{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.515025776436565,
  "eval_steps": 500,
  "global_step": 15000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 251.78126049041748,
      "epoch": 0.00033530323986755525,
      "grad_norm": 0.2821206415206205,
      "kl": 0.0,
      "learning_rate": 2.4691358024691357e-10,
      "loss": -0.0,
      "reward": 1.7428572177886963,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7517857365310192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2
    },
    {
      "completion_length": 251.26340675354004,
      "epoch": 0.0006706064797351105,
      "grad_norm": 0.23751592598417418,
      "kl": 2.596992999315262e-05,
      "learning_rate": 4.938271604938271e-10,
      "loss": 0.0,
      "reward": 1.7803571969270706,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4
    },
    {
      "completion_length": 255.4955472946167,
      "epoch": 0.0010059097196026656,
      "grad_norm": 0.34231130475220095,
      "kl": 0.00018531084060668945,
      "learning_rate": 7.407407407407407e-10,
      "loss": 0.0,
      "reward": 1.7053572311997414,
      "reward_std": 0.10354063473641872,
      "rewards/equation_reward_func": 0.7187500223517418,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6
    },
    {
      "completion_length": 242.8571538925171,
      "epoch": 0.001341212959470221,
      "grad_norm": 0.282011994670917,
      "kl": 0.0002923011779785156,
      "learning_rate": 9.876543209876543e-10,
      "loss": 0.0,
      "reward": 1.7816964909434319,
      "reward_std": 0.08649431029334664,
      "rewards/equation_reward_func": 0.793750025331974,
      "rewards/format_reward_func": 0.9879464358091354,
      "step": 8
    },
    {
      "completion_length": 246.70090103149414,
      "epoch": 0.0016765161993377762,
      "grad_norm": 0.4075528008052861,
      "kl": 0.0003095269203186035,
      "learning_rate": 1.2345679012345679e-09,
      "loss": 0.0,
      "reward": 1.7267858013510704,
      "reward_std": 0.11364216078072786,
      "rewards/equation_reward_func": 0.7401785962283611,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10
    },
    {
      "completion_length": 243.15179634094238,
      "epoch": 0.002011819439205331,
      "grad_norm": 0.2539315791340793,
      "kl": 0.00033289194107055664,
      "learning_rate": 1.4814814814814814e-09,
      "loss": 0.0,
      "reward": 1.775000050663948,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 12
    },
    {
      "completion_length": 246.633939743042,
      "epoch": 0.0023471226790728866,
      "grad_norm": 0.24597106986052078,
      "kl": 0.0003146529197692871,
      "learning_rate": 1.728395061728395e-09,
      "loss": 0.0,
      "reward": 1.687500074505806,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.691964328289032,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14
    },
    {
      "completion_length": 248.13840293884277,
      "epoch": 0.002682425918940442,
      "grad_norm": 0.2283280840772994,
      "kl": 0.0003129243850708008,
      "learning_rate": 1.9753086419753086e-09,
      "loss": 0.0,
      "reward": 1.7732143476605415,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7776786126196384,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 16
    },
    {
      "completion_length": 243.49108219146729,
      "epoch": 0.003017729158807997,
      "grad_norm": 0.25527874503701053,
      "kl": 0.0002974867820739746,
      "learning_rate": 2.222222222222222e-09,
      "loss": 0.0,
      "reward": 1.7375000715255737,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7419643178582191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 18
    },
    {
      "completion_length": 257.25000858306885,
      "epoch": 0.0033530323986755524,
      "grad_norm": 0.3755577292202996,
      "kl": 0.0003231167793273926,
      "learning_rate": 2.4691358024691357e-09,
      "loss": 0.0,
      "reward": 1.751785784959793,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500275671482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 20
    },
    {
      "completion_length": 248.70090293884277,
      "epoch": 0.0036883356385431073,
      "grad_norm": 0.25981511886182895,
      "kl": 0.00031244754791259766,
      "learning_rate": 2.7160493827160493e-09,
      "loss": 0.0,
      "reward": 1.7410715147852898,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7544643171131611,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 22
    },
    {
      "completion_length": 254.64733219146729,
      "epoch": 0.004023638878410662,
      "grad_norm": 0.15723937143886035,
      "kl": 0.0003066062927246094,
      "learning_rate": 2.962962962962963e-09,
      "loss": 0.0,
      "reward": 1.769642911851406,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7741071823984385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 24
    },
    {
      "completion_length": 240.6919765472412,
      "epoch": 0.004358942118278218,
      "grad_norm": 0.45759362392893915,
      "kl": 0.000310361385345459,
      "learning_rate": 3.209876543209876e-09,
      "loss": 0.0,
      "reward": 1.7950893640518188,
      "reward_std": 0.0776554741896689,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 26
    },
    {
      "completion_length": 250.5446548461914,
      "epoch": 0.004694245358145773,
      "grad_norm": 0.15906385599381753,
      "kl": 0.0002815127372741699,
      "learning_rate": 3.45679012345679e-09,
      "loss": 0.0,
      "reward": 1.7589286491274834,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928805589676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 28
    },
    {
      "completion_length": 245.8571548461914,
      "epoch": 0.005029548598013328,
      "grad_norm": 0.20213909633387014,
      "kl": 0.0003229975700378418,
      "learning_rate": 3.7037037037037036e-09,
      "loss": 0.0,
      "reward": 1.7946429252624512,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7991071566939354,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 30
    },
    {
      "completion_length": 239.7812614440918,
      "epoch": 0.005364851837880884,
      "grad_norm": 0.2975602670727807,
      "kl": 0.0003528594970703125,
      "learning_rate": 3.950617283950617e-09,
      "loss": 0.0,
      "reward": 1.7732143327593803,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7776786014437675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 32
    },
    {
      "completion_length": 244.4196548461914,
      "epoch": 0.005700155077748439,
      "grad_norm": 0.19631508669566625,
      "kl": 0.00030624866485595703,
      "learning_rate": 4.197530864197531e-09,
      "loss": 0.0,
      "reward": 1.7767857685685158,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7812500335276127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 34
    },
    {
      "completion_length": 248.04018878936768,
      "epoch": 0.006035458317615994,
      "grad_norm": 0.1693595352246246,
      "kl": 0.0003243684768676758,
      "learning_rate": 4.444444444444444e-09,
      "loss": 0.0,
      "reward": 1.6482143849134445,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.6616071630269289,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 36
    },
    {
      "completion_length": 246.6294765472412,
      "epoch": 0.006370761557483549,
      "grad_norm": 0.2152008240652874,
      "kl": 0.0003387331962585449,
      "learning_rate": 4.6913580246913574e-09,
      "loss": 0.0,
      "reward": 1.7517857775092125,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.765178594738245,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 38
    },
    {
      "completion_length": 242.79465579986572,
      "epoch": 0.006706064797351105,
      "grad_norm": 0.2081818106050967,
      "kl": 0.00032889842987060547,
      "learning_rate": 4.938271604938271e-09,
      "loss": 0.0,
      "reward": 1.7482143640518188,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7526786122471094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 40
    },
    {
      "completion_length": 247.04911708831787,
      "epoch": 0.00704136803721866,
      "grad_norm": 0.197765113235337,
      "kl": 0.00030040740966796875,
      "learning_rate": 5.1851851851851846e-09,
      "loss": 0.0,
      "reward": 1.8107143640518188,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.810714315623045,
      "rewards/format_reward_func": 1.0,
      "step": 42
    },
    {
      "completion_length": 233.49108219146729,
      "epoch": 0.007376671277086215,
      "grad_norm": 0.22273231251306044,
      "kl": 0.00031495094299316406,
      "learning_rate": 5.4320987654320985e-09,
      "loss": 0.0,
      "reward": 1.8017857447266579,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.8062500357627869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 44
    },
    {
      "completion_length": 237.5714406967163,
      "epoch": 0.00771197451695377,
      "grad_norm": 0.18181933051003635,
      "kl": 0.00033354759216308594,
      "learning_rate": 5.679012345679012e-09,
      "loss": 0.0,
      "reward": 1.7785715013742447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 46
    },
    {
      "completion_length": 246.5937614440918,
      "epoch": 0.008047277756821325,
      "grad_norm": 0.1606775878103379,
      "kl": 0.00032961368560791016,
      "learning_rate": 5.925925925925926e-09,
      "loss": 0.0,
      "reward": 1.7714286372065544,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286018162966,
      "rewards/format_reward_func": 1.0,
      "step": 48
    },
    {
      "completion_length": 239.64733219146729,
      "epoch": 0.008382580996688881,
      "grad_norm": 0.3020505688952993,
      "kl": 0.00031131505966186523,
      "learning_rate": 6.172839506172839e-09,
      "loss": 0.0,
      "reward": 1.7571429312229156,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 50
    },
    {
      "completion_length": 241.3169755935669,
      "epoch": 0.008717884236556436,
      "grad_norm": 0.28625665696322333,
      "kl": 0.0003256797790527344,
      "learning_rate": 6.419753086419752e-09,
      "loss": 0.0,
      "reward": 1.7821429148316383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 52
    },
    {
      "completion_length": 248.96429443359375,
      "epoch": 0.009053187476423991,
      "grad_norm": 0.26129573486345886,
      "kl": 0.0003027915954589844,
      "learning_rate": 6.666666666666667e-09,
      "loss": 0.0,
      "reward": 1.7571429088711739,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.766071455553174,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 54
    },
    {
      "completion_length": 250.23215293884277,
      "epoch": 0.009388490716291546,
      "grad_norm": 0.27548165311054307,
      "kl": 0.00035876035690307617,
      "learning_rate": 6.91358024691358e-09,
      "loss": 0.0,
      "reward": 1.7053572162985802,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7098214589059353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 56
    },
    {
      "completion_length": 244.34376049041748,
      "epoch": 0.009723793956159101,
      "grad_norm": 0.22151518083461771,
      "kl": 0.00031697750091552734,
      "learning_rate": 7.160493827160494e-09,
      "loss": 0.0,
      "reward": 1.8053571954369545,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8098214492201805,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 58
    },
    {
      "completion_length": 241.06697273254395,
      "epoch": 0.010059097196026656,
      "grad_norm": 0.22549107156528939,
      "kl": 0.000319063663482666,
      "learning_rate": 7.407407407407407e-09,
      "loss": 0.0,
      "reward": 1.7375000640749931,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643178582191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 60
    },
    {
      "completion_length": 243.7544755935669,
      "epoch": 0.010394400435894211,
      "grad_norm": 0.17997447995320268,
      "kl": 0.00031387805938720703,
      "learning_rate": 7.654320987654321e-09,
      "loss": 0.0,
      "reward": 1.741071492433548,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.74553575925529,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 62
    },
    {
      "completion_length": 239.93750953674316,
      "epoch": 0.010729703675761768,
      "grad_norm": 0.29247137737836243,
      "kl": 0.00034987926483154297,
      "learning_rate": 7.901234567901234e-09,
      "loss": 0.0,
      "reward": 1.7285714894533157,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7375000305473804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 64
    },
    {
      "completion_length": 246.06697273254395,
      "epoch": 0.011065006915629323,
      "grad_norm": 0.2353839170492452,
      "kl": 0.00032019615173339844,
      "learning_rate": 8.148148148148147e-09,
      "loss": 0.0,
      "reward": 1.7714286297559738,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714286129921675,
      "rewards/format_reward_func": 1.0,
      "step": 66
    },
    {
      "completion_length": 239.946439743042,
      "epoch": 0.011400310155496878,
      "grad_norm": 0.23810286576070455,
      "kl": 0.0003294944763183594,
      "learning_rate": 8.395061728395062e-09,
      "loss": 0.0,
      "reward": 1.7625000774860382,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7669643200933933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 68
    },
    {
      "completion_length": 244.83929538726807,
      "epoch": 0.011735613395364433,
      "grad_norm": 0.4474750559486092,
      "kl": 0.0003364682197570801,
      "learning_rate": 8.641975308641974e-09,
      "loss": 0.0,
      "reward": 1.7714286148548126,
      "reward_std": 0.11111677531152964,
      "rewards/equation_reward_func": 0.7714286185801029,
      "rewards/format_reward_func": 1.0,
      "step": 70
    },
    {
      "completion_length": 242.31697463989258,
      "epoch": 0.012070916635231988,
      "grad_norm": 0.23940161865679357,
      "kl": 0.000335693359375,
      "learning_rate": 8.888888888888889e-09,
      "loss": 0.0,
      "reward": 1.6892857998609543,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6892857477068901,
      "rewards/format_reward_func": 1.0,
      "step": 72
    },
    {
      "completion_length": 250.2500123977661,
      "epoch": 0.012406219875099543,
      "grad_norm": 0.3547914135677079,
      "kl": 0.00031691789627075195,
      "learning_rate": 9.135802469135803e-09,
      "loss": 0.0,
      "reward": 1.748214341700077,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7526785861700773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 74
    },
    {
      "completion_length": 247.17858505249023,
      "epoch": 0.012741523114967098,
      "grad_norm": 0.37705598365442405,
      "kl": 0.0003069639205932617,
      "learning_rate": 9.382716049382715e-09,
      "loss": 0.0,
      "reward": 1.7142857983708382,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7142857499420643,
      "rewards/format_reward_func": 1.0,
      "step": 76
    },
    {
      "completion_length": 240.8080472946167,
      "epoch": 0.013076826354834654,
      "grad_norm": 0.2500330365327647,
      "kl": 0.00031298398971557617,
      "learning_rate": 9.62962962962963e-09,
      "loss": 0.0,
      "reward": 1.7339286282658577,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7383928969502449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 78
    },
    {
      "completion_length": 252.24554824829102,
      "epoch": 0.01341212959470221,
      "grad_norm": 0.24171287363674568,
      "kl": 0.000303804874420166,
      "learning_rate": 9.876543209876543e-09,
      "loss": 0.0,
      "reward": 1.7433036342263222,
      "reward_std": 0.044825518038123846,
      "rewards/equation_reward_func": 0.7580357380211353,
      "rewards/format_reward_func": 0.9852678664028645,
      "step": 80
    },
    {
      "completion_length": 247.12054824829102,
      "epoch": 0.013747432834569764,
      "grad_norm": 0.3022600089442828,
      "kl": 0.00033527612686157227,
      "learning_rate": 1.0123456790123458e-08,
      "loss": 0.0,
      "reward": 1.6678572446107864,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6678571682423353,
      "rewards/format_reward_func": 1.0,
      "step": 82
    },
    {
      "completion_length": 245.23662281036377,
      "epoch": 0.01408273607443732,
      "grad_norm": 0.27039366876832605,
      "kl": 0.00029647350311279297,
      "learning_rate": 1.0370370370370369e-08,
      "loss": 0.0,
      "reward": 1.80892863124609,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.813392885029316,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 84
    },
    {
      "completion_length": 250.4330472946167,
      "epoch": 0.014418039314304874,
      "grad_norm": 0.7187583509006711,
      "kl": 0.0003364086151123047,
      "learning_rate": 1.0617283950617284e-08,
      "loss": 0.0,
      "reward": 1.7321429327130318,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7321428917348385,
      "rewards/format_reward_func": 1.0,
      "step": 86
    },
    {
      "completion_length": 247.8169765472412,
      "epoch": 0.01475334255417243,
      "grad_norm": 0.23848320000262993,
      "kl": 0.0003071427345275879,
      "learning_rate": 1.0864197530864197e-08,
      "loss": 0.0,
      "reward": 1.751785770058632,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7562500238418579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 88
    },
    {
      "completion_length": 240.4776906967163,
      "epoch": 0.015088645794039984,
      "grad_norm": 0.27460619568878064,
      "kl": 0.0003440380096435547,
      "learning_rate": 1.111111111111111e-08,
      "loss": 0.0,
      "reward": 1.798214353621006,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.8026785980910063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 90
    },
    {
      "completion_length": 251.821439743042,
      "epoch": 0.01542394903390754,
      "grad_norm": 0.29052291960498944,
      "kl": 0.00032770633697509766,
      "learning_rate": 1.1358024691358023e-08,
      "loss": 0.0,
      "reward": 1.7571429386734962,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7660714574158192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 92
    },
    {
      "completion_length": 245.17411613464355,
      "epoch": 0.015759252273775094,
      "grad_norm": 0.1360821787656637,
      "kl": 0.00029844045639038086,
      "learning_rate": 1.1604938271604938e-08,
      "loss": 0.0,
      "reward": 1.7446429282426834,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7491071783006191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 94
    },
    {
      "completion_length": 256.1517963409424,
      "epoch": 0.01609455551364265,
      "grad_norm": 0.27993944292911604,
      "kl": 0.0003134012222290039,
      "learning_rate": 1.1851851851851851e-08,
      "loss": 0.0,
      "reward": 1.7660714983940125,
      "reward_std": 0.09848987311124802,
      "rewards/equation_reward_func": 0.7794643267989159,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 96
    },
    {
      "completion_length": 242.4553689956665,
      "epoch": 0.016429858753510204,
      "grad_norm": 0.2106037899562362,
      "kl": 0.00028455257415771484,
      "learning_rate": 1.2098765432098765e-08,
      "loss": 0.0,
      "reward": 1.760714367032051,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 98
    },
    {
      "completion_length": 239.50447368621826,
      "epoch": 0.016765161993377763,
      "grad_norm": 0.3103782359638075,
      "kl": 0.0003159642219543457,
      "learning_rate": 1.2345679012345678e-08,
      "loss": 0.0,
      "reward": 1.764285795390606,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857693135738,
      "rewards/format_reward_func": 1.0,
      "step": 100
    },
    {
      "completion_length": 254.9285831451416,
      "epoch": 0.017100465233245318,
      "grad_norm": 0.17884850735394575,
      "kl": 0.00032651424407958984,
      "learning_rate": 1.2592592592592592e-08,
      "loss": 0.0,
      "reward": 1.8000000640749931,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000417232513,
      "rewards/format_reward_func": 1.0,
      "step": 102
    },
    {
      "completion_length": 248.58483219146729,
      "epoch": 0.017435768473112873,
      "grad_norm": 0.29105440062209287,
      "kl": 0.00033777952194213867,
      "learning_rate": 1.2839506172839504e-08,
      "loss": 0.0,
      "reward": 1.7629464864730835,
      "reward_std": 0.03219861118122935,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 104
    },
    {
      "completion_length": 251.37500953674316,
      "epoch": 0.017771071712980428,
      "grad_norm": 0.24668718286887112,
      "kl": 0.0003491640090942383,
      "learning_rate": 1.3086419753086419e-08,
      "loss": 0.0,
      "reward": 1.7071429342031479,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7071428932249546,
      "rewards/format_reward_func": 1.0,
      "step": 106
    },
    {
      "completion_length": 241.62501049041748,
      "epoch": 0.018106374952847983,
      "grad_norm": 0.2734650656617372,
      "kl": 0.0003235936164855957,
      "learning_rate": 1.3333333333333334e-08,
      "loss": 0.0,
      "reward": 1.7803572043776512,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 108
    },
    {
      "completion_length": 252.74108219146729,
      "epoch": 0.018441678192715538,
      "grad_norm": 0.2705035956067135,
      "kl": 0.00032901763916015625,
      "learning_rate": 1.3580246913580247e-08,
      "loss": 0.0,
      "reward": 1.7375000789761543,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.741964315995574,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 110
    },
    {
      "completion_length": 240.09822463989258,
      "epoch": 0.018776981432583092,
      "grad_norm": 0.14575037715148337,
      "kl": 0.0003132820129394531,
      "learning_rate": 1.382716049382716e-08,
      "loss": 0.0,
      "reward": 1.8000000640749931,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000342726707,
      "rewards/format_reward_func": 1.0,
      "step": 112
    },
    {
      "completion_length": 251.90626049041748,
      "epoch": 0.019112284672450647,
      "grad_norm": 0.28566608159303575,
      "kl": 0.0003253817558288574,
      "learning_rate": 1.4074074074074073e-08,
      "loss": 0.0,
      "reward": 1.7392857745289803,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.7482143174856901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 114
    },
    {
      "completion_length": 245.4419755935669,
      "epoch": 0.019447587912318202,
      "grad_norm": 0.22007001347735153,
      "kl": 0.0003275871276855469,
      "learning_rate": 1.4320987654320988e-08,
      "loss": 0.0,
      "reward": 1.7660714983940125,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7705357447266579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 116
    },
    {
      "completion_length": 246.47769165039062,
      "epoch": 0.019782891152185757,
      "grad_norm": 0.24532766247165583,
      "kl": 0.00031566619873046875,
      "learning_rate": 1.4567901234567901e-08,
      "loss": 0.0,
      "reward": 1.7267857939004898,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7312500327825546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 118
    },
    {
      "completion_length": 246.93304634094238,
      "epoch": 0.020118194392053312,
      "grad_norm": 0.21397852222502559,
      "kl": 0.00031769275665283203,
      "learning_rate": 1.4814814814814814e-08,
      "loss": 0.0,
      "reward": 1.7232143506407738,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7276785969734192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 120
    },
    {
      "completion_length": 245.8035831451416,
      "epoch": 0.020453497631920867,
      "grad_norm": 0.44949071446508276,
      "kl": 0.0003509521484375,
      "learning_rate": 1.5061728395061727e-08,
      "loss": 0.0,
      "reward": 1.7589286342263222,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928768336773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 122
    },
    {
      "completion_length": 250.5848331451416,
      "epoch": 0.020788800871788422,
      "grad_norm": 0.00010308908116451039,
      "kl": 0.00032585859298706055,
      "learning_rate": 1.5308641975308642e-08,
      "loss": 0.0,
      "reward": 1.732142947614193,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7321428954601288,
      "rewards/format_reward_func": 1.0,
      "step": 124
    },
    {
      "completion_length": 242.9062623977661,
      "epoch": 0.021124104111655977,
      "grad_norm": 0.2326498304971465,
      "kl": 0.0003529787063598633,
      "learning_rate": 1.5555555555555554e-08,
      "loss": 0.0,
      "reward": 1.7750000655651093,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.775000037625432,
      "rewards/format_reward_func": 1.0,
      "step": 126
    },
    {
      "completion_length": 248.71876049041748,
      "epoch": 0.021459407351523536,
      "grad_norm": 0.2724319389491217,
      "kl": 0.0003319978713989258,
      "learning_rate": 1.580246913580247e-08,
      "loss": 0.0,
      "reward": 1.767857201397419,
      "reward_std": 0.08586296439170837,
      "rewards/equation_reward_func": 0.7767857573926449,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 128
    },
    {
      "completion_length": 244.5982265472412,
      "epoch": 0.02179471059139109,
      "grad_norm": 0.28980555293971616,
      "kl": 0.00033468008041381836,
      "learning_rate": 1.6049382716049383e-08,
      "loss": 0.0,
      "reward": 1.796428620815277,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 130
    },
    {
      "completion_length": 246.5491189956665,
      "epoch": 0.022130013831258646,
      "grad_norm": 0.22034484681712568,
      "kl": 0.00032633543014526367,
      "learning_rate": 1.6296296296296295e-08,
      "loss": 0.0,
      "reward": 1.7267857939004898,
      "reward_std": 0.0732360603287816,
      "rewards/equation_reward_func": 0.7401786036789417,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 132
    },
    {
      "completion_length": 251.6919765472412,
      "epoch": 0.0224653170711262,
      "grad_norm": 0.21857858965530919,
      "kl": 0.00031572580337524414,
      "learning_rate": 1.654320987654321e-08,
      "loss": 0.0,
      "reward": 1.7660714909434319,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7705357261002064,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 134
    },
    {
      "completion_length": 244.83483505249023,
      "epoch": 0.022800620310993756,
      "grad_norm": 0.2697978233876309,
      "kl": 0.00033473968505859375,
      "learning_rate": 1.6790123456790124e-08,
      "loss": 0.0,
      "reward": 1.7500000447034836,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 136
    },
    {
      "completion_length": 256.58483505249023,
      "epoch": 0.02313592355086131,
      "grad_norm": 0.300841254760372,
      "kl": 0.0002970099449157715,
      "learning_rate": 1.7037037037037036e-08,
      "loss": 0.0,
      "reward": 1.7285715192556381,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7285714671015739,
      "rewards/format_reward_func": 1.0,
      "step": 138
    },
    {
      "completion_length": 246.47768783569336,
      "epoch": 0.023471226790728866,
      "grad_norm": 0.2083502710840338,
      "kl": 0.0003154277801513672,
      "learning_rate": 1.7283950617283947e-08,
      "loss": 0.0,
      "reward": 1.733928643167019,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.7383928969502449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 140
    },
    {
      "completion_length": 253.37054538726807,
      "epoch": 0.02380653003059642,
      "grad_norm": 0.19329695372740538,
      "kl": 0.00031769275665283203,
      "learning_rate": 1.7530864197530862e-08,
      "loss": 0.0,
      "reward": 1.7571429386734962,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 142
    },
    {
      "completion_length": 244.33483123779297,
      "epoch": 0.024141833270463976,
      "grad_norm": 0.2657022566055933,
      "kl": 0.00033587217330932617,
      "learning_rate": 1.7777777777777777e-08,
      "loss": 0.0,
      "reward": 1.7642857655882835,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.773214302957058,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 144
    },
    {
      "completion_length": 241.88840293884277,
      "epoch": 0.02447713651033153,
      "grad_norm": 0.24003225151699342,
      "kl": 0.00033402442932128906,
      "learning_rate": 1.8024691358024692e-08,
      "loss": 0.0,
      "reward": 1.7392857894301414,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7482143193483353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 146
    },
    {
      "completion_length": 254.4598331451416,
      "epoch": 0.024812439750199086,
      "grad_norm": 0.2815512180517347,
      "kl": 0.0003273487091064453,
      "learning_rate": 1.8271604938271607e-08,
      "loss": 0.0,
      "reward": 1.7017857804894447,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7062500528991222,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 148
    },
    {
      "completion_length": 235.10268878936768,
      "epoch": 0.02514774299006664,
      "grad_norm": 0.32352647991566374,
      "kl": 0.00031512975692749023,
      "learning_rate": 1.8518518518518518e-08,
      "loss": 0.0,
      "reward": 1.7821429073810577,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428962051868,
      "rewards/format_reward_func": 1.0,
      "step": 150
    },
    {
      "completion_length": 243.23215198516846,
      "epoch": 0.025483046229934195,
      "grad_norm": 0.1719160101460022,
      "kl": 0.0003261566162109375,
      "learning_rate": 1.876543209876543e-08,
      "loss": 0.0,
      "reward": 1.7571429312229156,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.757142897695303,
      "rewards/format_reward_func": 1.0,
      "step": 152
    },
    {
      "completion_length": 247.1071548461914,
      "epoch": 0.02581834946980175,
      "grad_norm": 0.2568745935002292,
      "kl": 0.00030517578125,
      "learning_rate": 1.9012345679012344e-08,
      "loss": 0.0,
      "reward": 1.7089286521077156,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7133929040282965,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 154
    },
    {
      "completion_length": 242.92411994934082,
      "epoch": 0.02615365270966931,
      "grad_norm": 0.32403116735365,
      "kl": 0.00030052661895751953,
      "learning_rate": 1.925925925925926e-08,
      "loss": 0.0,
      "reward": 1.783928632736206,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 156
    },
    {
      "completion_length": 245.2053680419922,
      "epoch": 0.026488955949536864,
      "grad_norm": 0.2945305207519835,
      "kl": 0.00034099817276000977,
      "learning_rate": 1.950617283950617e-08,
      "loss": 0.0,
      "reward": 1.760714367032051,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 158
    },
    {
      "completion_length": 250.0312623977661,
      "epoch": 0.02682425918940442,
      "grad_norm": 0.2981654874681236,
      "kl": 0.00033992528915405273,
      "learning_rate": 1.9753086419753086e-08,
      "loss": 0.0,
      "reward": 1.7785714864730835,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7875000201165676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 160
    },
    {
      "completion_length": 252.8616180419922,
      "epoch": 0.027159562429271974,
      "grad_norm": 0.20431294977682624,
      "kl": 0.0002976059913635254,
      "learning_rate": 2e-08,
      "loss": 0.0,
      "reward": 1.7964286357164383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 162
    },
    {
      "completion_length": 246.81697368621826,
      "epoch": 0.02749486566913953,
      "grad_norm": 0.1897492145271756,
      "kl": 0.00032776594161987305,
      "learning_rate": 2.0246913580246915e-08,
      "loss": 0.0,
      "reward": 1.7375000715255737,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7419643253087997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 164
    },
    {
      "completion_length": 242.5178689956665,
      "epoch": 0.027830168909007084,
      "grad_norm": 0.1981890035443643,
      "kl": 0.00030219554901123047,
      "learning_rate": 2.0493827160493823e-08,
      "loss": 0.0,
      "reward": 1.7089286670088768,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.713392898440361,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 166
    },
    {
      "completion_length": 242.95536994934082,
      "epoch": 0.02816547214887464,
      "grad_norm": 0.27498948462332695,
      "kl": 0.00031065940856933594,
      "learning_rate": 2.0740740740740738e-08,
      "loss": 0.0,
      "reward": 1.7267858237028122,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.731250025331974,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 168
    },
    {
      "completion_length": 244.3169765472412,
      "epoch": 0.028500775388742194,
      "grad_norm": 0.18119311717593303,
      "kl": 0.0003190040588378906,
      "learning_rate": 2.0987654320987653e-08,
      "loss": 0.0,
      "reward": 1.7535715103149414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 170
    },
    {
      "completion_length": 250.15179824829102,
      "epoch": 0.02883607862860975,
      "grad_norm": 0.1600883159721621,
      "kl": 0.0003666877746582031,
      "learning_rate": 2.1234567901234568e-08,
      "loss": 0.0,
      "reward": 1.775000087916851,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 172
    },
    {
      "completion_length": 252.1607255935669,
      "epoch": 0.029171381868477304,
      "grad_norm": 0.20143375779388584,
      "kl": 0.00031435489654541016,
      "learning_rate": 2.148148148148148e-08,
      "loss": 0.0,
      "reward": 1.7308036610484123,
      "reward_std": 0.0473508988507092,
      "rewards/equation_reward_func": 0.7321428842842579,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 174
    },
    {
      "completion_length": 245.46429538726807,
      "epoch": 0.02950668510834486,
      "grad_norm": 0.38570735794849914,
      "kl": 0.0003286004066467285,
      "learning_rate": 2.1728395061728394e-08,
      "loss": 0.0,
      "reward": 1.7125001028180122,
      "reward_std": 0.10354063380509615,
      "rewards/equation_reward_func": 0.7169643118977547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 176
    },
    {
      "completion_length": 247.22769165039062,
      "epoch": 0.029841988348212414,
      "grad_norm": 0.3972032973187878,
      "kl": 0.0003095269203186035,
      "learning_rate": 2.197530864197531e-08,
      "loss": 0.0,
      "reward": 1.748214341700077,
      "reward_std": 0.10354063287377357,
      "rewards/equation_reward_func": 0.7526786141097546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 178
    },
    {
      "completion_length": 243.81697463989258,
      "epoch": 0.03017729158807997,
      "grad_norm": 0.3034668175618403,
      "kl": 0.00033104419708251953,
      "learning_rate": 2.222222222222222e-08,
      "loss": 0.0,
      "reward": 1.7535714954137802,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 180
    },
    {
      "completion_length": 244.0089406967163,
      "epoch": 0.030512594827947524,
      "grad_norm": 0.20606761170561028,
      "kl": 0.00030857324600219727,
      "learning_rate": 2.2469135802469135e-08,
      "loss": 0.0,
      "reward": 1.7875000536441803,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7919643055647612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 182
    },
    {
      "completion_length": 250.7098331451416,
      "epoch": 0.03084789806781508,
      "grad_norm": 0.20840818777387127,
      "kl": 0.00032150745391845703,
      "learning_rate": 2.2716049382716047e-08,
      "loss": 0.0,
      "reward": 1.7678572088479996,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571604192257,
      "rewards/format_reward_func": 1.0,
      "step": 184
    },
    {
      "completion_length": 238.39733219146729,
      "epoch": 0.031183201307682637,
      "grad_norm": 0.30613951822839514,
      "kl": 0.00033026933670043945,
      "learning_rate": 2.296296296296296e-08,
      "loss": 0.0,
      "reward": 1.8000000566244125,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8000000156462193,
      "rewards/format_reward_func": 1.0,
      "step": 186
    },
    {
      "completion_length": 242.68751335144043,
      "epoch": 0.03151850454755019,
      "grad_norm": 0.24889609063442605,
      "kl": 0.0003234744071960449,
      "learning_rate": 2.3209876543209876e-08,
      "loss": 0.0,
      "reward": 1.7571429163217545,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428958326578,
      "rewards/format_reward_func": 1.0,
      "step": 188
    },
    {
      "completion_length": 236.5714406967163,
      "epoch": 0.03185380778741775,
      "grad_norm": 0.25782909812310634,
      "kl": 0.0003020763397216797,
      "learning_rate": 2.345679012345679e-08,
      "loss": 0.0,
      "reward": 1.7535714954137802,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714693367481,
      "rewards/format_reward_func": 1.0,
      "step": 190
    },
    {
      "completion_length": 242.67412090301514,
      "epoch": 0.0321891110272853,
      "grad_norm": 0.16179825374337622,
      "kl": 0.0003166794776916504,
      "learning_rate": 2.3703703703703703e-08,
      "loss": 0.0,
      "reward": 1.8178571984171867,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.8267857432365417,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 192
    },
    {
      "completion_length": 246.08929634094238,
      "epoch": 0.03252441426715286,
      "grad_norm": 0.20594663668147256,
      "kl": 0.00032722949981689453,
      "learning_rate": 2.3950617283950614e-08,
      "loss": 0.0,
      "reward": 1.8089286237955093,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8133928775787354,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 194
    },
    {
      "completion_length": 249.73215770721436,
      "epoch": 0.03285971750702041,
      "grad_norm": 0.2088766158551277,
      "kl": 0.00032842159271240234,
      "learning_rate": 2.419753086419753e-08,
      "loss": 0.0,
      "reward": 1.7089286595582962,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7133928947150707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 196
    },
    {
      "completion_length": 246.8571548461914,
      "epoch": 0.03319502074688797,
      "grad_norm": 0.19387291215229924,
      "kl": 0.00029861927032470703,
      "learning_rate": 2.4444444444444444e-08,
      "loss": 0.0,
      "reward": 1.7946429252624512,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7991071604192257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 198
    },
    {
      "completion_length": 246.4910831451416,
      "epoch": 0.033530323986755525,
      "grad_norm": 0.14395851317716862,
      "kl": 0.00030988454818725586,
      "learning_rate": 2.4691358024691355e-08,
      "loss": 0.0,
      "reward": 1.7553572207689285,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214540630579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 200
    },
    {
      "completion_length": 234.1294765472412,
      "epoch": 0.03386562722662308,
      "grad_norm": 0.24436453124931118,
      "kl": 0.00032639503479003906,
      "learning_rate": 2.493827160493827e-08,
      "loss": 0.0,
      "reward": 1.7517857775092125,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.75625004991889,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 202
    },
    {
      "completion_length": 248.93304634094238,
      "epoch": 0.034200930466490635,
      "grad_norm": 0.2248675849511274,
      "kl": 0.0003234744071960449,
      "learning_rate": 2.5185185185185185e-08,
      "loss": 0.0,
      "reward": 1.7285715267062187,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7285714596509933,
      "rewards/format_reward_func": 1.0,
      "step": 204
    },
    {
      "completion_length": 246.19197368621826,
      "epoch": 0.03453623370635819,
      "grad_norm": 0.22934369353586878,
      "kl": 0.0003287792205810547,
      "learning_rate": 2.54320987654321e-08,
      "loss": 0.0,
      "reward": 1.7500000894069672,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 206
    },
    {
      "completion_length": 245.40179634094238,
      "epoch": 0.034871536946225745,
      "grad_norm": 0.29961009398828276,
      "kl": 0.00034368038177490234,
      "learning_rate": 2.5679012345679008e-08,
      "loss": 0.0,
      "reward": 1.7616072222590446,
      "reward_std": 0.06944798538461328,
      "rewards/equation_reward_func": 0.7723214626312256,
      "rewards/format_reward_func": 0.9892857223749161,
      "step": 208
    },
    {
      "completion_length": 248.41518878936768,
      "epoch": 0.0352068401860933,
      "grad_norm": 0.15792865989980773,
      "kl": 0.00034034252166748047,
      "learning_rate": 2.5925925925925923e-08,
      "loss": 0.0,
      "reward": 1.769642896950245,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7741071861237288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 210
    },
    {
      "completion_length": 242.93304538726807,
      "epoch": 0.035542143425960855,
      "grad_norm": 0.3301838137697797,
      "kl": 0.0003197193145751953,
      "learning_rate": 2.6172839506172838e-08,
      "loss": 0.0,
      "reward": 1.750446505844593,
      "reward_std": 0.07007933082059026,
      "rewards/equation_reward_func": 0.7562500145286322,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 212
    },
    {
      "completion_length": 248.7276906967163,
      "epoch": 0.03587744666582841,
      "grad_norm": 0.35616311274054235,
      "kl": 0.0003361701965332031,
      "learning_rate": 2.6419753086419752e-08,
      "loss": 0.0,
      "reward": 1.7678572162985802,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 214
    },
    {
      "completion_length": 248.99108505249023,
      "epoch": 0.036212749905695965,
      "grad_norm": 0.23756977130728407,
      "kl": 0.00034689903259277344,
      "learning_rate": 2.6666666666666667e-08,
      "loss": 0.0,
      "reward": 1.7017857804894447,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7062500342726707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 216
    },
    {
      "completion_length": 248.22768688201904,
      "epoch": 0.03654805314556352,
      "grad_norm": 0.2756674602915548,
      "kl": 0.00029724836349487305,
      "learning_rate": 2.691358024691358e-08,
      "loss": 0.0,
      "reward": 1.705357238650322,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7098214626312256,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 218
    },
    {
      "completion_length": 233.95983219146729,
      "epoch": 0.036883356385431075,
      "grad_norm": 0.2544126349698565,
      "kl": 0.0003007054328918457,
      "learning_rate": 2.7160493827160494e-08,
      "loss": 0.0,
      "reward": 1.7625000700354576,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7669643107801676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 220
    },
    {
      "completion_length": 250.9553680419922,
      "epoch": 0.03721865962529863,
      "grad_norm": 0.213788279693055,
      "kl": 0.0003082156181335449,
      "learning_rate": 2.740740740740741e-08,
      "loss": 0.0,
      "reward": 1.7589286267757416,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7633928842842579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 222
    },
    {
      "completion_length": 249.2500123977661,
      "epoch": 0.037553962865166185,
      "grad_norm": 0.2610069019462948,
      "kl": 0.0003325939178466797,
      "learning_rate": 2.765432098765432e-08,
      "loss": 0.0,
      "reward": 1.72857154160738,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.728571455925703,
      "rewards/format_reward_func": 1.0,
      "step": 224
    },
    {
      "completion_length": 247.03572368621826,
      "epoch": 0.037889266105033736,
      "grad_norm": 0.2206000139730647,
      "kl": 0.0003177523612976074,
      "learning_rate": 2.790123456790123e-08,
      "loss": 0.0,
      "reward": 1.773214340209961,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7776786051690578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 226
    },
    {
      "completion_length": 243.68304634094238,
      "epoch": 0.038224569344901295,
      "grad_norm": 0.34760430842090373,
      "kl": 0.0003180503845214844,
      "learning_rate": 2.8148148148148146e-08,
      "loss": 0.0,
      "reward": 1.7267857789993286,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.731250025331974,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 228
    },
    {
      "completion_length": 259.589298248291,
      "epoch": 0.03855987258476885,
      "grad_norm": 0.20944117193770692,
      "kl": 0.000335693359375,
      "learning_rate": 2.839506172839506e-08,
      "loss": 0.0,
      "reward": 1.6821429431438446,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.6910714581608772,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 230
    },
    {
      "completion_length": 241.665189743042,
      "epoch": 0.038895175824636405,
      "grad_norm": 0.2512126660543398,
      "kl": 0.00031304359436035156,
      "learning_rate": 2.8641975308641976e-08,
      "loss": 0.0,
      "reward": 1.7285714894533157,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7285714671015739,
      "rewards/format_reward_func": 1.0,
      "step": 232
    },
    {
      "completion_length": 234.51786708831787,
      "epoch": 0.03923047906450396,
      "grad_norm": 0.18261909607261712,
      "kl": 0.000345766544342041,
      "learning_rate": 2.8888888888888887e-08,
      "loss": 0.0,
      "reward": 1.73392865806818,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7375000305473804,
      "rewards/format_reward_func": 0.9964285790920258,
      "step": 234
    },
    {
      "completion_length": 252.02679634094238,
      "epoch": 0.039565782304371515,
      "grad_norm": 0.20775712922379,
      "kl": 0.0003249645233154297,
      "learning_rate": 2.9135802469135802e-08,
      "loss": 0.0,
      "reward": 1.7589286416769028,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928880095482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 236
    },
    {
      "completion_length": 242.97768783569336,
      "epoch": 0.03990108554423907,
      "grad_norm": 0.1833960305143577,
      "kl": 0.00034111738204956055,
      "learning_rate": 2.9382716049382714e-08,
      "loss": 0.0,
      "reward": 1.7446429207921028,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071820259094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 238
    },
    {
      "completion_length": 244.35269165039062,
      "epoch": 0.040236388784106625,
      "grad_norm": 0.08606613377561309,
      "kl": 0.0003235340118408203,
      "learning_rate": 2.962962962962963e-08,
      "loss": 0.0,
      "reward": 1.750000074505806,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000186264515,
      "rewards/format_reward_func": 1.0,
      "step": 240
    },
    {
      "completion_length": 242.70090675354004,
      "epoch": 0.04057169202397418,
      "grad_norm": 0.22739026239901686,
      "kl": 0.00033462047576904297,
      "learning_rate": 2.987654320987654e-08,
      "loss": 0.0,
      "reward": 1.7982143387198448,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.802678607404232,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 242
    },
    {
      "completion_length": 240.8169755935669,
      "epoch": 0.040906995263841735,
      "grad_norm": 0.23189861598153808,
      "kl": 0.00033158063888549805,
      "learning_rate": 3.0123456790123455e-08,
      "loss": 0.0,
      "reward": 1.7357143685221672,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7446428909897804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 244
    },
    {
      "completion_length": 245.93305015563965,
      "epoch": 0.04124229850370929,
      "grad_norm": 0.3452260535215257,
      "kl": 0.0003554821014404297,
      "learning_rate": 3.037037037037037e-08,
      "loss": 0.0,
      "reward": 1.7750000730156898,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 246
    },
    {
      "completion_length": 254.03572750091553,
      "epoch": 0.041577601743576845,
      "grad_norm": 0.3311219707962642,
      "kl": 0.0003204345703125,
      "learning_rate": 3.0617283950617284e-08,
      "loss": 0.0,
      "reward": 1.6928572431206703,
      "reward_std": 0.10101525112986565,
      "rewards/equation_reward_func": 0.6928571779280901,
      "rewards/format_reward_func": 1.0,
      "step": 248
    },
    {
      "completion_length": 249.513409614563,
      "epoch": 0.0419129049834444,
      "grad_norm": 0.37071766869235806,
      "kl": 0.0003209114074707031,
      "learning_rate": 3.086419753086419e-08,
      "loss": 0.0,
      "reward": 1.7535714954137802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 250
    },
    {
      "completion_length": 244.571439743042,
      "epoch": 0.042248208223311955,
      "grad_norm": 0.2739936128591675,
      "kl": 0.0003472566604614258,
      "learning_rate": 3.111111111111111e-08,
      "loss": 0.0,
      "reward": 1.7607143446803093,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143130153418,
      "rewards/format_reward_func": 1.0,
      "step": 252
    },
    {
      "completion_length": 232.39733028411865,
      "epoch": 0.04258351146317951,
      "grad_norm": 0.17136269147118577,
      "kl": 0.00031685829162597656,
      "learning_rate": 3.135802469135802e-08,
      "loss": 0.0,
      "reward": 1.7750000655651093,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 254
    },
    {
      "completion_length": 242.82144165039062,
      "epoch": 0.04291881470304707,
      "grad_norm": 0.24005201682447744,
      "kl": 0.0003197789192199707,
      "learning_rate": 3.160493827160494e-08,
      "loss": 0.0,
      "reward": 1.735714353621006,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143275439739,
      "rewards/format_reward_func": 1.0,
      "step": 256
    },
    {
      "completion_length": 247.8571538925171,
      "epoch": 0.04325411794291462,
      "grad_norm": 0.25000441328741113,
      "kl": 0.0003287792205810547,
      "learning_rate": 3.185185185185185e-08,
      "loss": 0.0,
      "reward": 1.7410714998841286,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7455357387661934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 258
    },
    {
      "completion_length": 246.6428680419922,
      "epoch": 0.04358942118278218,
      "grad_norm": 0.18778229429765905,
      "kl": 0.0002881288528442383,
      "learning_rate": 3.2098765432098767e-08,
      "loss": 0.0,
      "reward": 1.7500000894069672,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7589285969734192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 260
    },
    {
      "completion_length": 252.77679634094238,
      "epoch": 0.04392472442264973,
      "grad_norm": 0.2507058768188766,
      "kl": 0.0003418922424316406,
      "learning_rate": 3.234567901234568e-08,
      "loss": 0.0,
      "reward": 1.6964286491274834,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6964286118745804,
      "rewards/format_reward_func": 1.0,
      "step": 262
    },
    {
      "completion_length": 250.5223331451416,
      "epoch": 0.04426002766251729,
      "grad_norm": 0.2145908398443362,
      "kl": 0.00032597780227661133,
      "learning_rate": 3.259259259259259e-08,
      "loss": 0.0,
      "reward": 1.7250000685453415,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7339286059141159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 264
    },
    {
      "completion_length": 244.2946538925171,
      "epoch": 0.04459533090238484,
      "grad_norm": 0.1814048605402375,
      "kl": 0.0003126859664916992,
      "learning_rate": 3.2839506172839504e-08,
      "loss": 0.0,
      "reward": 1.7357143461704254,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7446428835391998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 266
    },
    {
      "completion_length": 240.95090579986572,
      "epoch": 0.0449306341422524,
      "grad_norm": 0.573185463890021,
      "kl": 0.00038886070251464844,
      "learning_rate": 3.308641975308642e-08,
      "loss": 0.0,
      "reward": 1.7875000461935997,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7919642999768257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 268
    },
    {
      "completion_length": 251.35269165039062,
      "epoch": 0.04526593738211995,
      "grad_norm": 0.345458955989029,
      "kl": 0.0003148317337036133,
      "learning_rate": 3.3333333333333334e-08,
      "loss": 0.0,
      "reward": 1.7607143595814705,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 270
    },
    {
      "completion_length": 240.8303680419922,
      "epoch": 0.04560124062198751,
      "grad_norm": 0.21387981859355704,
      "kl": 0.00033462047576904297,
      "learning_rate": 3.358024691358025e-08,
      "loss": 0.0,
      "reward": 1.7678572237491608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 272
    },
    {
      "completion_length": 243.8482265472412,
      "epoch": 0.04593654386185506,
      "grad_norm": 0.1372445448617604,
      "kl": 0.00030040740966796875,
      "learning_rate": 3.382716049382716e-08,
      "loss": 0.0,
      "reward": 1.8017857745289803,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.806250024586916,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 274
    },
    {
      "completion_length": 242.38393878936768,
      "epoch": 0.04627184710172262,
      "grad_norm": 0.2538804221997863,
      "kl": 0.0003288388252258301,
      "learning_rate": 3.407407407407407e-08,
      "loss": 0.0,
      "reward": 1.7482143491506577,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7526785973459482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 276
    },
    {
      "completion_length": 243.4553680419922,
      "epoch": 0.04660715034159017,
      "grad_norm": 0.17161419413289314,
      "kl": 0.0003286600112915039,
      "learning_rate": 3.4320987654320987e-08,
      "loss": 0.0,
      "reward": 1.739285796880722,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857521772385,
      "rewards/format_reward_func": 1.0,
      "step": 278
    },
    {
      "completion_length": 253.1071548461914,
      "epoch": 0.04694245358145773,
      "grad_norm": 0.1583888212171527,
      "kl": 0.0003293752670288086,
      "learning_rate": 3.4567901234567895e-08,
      "loss": 0.0,
      "reward": 1.7857143580913544,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143152505159,
      "rewards/format_reward_func": 1.0,
      "step": 280
    },
    {
      "completion_length": 247.33483600616455,
      "epoch": 0.04727775682132528,
      "grad_norm": 0.2364266681687748,
      "kl": 0.0003266334533691406,
      "learning_rate": 3.481481481481481e-08,
      "loss": 0.0,
      "reward": 1.725000075995922,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7250000461935997,
      "rewards/format_reward_func": 1.0,
      "step": 282
    },
    {
      "completion_length": 247.9196548461914,
      "epoch": 0.04761306006119284,
      "grad_norm": 0.15755245023632508,
      "kl": 0.0003324151039123535,
      "learning_rate": 3.5061728395061724e-08,
      "loss": 0.0,
      "reward": 1.7964286282658577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 284
    },
    {
      "completion_length": 244.0982265472412,
      "epoch": 0.0479483633010604,
      "grad_norm": 0.248909793697624,
      "kl": 0.0002993345260620117,
      "learning_rate": 3.530864197530864e-08,
      "loss": 0.0,
      "reward": 1.7017857879400253,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7062500305473804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 286
    },
    {
      "completion_length": 245.4866180419922,
      "epoch": 0.04828366654092795,
      "grad_norm": 0.2979174520331835,
      "kl": 0.0003222227096557617,
      "learning_rate": 3.5555555555555554e-08,
      "loss": 0.0,
      "reward": 1.7160715088248253,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7205357514321804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 288
    },
    {
      "completion_length": 245.1696548461914,
      "epoch": 0.04861896978079551,
      "grad_norm": 0.2121228295030136,
      "kl": 0.0003127455711364746,
      "learning_rate": 3.580246913580247e-08,
      "loss": 0.0,
      "reward": 1.7750000730156898,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 290
    },
    {
      "completion_length": 254.25001621246338,
      "epoch": 0.04895427302066306,
      "grad_norm": 0.2591013783701026,
      "kl": 0.0003070831298828125,
      "learning_rate": 3.6049382716049384e-08,
      "loss": 0.0,
      "reward": 1.7107143625617027,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7107143253087997,
      "rewards/format_reward_func": 1.0,
      "step": 292
    },
    {
      "completion_length": 248.9285831451416,
      "epoch": 0.04928957626053062,
      "grad_norm": 0.25970339981154605,
      "kl": 0.00030744075775146484,
      "learning_rate": 3.62962962962963e-08,
      "loss": 0.0,
      "reward": 1.7142858132719994,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7142857424914837,
      "rewards/format_reward_func": 1.0,
      "step": 294
    },
    {
      "completion_length": 245.6607255935669,
      "epoch": 0.04962487950039817,
      "grad_norm": 0.1693241175450565,
      "kl": 0.0003269314765930176,
      "learning_rate": 3.6543209876543213e-08,
      "loss": 0.0,
      "reward": 1.7928572073578835,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 296
    },
    {
      "completion_length": 244.47768878936768,
      "epoch": 0.04996018274026573,
      "grad_norm": 0.28498148563751646,
      "kl": 0.00032448768615722656,
      "learning_rate": 3.679012345679012e-08,
      "loss": 0.0,
      "reward": 1.7232143804430962,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7276786062866449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 298
    },
    {
      "completion_length": 249.23662090301514,
      "epoch": 0.05029548598013328,
      "grad_norm": 0.17194495222367387,
      "kl": 0.00031006336212158203,
      "learning_rate": 3.7037037037037036e-08,
      "loss": 0.0,
      "reward": 1.7250000908970833,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.725000036880374,
      "rewards/format_reward_func": 1.0,
      "step": 300
    },
    {
      "completion_length": 241.9509038925171,
      "epoch": 0.05063078922000084,
      "grad_norm": 0.09369687516121022,
      "kl": 0.0003318190574645996,
      "learning_rate": 3.7283950617283945e-08,
      "loss": 0.0,
      "reward": 1.7589286640286446,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7633928805589676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 302
    },
    {
      "completion_length": 245.50893878936768,
      "epoch": 0.05096609245986839,
      "grad_norm": 0.20374576311730827,
      "kl": 0.0003292560577392578,
      "learning_rate": 3.753086419753086e-08,
      "loss": 0.0,
      "reward": 1.762500062584877,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643182307482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 304
    },
    {
      "completion_length": 246.89286708831787,
      "epoch": 0.05130139569973595,
      "grad_norm": 0.33707103491103274,
      "kl": 0.00032824277877807617,
      "learning_rate": 3.7777777777777774e-08,
      "loss": 0.0,
      "reward": 1.7375000640749931,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7419643141329288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 306
    },
    {
      "completion_length": 251.9910831451416,
      "epoch": 0.0516366989396035,
      "grad_norm": 0.16043020844799977,
      "kl": 0.0003122687339782715,
      "learning_rate": 3.802469135802469e-08,
      "loss": 0.0,
      "reward": 1.7214286476373672,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7214286103844643,
      "rewards/format_reward_func": 1.0,
      "step": 308
    },
    {
      "completion_length": 252.6696548461914,
      "epoch": 0.05197200217947106,
      "grad_norm": 0.32476087022862743,
      "kl": 0.00031131505966186523,
      "learning_rate": 3.8271604938271604e-08,
      "loss": 0.0,
      "reward": 1.7142858058214188,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7232143059372902,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 310
    },
    {
      "completion_length": 250.33037090301514,
      "epoch": 0.05230730541933862,
      "grad_norm": 0.20662703392624346,
      "kl": 0.00035762786865234375,
      "learning_rate": 3.851851851851852e-08,
      "loss": 0.0,
      "reward": 1.7232143431901932,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.727678619325161,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 312
    },
    {
      "completion_length": 245.37947368621826,
      "epoch": 0.05264260865920617,
      "grad_norm": 0.3116052051179476,
      "kl": 0.00034052133560180664,
      "learning_rate": 3.876543209876543e-08,
      "loss": 0.0,
      "reward": 1.7267857939004898,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7312500216066837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 314
    },
    {
      "completion_length": 254.11608219146729,
      "epoch": 0.05297791189907373,
      "grad_norm": 0.23588516688376734,
      "kl": 0.00032001733779907227,
      "learning_rate": 3.901234567901234e-08,
      "loss": 0.0,
      "reward": 1.7589286491274834,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.772321455180645,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 316
    },
    {
      "completion_length": 250.17858409881592,
      "epoch": 0.05331321513894128,
      "grad_norm": 0.21352114622407678,
      "kl": 0.0003173947334289551,
      "learning_rate": 3.9259259259259256e-08,
      "loss": 0.0,
      "reward": 1.708928644657135,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7133928779512644,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 318
    },
    {
      "completion_length": 241.33036708831787,
      "epoch": 0.05364851837880884,
      "grad_norm": 0.19064104806150775,
      "kl": 0.000352323055267334,
      "learning_rate": 3.950617283950617e-08,
      "loss": 0.0,
      "reward": 1.782142922282219,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 320
    },
    {
      "completion_length": 247.5714406967163,
      "epoch": 0.05398382161867639,
      "grad_norm": 0.09471434497682275,
      "kl": 0.00035202503204345703,
      "learning_rate": 3.9753086419753086e-08,
      "loss": 0.0,
      "reward": 1.7357143759727478,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357142977416515,
      "rewards/format_reward_func": 1.0,
      "step": 322
    },
    {
      "completion_length": 247.8392972946167,
      "epoch": 0.05431912485854395,
      "grad_norm": 0.16439367857374707,
      "kl": 0.0003255605697631836,
      "learning_rate": 4e-08,
      "loss": 0.0,
      "reward": 1.7250000685453415,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7339286040514708,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 324
    },
    {
      "completion_length": 244.4821538925171,
      "epoch": 0.0546544280984115,
      "grad_norm": 0.28376400955422515,
      "kl": 0.0003072023391723633,
      "learning_rate": 4.0246913580246916e-08,
      "loss": 0.0,
      "reward": 1.7357143610715866,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7446429003030062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 326
    },
    {
      "completion_length": 241.10715293884277,
      "epoch": 0.05498973133827906,
      "grad_norm": 0.21232563662149967,
      "kl": 0.00036221742630004883,
      "learning_rate": 4.049382716049383e-08,
      "loss": 0.0,
      "reward": 1.76071435213089,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 328
    },
    {
      "completion_length": 244.8259048461914,
      "epoch": 0.05532503457814661,
      "grad_norm": 0.31284424481049833,
      "kl": 0.00032901763916015625,
      "learning_rate": 4.0740740740740745e-08,
      "loss": 0.0,
      "reward": 1.7642857655882835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 330
    },
    {
      "completion_length": 233.58036708831787,
      "epoch": 0.05566033781801417,
      "grad_norm": 0.14740794798794307,
      "kl": 0.00032466650009155273,
      "learning_rate": 4.098765432098765e-08,
      "loss": 0.0,
      "reward": 1.7500000670552254,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000484287739,
      "rewards/format_reward_func": 1.0,
      "step": 332
    },
    {
      "completion_length": 247.05358219146729,
      "epoch": 0.05599564105788172,
      "grad_norm": 0.21326826487289274,
      "kl": 0.0003064870834350586,
      "learning_rate": 4.123456790123456e-08,
      "loss": 0.0,
      "reward": 1.7660714760422707,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7705357521772385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 334
    },
    {
      "completion_length": 252.5134038925171,
      "epoch": 0.05633094429774928,
      "grad_norm": 0.2820975150743322,
      "kl": 0.0003514289855957031,
      "learning_rate": 4.1481481481481476e-08,
      "loss": 0.0,
      "reward": 1.737500086426735,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7419643178582191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 336
    },
    {
      "completion_length": 245.87054538726807,
      "epoch": 0.05666624753761683,
      "grad_norm": 0.28892632014326425,
      "kl": 0.00032830238342285156,
      "learning_rate": 4.172839506172839e-08,
      "loss": 0.0,
      "reward": 1.7392857894301414,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.739285733550787,
      "rewards/format_reward_func": 1.0,
      "step": 338
    },
    {
      "completion_length": 257.6428699493408,
      "epoch": 0.05700155077748439,
      "grad_norm": 0.20801041134465126,
      "kl": 0.0003235936164855957,
      "learning_rate": 4.1975308641975306e-08,
      "loss": 0.0,
      "reward": 1.7035714983940125,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.712500024586916,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 340
    },
    {
      "completion_length": 239.0446538925171,
      "epoch": 0.057336854017351946,
      "grad_norm": 0.20466302680456666,
      "kl": 0.00033277273178100586,
      "learning_rate": 4.222222222222222e-08,
      "loss": 0.0,
      "reward": 1.7785714864730835,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7875000350177288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 342
    },
    {
      "completion_length": 247.04911994934082,
      "epoch": 0.0576721572572195,
      "grad_norm": 0.2544168425529169,
      "kl": 0.0003313422203063965,
      "learning_rate": 4.2469135802469136e-08,
      "loss": 0.0,
      "reward": 1.7500000670552254,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 344
    },
    {
      "completion_length": 247.80805110931396,
      "epoch": 0.058007460497087056,
      "grad_norm": 0.28697432321700045,
      "kl": 0.0003412961959838867,
      "learning_rate": 4.271604938271605e-08,
      "loss": 0.0,
      "reward": 1.7464286461472511,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7464286126196384,
      "rewards/format_reward_func": 1.0,
      "step": 346
    },
    {
      "completion_length": 241.80358409881592,
      "epoch": 0.05834276373695461,
      "grad_norm": 0.2989078442481178,
      "kl": 0.0003070235252380371,
      "learning_rate": 4.296296296296296e-08,
      "loss": 0.0,
      "reward": 1.771428644657135,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285887777805,
      "rewards/format_reward_func": 1.0,
      "step": 348
    },
    {
      "completion_length": 241.81697463989258,
      "epoch": 0.058678066976822166,
      "grad_norm": 0.18283142606513708,
      "kl": 0.0003039836883544922,
      "learning_rate": 4.3209876543209874e-08,
      "loss": 0.0,
      "reward": 1.7857143431901932,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 350
    },
    {
      "completion_length": 245.33929634094238,
      "epoch": 0.05901337021668972,
      "grad_norm": 0.2445142243174735,
      "kl": 0.0003504753112792969,
      "learning_rate": 4.345679012345679e-08,
      "loss": 0.0,
      "reward": 1.771428644657135,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 352
    },
    {
      "completion_length": 246.90179920196533,
      "epoch": 0.059348673456557276,
      "grad_norm": 0.16569502991379484,
      "kl": 0.00032913684844970703,
      "learning_rate": 4.37037037037037e-08,
      "loss": 0.0,
      "reward": 1.7571429312229156,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571429088711739,
      "rewards/format_reward_func": 1.0,
      "step": 354
    },
    {
      "completion_length": 240.02679634094238,
      "epoch": 0.05968397669642483,
      "grad_norm": 0.2585568390691014,
      "kl": 0.0003421306610107422,
      "learning_rate": 4.395061728395062e-08,
      "loss": 0.0,
      "reward": 1.758928656578064,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7633928880095482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 356
    },
    {
      "completion_length": 245.41072463989258,
      "epoch": 0.060019279936292386,
      "grad_norm": 0.13192074596224,
      "kl": 0.00033915042877197266,
      "learning_rate": 4.419753086419753e-08,
      "loss": 0.0,
      "reward": 1.7678571939468384,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 358
    },
    {
      "completion_length": 235.59375953674316,
      "epoch": 0.06035458317615994,
      "grad_norm": 0.1860175287251393,
      "kl": 0.00028246641159057617,
      "learning_rate": 4.444444444444444e-08,
      "loss": 0.0,
      "reward": 1.7875000536441803,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7919643186032772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 360
    },
    {
      "completion_length": 250.3169755935669,
      "epoch": 0.060689886416027496,
      "grad_norm": 0.296173081498714,
      "kl": 0.0003261566162109375,
      "learning_rate": 4.4691358024691356e-08,
      "loss": 0.0,
      "reward": 1.7500000521540642,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 362
    },
    {
      "completion_length": 245.74108028411865,
      "epoch": 0.06102518965589505,
      "grad_norm": 0.17763701007983315,
      "kl": 0.0003126859664916992,
      "learning_rate": 4.493827160493827e-08,
      "loss": 0.0,
      "reward": 1.7714286372065544,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7803571783006191,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 364
    },
    {
      "completion_length": 249.30358219146729,
      "epoch": 0.061360492895762606,
      "grad_norm": 0.2121971068500702,
      "kl": 0.00033473968505859375,
      "learning_rate": 4.518518518518518e-08,
      "loss": 0.0,
      "reward": 1.751785770058632,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500387430191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 366
    },
    {
      "completion_length": 243.9285831451416,
      "epoch": 0.06169579613563016,
      "grad_norm": 0.15114489649405735,
      "kl": 0.00029456615447998047,
      "learning_rate": 4.5432098765432094e-08,
      "loss": 0.0,
      "reward": 1.7892857640981674,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 368
    },
    {
      "completion_length": 248.87500953674316,
      "epoch": 0.062031099375497716,
      "grad_norm": 0.2841277990992117,
      "kl": 0.00031626224517822266,
      "learning_rate": 4.567901234567901e-08,
      "loss": 0.0,
      "reward": 1.741071492433548,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7544643115252256,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 370
    },
    {
      "completion_length": 244.41965293884277,
      "epoch": 0.062366402615365274,
      "grad_norm": 0.4126531872874137,
      "kl": 0.0003432035446166992,
      "learning_rate": 4.592592592592592e-08,
      "loss": 0.0,
      "reward": 1.7303572222590446,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7348214574158192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 372
    },
    {
      "completion_length": 249.9821548461914,
      "epoch": 0.06270170585523283,
      "grad_norm": 0.25716336513716315,
      "kl": 0.0003498196601867676,
      "learning_rate": 4.617283950617284e-08,
      "loss": 0.0,
      "reward": 1.7821429446339607,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.782142873853445,
      "rewards/format_reward_func": 1.0,
      "step": 374
    },
    {
      "completion_length": 258.8259086608887,
      "epoch": 0.06303700909510038,
      "grad_norm": 0.209967257582214,
      "kl": 0.00030928850173950195,
      "learning_rate": 4.641975308641975e-08,
      "loss": 0.0,
      "reward": 1.696428656578064,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7053571715950966,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 376
    },
    {
      "completion_length": 243.48661613464355,
      "epoch": 0.06337231233496794,
      "grad_norm": 0.357820262167908,
      "kl": 0.00035440921783447266,
      "learning_rate": 4.666666666666667e-08,
      "loss": 0.0,
      "reward": 1.7553572058677673,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.759821455925703,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 378
    },
    {
      "completion_length": 250.60269165039062,
      "epoch": 0.0637076155748355,
      "grad_norm": 0.2806225729005327,
      "kl": 0.0003170967102050781,
      "learning_rate": 4.691358024691358e-08,
      "loss": 0.0,
      "reward": 1.760714367032051,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143092900515,
      "rewards/format_reward_func": 1.0,
      "step": 380
    },
    {
      "completion_length": 243.09376049041748,
      "epoch": 0.06404291881470305,
      "grad_norm": 0.18647090907402403,
      "kl": 0.0003414154052734375,
      "learning_rate": 4.716049382716049e-08,
      "loss": 0.0,
      "reward": 1.725000061094761,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7339285928755999,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 382
    },
    {
      "completion_length": 246.5089406967163,
      "epoch": 0.0643782220545706,
      "grad_norm": 0.6344869457901994,
      "kl": 0.0003355741500854492,
      "learning_rate": 4.7407407407407405e-08,
      "loss": 0.0,
      "reward": 1.725000075995922,
      "reward_std": 0.08586296439170837,
      "rewards/equation_reward_func": 0.733928594738245,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 384
    },
    {
      "completion_length": 242.91519260406494,
      "epoch": 0.06471352529443816,
      "grad_norm": 0.3172293127272545,
      "kl": 0.0003371238708496094,
      "learning_rate": 4.765432098765432e-08,
      "loss": 0.0,
      "reward": 1.762500062584877,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643200933933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 386
    },
    {
      "completion_length": 238.28126335144043,
      "epoch": 0.06504882853430571,
      "grad_norm": 0.2713314165205857,
      "kl": 0.0003418922424316406,
      "learning_rate": 4.790123456790123e-08,
      "loss": 0.0,
      "reward": 1.7482143566012383,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7526786141097546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 388
    },
    {
      "completion_length": 243.88393783569336,
      "epoch": 0.06538413177417327,
      "grad_norm": 0.15007989523396975,
      "kl": 0.0003464221954345703,
      "learning_rate": 4.814814814814814e-08,
      "loss": 0.0,
      "reward": 1.8035714998841286,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 390
    },
    {
      "completion_length": 247.31697845458984,
      "epoch": 0.06571943501404082,
      "grad_norm": 0.4454137762015419,
      "kl": 0.0003800392150878906,
      "learning_rate": 4.839506172839506e-08,
      "loss": 0.0,
      "reward": 1.7357143759727478,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.7446428947150707,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 392
    },
    {
      "completion_length": 252.01340579986572,
      "epoch": 0.06605473825390838,
      "grad_norm": 0.5044913641727443,
      "kl": 0.00032645463943481445,
      "learning_rate": 4.864197530864197e-08,
      "loss": 0.0,
      "reward": 1.698214367032051,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7026786152273417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 394
    },
    {
      "completion_length": 241.75001430511475,
      "epoch": 0.06639004149377593,
      "grad_norm": 0.19490645814902496,
      "kl": 0.0003345012664794922,
      "learning_rate": 4.888888888888889e-08,
      "loss": 0.0,
      "reward": 1.733928643167019,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7383928932249546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 396
    },
    {
      "completion_length": 248.08036994934082,
      "epoch": 0.06672534473364349,
      "grad_norm": 0.2114488097393688,
      "kl": 0.0003476142883300781,
      "learning_rate": 4.91358024691358e-08,
      "loss": 0.0,
      "reward": 1.775000087916851,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 398
    },
    {
      "completion_length": 249.24554920196533,
      "epoch": 0.06706064797351105,
      "grad_norm": 0.3736813422927126,
      "kl": 0.0003236532211303711,
      "learning_rate": 4.938271604938271e-08,
      "loss": 0.0,
      "reward": 1.7535715028643608,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 400
    },
    {
      "completion_length": 250.5491189956665,
      "epoch": 0.0673959512133786,
      "grad_norm": 0.09379494996574526,
      "kl": 0.0003262758255004883,
      "learning_rate": 4.9629629629629626e-08,
      "loss": 0.0,
      "reward": 1.8142857626080513,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8232143111526966,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 402
    },
    {
      "completion_length": 242.2366180419922,
      "epoch": 0.06773125445324615,
      "grad_norm": 0.21504913509015106,
      "kl": 0.0003097057342529297,
      "learning_rate": 4.987654320987654e-08,
      "loss": 0.0,
      "reward": 1.744642935693264,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 404
    },
    {
      "completion_length": 249.0446548461914,
      "epoch": 0.0680665576931137,
      "grad_norm": 0.25395455275495804,
      "kl": 0.0003694295883178711,
      "learning_rate": 5.0123456790123455e-08,
      "loss": 0.0,
      "reward": 1.7357143610715866,
      "reward_std": 0.0909137288108468,
      "rewards/equation_reward_func": 0.7535714469850063,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 406
    },
    {
      "completion_length": 238.26340579986572,
      "epoch": 0.06840186093298127,
      "grad_norm": 0.33754900952521866,
      "kl": 0.0003122687339782715,
      "learning_rate": 5.037037037037037e-08,
      "loss": 0.0,
      "reward": 1.7535715103149414,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7535714600235224,
      "rewards/format_reward_func": 1.0,
      "step": 408
    },
    {
      "completion_length": 249.81697845458984,
      "epoch": 0.06873716417284882,
      "grad_norm": 0.1367116643336248,
      "kl": 0.0003184080123901367,
      "learning_rate": 5.0617283950617285e-08,
      "loss": 0.0,
      "reward": 1.728571504354477,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714522004128,
      "rewards/format_reward_func": 1.0,
      "step": 410
    },
    {
      "completion_length": 246.19197463989258,
      "epoch": 0.06907246741271637,
      "grad_norm": 0.23975839196424972,
      "kl": 0.00031936168670654297,
      "learning_rate": 5.08641975308642e-08,
      "loss": 0.0,
      "reward": 1.7946428880095482,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.799107177183032,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 412
    },
    {
      "completion_length": 245.99108505249023,
      "epoch": 0.06940777065258392,
      "grad_norm": 0.35602425619406325,
      "kl": 0.00033867359161376953,
      "learning_rate": 5.1111111111111114e-08,
      "loss": 0.0,
      "reward": 1.8196429014205933,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8241071589291096,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 414
    },
    {
      "completion_length": 250.1562614440918,
      "epoch": 0.06974307389245149,
      "grad_norm": 0.24837916834782592,
      "kl": 0.0003298521041870117,
      "learning_rate": 5.1358024691358016e-08,
      "loss": 0.0,
      "reward": 1.735714353621006,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.7446428872644901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 416
    },
    {
      "completion_length": 251.49107837677002,
      "epoch": 0.07007837713231904,
      "grad_norm": 0.35602402754236945,
      "kl": 0.0003180503845214844,
      "learning_rate": 5.160493827160493e-08,
      "loss": 0.0,
      "reward": 1.751785784959793,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7562500275671482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 418
    },
    {
      "completion_length": 238.81251049041748,
      "epoch": 0.0704136803721866,
      "grad_norm": 0.3111388382821028,
      "kl": 0.0003452301025390625,
      "learning_rate": 5.1851851851851846e-08,
      "loss": 0.0,
      "reward": 1.7392857894301414,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7392857633531094,
      "rewards/format_reward_func": 1.0,
      "step": 420
    },
    {
      "completion_length": 241.9241180419922,
      "epoch": 0.07074898361205414,
      "grad_norm": 0.22274741286610353,
      "kl": 0.0003383755683898926,
      "learning_rate": 5.209876543209876e-08,
      "loss": 0.0,
      "reward": 1.7571429386734962,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428753435612,
      "rewards/format_reward_func": 1.0,
      "step": 422
    },
    {
      "completion_length": 244.17411708831787,
      "epoch": 0.07108428685192171,
      "grad_norm": 0.19095279674450452,
      "kl": 0.00030684471130371094,
      "learning_rate": 5.2345679012345675e-08,
      "loss": 0.0,
      "reward": 1.703571505844593,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7125000283122063,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 424
    },
    {
      "completion_length": 245.1562623977661,
      "epoch": 0.07141959009178926,
      "grad_norm": 0.21128512978496952,
      "kl": 0.0003224611282348633,
      "learning_rate": 5.259259259259259e-08,
      "loss": 0.0,
      "reward": 1.8107143491506577,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.810714315623045,
      "rewards/format_reward_func": 1.0,
      "step": 426
    },
    {
      "completion_length": 232.4821548461914,
      "epoch": 0.07175489333165681,
      "grad_norm": 0.2414527596562914,
      "kl": 0.0003307461738586426,
      "learning_rate": 5.2839506172839505e-08,
      "loss": 0.0,
      "reward": 1.7964286357164383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964286096394062,
      "rewards/format_reward_func": 1.0,
      "step": 428
    },
    {
      "completion_length": 253.75447750091553,
      "epoch": 0.07209019657152438,
      "grad_norm": 0.2552537262189747,
      "kl": 0.00033414363861083984,
      "learning_rate": 5.308641975308642e-08,
      "loss": 0.0,
      "reward": 1.707142949104309,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7160714641213417,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 430
    },
    {
      "completion_length": 251.34822368621826,
      "epoch": 0.07242549981139193,
      "grad_norm": 0.1560573542361731,
      "kl": 0.0003288388252258301,
      "learning_rate": 5.3333333333333334e-08,
      "loss": 0.0,
      "reward": 1.742857202887535,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 432
    },
    {
      "completion_length": 247.5044755935669,
      "epoch": 0.07276080305125948,
      "grad_norm": 0.30241987654645447,
      "kl": 0.0003216266632080078,
      "learning_rate": 5.358024691358024e-08,
      "loss": 0.0,
      "reward": 1.7660715207457542,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7705357484519482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 434
    },
    {
      "completion_length": 246.5312614440918,
      "epoch": 0.07309610629112703,
      "grad_norm": 0.33054557128364087,
      "kl": 0.0003314018249511719,
      "learning_rate": 5.382716049382716e-08,
      "loss": 0.0,
      "reward": 1.7250000536441803,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7339286021888256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 436
    },
    {
      "completion_length": 243.68304920196533,
      "epoch": 0.0734314095309946,
      "grad_norm": 0.19851593896021988,
      "kl": 0.00032967329025268555,
      "learning_rate": 5.407407407407407e-08,
      "loss": 0.0,
      "reward": 1.7107143625617027,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7107143215835094,
      "rewards/format_reward_func": 1.0,
      "step": 438
    },
    {
      "completion_length": 249.64287185668945,
      "epoch": 0.07376671277086215,
      "grad_norm": 0.1713552603952388,
      "kl": 0.0003451108932495117,
      "learning_rate": 5.432098765432099e-08,
      "loss": 0.0,
      "reward": 1.7750000357627869,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7839285992085934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 440
    },
    {
      "completion_length": 240.55358409881592,
      "epoch": 0.0741020160107297,
      "grad_norm": 0.21027297014803956,
      "kl": 0.000316619873046875,
      "learning_rate": 5.45679012345679e-08,
      "loss": 0.0,
      "reward": 1.7821429297327995,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428962051868,
      "rewards/format_reward_func": 1.0,
      "step": 442
    },
    {
      "completion_length": 245.67411708831787,
      "epoch": 0.07443731925059725,
      "grad_norm": 0.33966947103680717,
      "kl": 0.00033915042877197266,
      "learning_rate": 5.481481481481482e-08,
      "loss": 0.0,
      "reward": 1.7660714983940125,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7705357447266579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 444
    },
    {
      "completion_length": 254.05358028411865,
      "epoch": 0.07477262249046482,
      "grad_norm": 0.4185328389589341,
      "kl": 0.00033539533615112305,
      "learning_rate": 5.5061728395061725e-08,
      "loss": 0.0,
      "reward": 1.6982143595814705,
      "reward_std": 0.09848987031728029,
      "rewards/equation_reward_func": 0.7116071805357933,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 446
    },
    {
      "completion_length": 235.8437614440918,
      "epoch": 0.07510792573033237,
      "grad_norm": 0.31327344899543313,
      "kl": 0.00031751394271850586,
      "learning_rate": 5.530864197530864e-08,
      "loss": 0.0,
      "reward": 1.785714365541935,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7857143245637417,
      "rewards/format_reward_func": 1.0,
      "step": 448
    },
    {
      "completion_length": 244.25894260406494,
      "epoch": 0.07544322897019992,
      "grad_norm": 0.168519024379435,
      "kl": 0.00033473968505859375,
      "learning_rate": 5.555555555555555e-08,
      "loss": 0.0,
      "reward": 1.7428572103381157,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 450
    },
    {
      "completion_length": 241.50447463989258,
      "epoch": 0.07577853221006747,
      "grad_norm": 0.2579852036649518,
      "kl": 0.0003389120101928711,
      "learning_rate": 5.580246913580246e-08,
      "loss": 0.0,
      "reward": 1.7750000730156898,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7750000394880772,
      "rewards/format_reward_func": 1.0,
      "step": 452
    },
    {
      "completion_length": 246.60715103149414,
      "epoch": 0.07611383544993504,
      "grad_norm": 0.2946907618719622,
      "kl": 0.0003083944320678711,
      "learning_rate": 5.604938271604938e-08,
      "loss": 0.0,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 454
    },
    {
      "completion_length": 243.24554634094238,
      "epoch": 0.07644913868980259,
      "grad_norm": 0.4449621858359876,
      "kl": 0.0003170967102050781,
      "learning_rate": 5.629629629629629e-08,
      "loss": 0.0,
      "reward": 1.7464286386966705,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.746428607031703,
      "rewards/format_reward_func": 1.0,
      "step": 456
    },
    {
      "completion_length": 244.9017972946167,
      "epoch": 0.07678444192967014,
      "grad_norm": 0.27342841301104115,
      "kl": 0.00032132863998413086,
      "learning_rate": 5.654320987654321e-08,
      "loss": 0.0,
      "reward": 1.778571479022503,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 458
    },
    {
      "completion_length": 246.16518783569336,
      "epoch": 0.0771197451695377,
      "grad_norm": 0.19558656358425155,
      "kl": 0.0003364682197570801,
      "learning_rate": 5.679012345679012e-08,
      "loss": 0.0,
      "reward": 1.7714286297559738,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7803571783006191,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 460
    },
    {
      "completion_length": 241.52233123779297,
      "epoch": 0.07745504840940526,
      "grad_norm": 0.169838691432525,
      "kl": 0.00031113624572753906,
      "learning_rate": 5.703703703703704e-08,
      "loss": 0.0,
      "reward": 1.7803572416305542,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7848214581608772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 462
    },
    {
      "completion_length": 253.54019165039062,
      "epoch": 0.07779035164927281,
      "grad_norm": 0.19518278420827403,
      "kl": 0.00032633543014526367,
      "learning_rate": 5.728395061728395e-08,
      "loss": 0.0,
      "reward": 1.7000001072883606,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7089285887777805,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 464
    },
    {
      "completion_length": 236.49554634094238,
      "epoch": 0.07812565488914036,
      "grad_norm": 0.21490568594797874,
      "kl": 0.00033402442932128906,
      "learning_rate": 5.7530864197530866e-08,
      "loss": 0.0,
      "reward": 1.8160714954137802,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8205357305705547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 466
    },
    {
      "completion_length": 244.2857255935669,
      "epoch": 0.07846095812900793,
      "grad_norm": 0.11754674459427326,
      "kl": 0.00033402442932128906,
      "learning_rate": 5.7777777777777775e-08,
      "loss": 0.0,
      "reward": 1.8017857447266579,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8062500320374966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 468
    },
    {
      "completion_length": 244.0848331451416,
      "epoch": 0.07879626136887548,
      "grad_norm": 0.23226873868439502,
      "kl": 0.00031507015228271484,
      "learning_rate": 5.802469135802469e-08,
      "loss": 0.0,
      "reward": 1.70000009983778,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7089286036789417,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 470
    },
    {
      "completion_length": 252.51340770721436,
      "epoch": 0.07913156460874303,
      "grad_norm": 0.2389300517769942,
      "kl": 0.00030809640884399414,
      "learning_rate": 5.8271604938271604e-08,
      "loss": 0.0,
      "reward": 1.7339286506175995,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7562500387430191,
      "rewards/format_reward_func": 0.9776785783469677,
      "step": 472
    },
    {
      "completion_length": 258.8705463409424,
      "epoch": 0.07946686784861058,
      "grad_norm": 0.21870083869491905,
      "kl": 0.00032639503479003906,
      "learning_rate": 5.851851851851851e-08,
      "loss": 0.0,
      "reward": 1.6946429386734962,
      "reward_std": 0.09848987124860287,
      "rewards/equation_reward_func": 0.7080357503145933,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 474
    },
    {
      "completion_length": 240.48661994934082,
      "epoch": 0.07980217108847815,
      "grad_norm": 0.22507093576887516,
      "kl": 0.00033414363861083984,
      "learning_rate": 5.876543209876543e-08,
      "loss": 0.0,
      "reward": 1.7571429088711739,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428772062063,
      "rewards/format_reward_func": 1.0,
      "step": 476
    },
    {
      "completion_length": 249.78572750091553,
      "epoch": 0.0801374743283457,
      "grad_norm": 0.2081506142291193,
      "kl": 0.000325620174407959,
      "learning_rate": 5.901234567901234e-08,
      "loss": 0.0,
      "reward": 1.7232143580913544,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7276785969734192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 478
    },
    {
      "completion_length": 248.21875762939453,
      "epoch": 0.08047277756821325,
      "grad_norm": 0.29807364642733086,
      "kl": 0.0003132820129394531,
      "learning_rate": 5.925925925925926e-08,
      "loss": 0.0,
      "reward": 1.6535715386271477,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.6535714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 480
    },
    {
      "completion_length": 243.1919765472412,
      "epoch": 0.08080808080808081,
      "grad_norm": 0.20914536774706063,
      "kl": 0.000324249267578125,
      "learning_rate": 5.950617283950617e-08,
      "loss": 0.0,
      "reward": 1.7928571924567223,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 482
    },
    {
      "completion_length": 247.24554443359375,
      "epoch": 0.08114338404794837,
      "grad_norm": 0.235749099937035,
      "kl": 0.00035053491592407227,
      "learning_rate": 5.975308641975308e-08,
      "loss": 0.0,
      "reward": 1.7571429312229156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 484
    },
    {
      "completion_length": 234.6830472946167,
      "epoch": 0.08147868728781592,
      "grad_norm": 0.28977720166304766,
      "kl": 0.0003306269645690918,
      "learning_rate": 6e-08,
      "loss": 0.0,
      "reward": 1.7428572252392769,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 486
    },
    {
      "completion_length": 249.5044755935669,
      "epoch": 0.08181399052768347,
      "grad_norm": 0.26207066498924353,
      "kl": 0.00032585859298706055,
      "learning_rate": 6.024691358024691e-08,
      "loss": 0.0,
      "reward": 1.716071479022503,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.729464327916503,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 488
    },
    {
      "completion_length": 247.4196548461914,
      "epoch": 0.08214929376755103,
      "grad_norm": 0.18458472700771195,
      "kl": 0.00031828880310058594,
      "learning_rate": 6.049382716049382e-08,
      "loss": 0.0,
      "reward": 1.776785783469677,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7812500298023224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 490
    },
    {
      "completion_length": 235.79018783569336,
      "epoch": 0.08248459700741859,
      "grad_norm": 0.2414137740023195,
      "kl": 0.00032210350036621094,
      "learning_rate": 6.074074074074074e-08,
      "loss": 0.0,
      "reward": 1.7892857566475868,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7892857510596514,
      "rewards/format_reward_func": 1.0,
      "step": 492
    },
    {
      "completion_length": 244.3482255935669,
      "epoch": 0.08281990024728614,
      "grad_norm": 0.223221207261264,
      "kl": 0.00033855438232421875,
      "learning_rate": 6.098765432098765e-08,
      "loss": 0.0,
      "reward": 1.721428669989109,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7214285992085934,
      "rewards/format_reward_func": 1.0,
      "step": 494
    },
    {
      "completion_length": 250.5089406967163,
      "epoch": 0.08315520348715369,
      "grad_norm": 0.16481525246585177,
      "kl": 0.0003097057342529297,
      "learning_rate": 6.123456790123457e-08,
      "loss": 0.0,
      "reward": 1.7410714849829674,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7544643208384514,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 496
    },
    {
      "completion_length": 243.70090198516846,
      "epoch": 0.08349050672702125,
      "grad_norm": 0.23407503260150778,
      "kl": 0.000316619873046875,
      "learning_rate": 6.148148148148148e-08,
      "loss": 0.0,
      "reward": 1.7178572192788124,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7178571745753288,
      "rewards/format_reward_func": 1.0,
      "step": 498
    },
    {
      "completion_length": 251.65625953674316,
      "epoch": 0.0838258099668888,
      "grad_norm": 0.21363418666430034,
      "kl": 0.00032788515090942383,
      "learning_rate": 6.172839506172839e-08,
      "loss": 0.0,
      "reward": 1.7857143506407738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143245637417,
      "rewards/format_reward_func": 1.0,
      "step": 500
    },
    {
      "completion_length": 234.915189743042,
      "epoch": 0.08416111320675636,
      "grad_norm": 0.2728153197023007,
      "kl": 0.00033909082412719727,
      "learning_rate": 6.19753086419753e-08,
      "loss": 0.0,
      "reward": 1.7392858266830444,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 1.0,
      "step": 502
    },
    {
      "completion_length": 245.33483409881592,
      "epoch": 0.08449641644662391,
      "grad_norm": 0.26575297012927046,
      "kl": 0.00031566619873046875,
      "learning_rate": 6.222222222222221e-08,
      "loss": 0.0,
      "reward": 1.7839286252856255,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7883928809314966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 504
    },
    {
      "completion_length": 248.352689743042,
      "epoch": 0.08483171968649147,
      "grad_norm": 0.1380297498926032,
      "kl": 0.0003465414047241211,
      "learning_rate": 6.246913580246913e-08,
      "loss": 0.0,
      "reward": 1.785714365541935,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143077999353,
      "rewards/format_reward_func": 1.0,
      "step": 506
    },
    {
      "completion_length": 241.34375858306885,
      "epoch": 0.08516702292635903,
      "grad_norm": 0.2441522470456551,
      "kl": 0.00032442808151245117,
      "learning_rate": 6.271604938271604e-08,
      "loss": 0.0,
      "reward": 1.785714365541935,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7857143059372902,
      "rewards/format_reward_func": 1.0,
      "step": 508
    },
    {
      "completion_length": 246.17412090301514,
      "epoch": 0.08550232616622658,
      "grad_norm": 0.21561174473883762,
      "kl": 0.0003211498260498047,
      "learning_rate": 6.296296296296296e-08,
      "loss": 0.0,
      "reward": 1.7250000685453415,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7250000294297934,
      "rewards/format_reward_func": 1.0,
      "step": 510
    },
    {
      "completion_length": 248.7991180419922,
      "epoch": 0.08583762940609414,
      "grad_norm": 0.1975704900557329,
      "kl": 0.00030416250228881836,
      "learning_rate": 6.320987654320987e-08,
      "loss": 0.0,
      "reward": 1.79464291036129,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7991071753203869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 512
    },
    {
      "completion_length": 244.66518878936768,
      "epoch": 0.0861729326459617,
      "grad_norm": 0.31873917088865067,
      "kl": 0.0003369450569152832,
      "learning_rate": 6.345679012345679e-08,
      "loss": 0.0,
      "reward": 1.6946429535746574,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.6991071738302708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 514
    },
    {
      "completion_length": 249.6785831451416,
      "epoch": 0.08650823588582925,
      "grad_norm": 0.1803545586351522,
      "kl": 0.00034105777740478516,
      "learning_rate": 6.37037037037037e-08,
      "loss": 0.0,
      "reward": 1.796428620815277,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285872876644,
      "rewards/format_reward_func": 1.0,
      "step": 516
    },
    {
      "completion_length": 238.2366180419922,
      "epoch": 0.0868435391256968,
      "grad_norm": 0.19465113178922888,
      "kl": 0.0003395676612854004,
      "learning_rate": 6.39506172839506e-08,
      "loss": 0.0,
      "reward": 1.803571492433548,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 518
    },
    {
      "completion_length": 243.0982255935669,
      "epoch": 0.08717884236556436,
      "grad_norm": 0.19674061927229855,
      "kl": 0.00034350156784057617,
      "learning_rate": 6.419753086419753e-08,
      "loss": 0.0,
      "reward": 1.7714286223053932,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714286055415869,
      "rewards/format_reward_func": 1.0,
      "step": 520
    },
    {
      "completion_length": 248.9866180419922,
      "epoch": 0.08751414560543191,
      "grad_norm": 0.282739041232047,
      "kl": 0.00033104419708251953,
      "learning_rate": 6.444444444444443e-08,
      "loss": 0.0,
      "reward": 1.7089286670088768,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7133928947150707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 522
    },
    {
      "completion_length": 237.4553689956665,
      "epoch": 0.08784944884529947,
      "grad_norm": 0.2913300631224223,
      "kl": 0.00030612945556640625,
      "learning_rate": 6.469135802469136e-08,
      "loss": 0.0,
      "reward": 1.7995536252856255,
      "reward_std": 0.06124049657955766,
      "rewards/equation_reward_func": 0.8026785850524902,
      "rewards/format_reward_func": 0.9968750029802322,
      "step": 524
    },
    {
      "completion_length": 248.6517972946167,
      "epoch": 0.08818475208516702,
      "grad_norm": 0.2988093886621688,
      "kl": 0.0003228187561035156,
      "learning_rate": 6.493827160493826e-08,
      "loss": 0.0,
      "reward": 1.7535714954137802,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 526
    },
    {
      "completion_length": 247.02233505249023,
      "epoch": 0.08852005532503458,
      "grad_norm": 0.31123073954807806,
      "kl": 0.00034630298614501953,
      "learning_rate": 6.518518518518518e-08,
      "loss": 0.0,
      "reward": 1.7321429252624512,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321428917348385,
      "rewards/format_reward_func": 1.0,
      "step": 528
    },
    {
      "completion_length": 240.74108123779297,
      "epoch": 0.08885535856490213,
      "grad_norm": 0.2844834020779769,
      "kl": 0.0003533363342285156,
      "learning_rate": 6.54320987654321e-08,
      "loss": 0.0,
      "reward": 1.8178571909666061,
      "reward_std": 0.0858629634603858,
      "rewards/equation_reward_func": 0.817857164889574,
      "rewards/format_reward_func": 1.0,
      "step": 530
    },
    {
      "completion_length": 247.12947845458984,
      "epoch": 0.08919066180476969,
      "grad_norm": 0.3683831687012005,
      "kl": 0.00037026405334472656,
      "learning_rate": 6.567901234567901e-08,
      "loss": 0.0,
      "reward": 1.7214286252856255,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7392857503145933,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 532
    },
    {
      "completion_length": 239.91965293884277,
      "epoch": 0.08952596504463724,
      "grad_norm": 0.23647114042655057,
      "kl": 0.00031244754791259766,
      "learning_rate": 6.592592592592592e-08,
      "loss": 0.0,
      "reward": 1.737500049173832,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643066823483,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 534
    },
    {
      "completion_length": 239.70537090301514,
      "epoch": 0.0898612682845048,
      "grad_norm": 0.2744914568098237,
      "kl": 0.00034332275390625,
      "learning_rate": 6.617283950617284e-08,
      "loss": 0.0,
      "reward": 1.7750000730156898,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 536
    },
    {
      "completion_length": 235.54465579986572,
      "epoch": 0.09019657152437235,
      "grad_norm": 0.24470360924408466,
      "kl": 0.00032085180282592773,
      "learning_rate": 6.641975308641975e-08,
      "loss": 0.0,
      "reward": 1.7714286521077156,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7714285925030708,
      "rewards/format_reward_func": 1.0,
      "step": 538
    },
    {
      "completion_length": 252.58929634094238,
      "epoch": 0.0905318747642399,
      "grad_norm": 0.19399993162010581,
      "kl": 0.0003203749656677246,
      "learning_rate": 6.666666666666667e-08,
      "loss": 0.0,
      "reward": 1.7142857536673546,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7142857536673546,
      "rewards/format_reward_func": 1.0,
      "step": 540
    },
    {
      "completion_length": 242.52679443359375,
      "epoch": 0.09086717800410747,
      "grad_norm": 0.1805765854243748,
      "kl": 0.0003228187561035156,
      "learning_rate": 6.691358024691358e-08,
      "loss": 0.0,
      "reward": 1.803571492433548,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 542
    },
    {
      "completion_length": 250.33930015563965,
      "epoch": 0.09120248124397502,
      "grad_norm": 0.2510144798661156,
      "kl": 0.00032061338424682617,
      "learning_rate": 6.71604938271605e-08,
      "loss": 0.0,
      "reward": 1.753571517765522,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7625000290572643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 544
    },
    {
      "completion_length": 248.883939743042,
      "epoch": 0.09153778448384257,
      "grad_norm": 0.2655297568235366,
      "kl": 0.0003451108932495117,
      "learning_rate": 6.74074074074074e-08,
      "loss": 0.0,
      "reward": 1.751785784959793,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500312924385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 546
    },
    {
      "completion_length": 249.2009038925171,
      "epoch": 0.09187308772371013,
      "grad_norm": 0.26878506330376745,
      "kl": 0.0003330707550048828,
      "learning_rate": 6.765432098765431e-08,
      "loss": 0.0,
      "reward": 1.7482143640518188,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 548
    },
    {
      "completion_length": 237.3348331451416,
      "epoch": 0.09220839096357769,
      "grad_norm": 0.2253938894358309,
      "kl": 0.00033724308013916016,
      "learning_rate": 6.790123456790123e-08,
      "loss": 0.0,
      "reward": 1.757142923772335,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571429051458836,
      "rewards/format_reward_func": 1.0,
      "step": 550
    },
    {
      "completion_length": 251.540189743042,
      "epoch": 0.09254369420344524,
      "grad_norm": 0.22765833106871078,
      "kl": 0.00031238794326782227,
      "learning_rate": 6.814814814814814e-08,
      "loss": 0.0,
      "reward": 1.762500062584877,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7669643145054579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 552
    },
    {
      "completion_length": 243.71876430511475,
      "epoch": 0.0928789974433128,
      "grad_norm": 0.2972380601698253,
      "kl": 0.00034296512603759766,
      "learning_rate": 6.839506172839506e-08,
      "loss": 0.0,
      "reward": 1.7625000700354576,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 554
    },
    {
      "completion_length": 241.85269260406494,
      "epoch": 0.09321430068318035,
      "grad_norm": 0.18903812105876086,
      "kl": 0.0003712177276611328,
      "learning_rate": 6.864197530864197e-08,
      "loss": 0.0,
      "reward": 1.7821429148316383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428999304771,
      "rewards/format_reward_func": 1.0,
      "step": 556
    },
    {
      "completion_length": 240.98661708831787,
      "epoch": 0.09354960392304791,
      "grad_norm": 0.2808854496545026,
      "kl": 0.00033342838287353516,
      "learning_rate": 6.888888888888889e-08,
      "loss": 0.0,
      "reward": 1.7535715028643608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714488476515,
      "rewards/format_reward_func": 1.0,
      "step": 558
    },
    {
      "completion_length": 247.7678680419922,
      "epoch": 0.09388490716291546,
      "grad_norm": 0.2433063193601212,
      "kl": 0.00035250186920166016,
      "learning_rate": 6.913580246913579e-08,
      "loss": 0.0,
      "reward": 1.7250000834465027,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000312924385,
      "rewards/format_reward_func": 1.0,
      "step": 560
    },
    {
      "completion_length": 239.64733219146729,
      "epoch": 0.09422021040278301,
      "grad_norm": 0.2328401577713564,
      "kl": 0.00031560659408569336,
      "learning_rate": 6.938271604938272e-08,
      "loss": 0.0,
      "reward": 1.793750062584877,
      "reward_std": 0.03914341004565358,
      "rewards/equation_reward_func": 0.795535746961832,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 562
    },
    {
      "completion_length": 243.7321548461914,
      "epoch": 0.09455551364265057,
      "grad_norm": 0.27133592508850835,
      "kl": 0.00033986568450927734,
      "learning_rate": 6.962962962962962e-08,
      "loss": 0.0,
      "reward": 1.8071429282426834,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8071428686380386,
      "rewards/format_reward_func": 1.0,
      "step": 564
    },
    {
      "completion_length": 255.19197368621826,
      "epoch": 0.09489081688251813,
      "grad_norm": 0.2894428896849047,
      "kl": 0.0003210902214050293,
      "learning_rate": 6.987654320987655e-08,
      "loss": 0.0,
      "reward": 1.7375000789761543,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.74196432903409,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 566
    },
    {
      "completion_length": 256.6250114440918,
      "epoch": 0.09522612012238568,
      "grad_norm": 0.24724991177580125,
      "kl": 0.00033986568450927734,
      "learning_rate": 7.012345679012345e-08,
      "loss": 0.0,
      "reward": 1.721428632736206,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7214285992085934,
      "rewards/format_reward_func": 1.0,
      "step": 568
    },
    {
      "completion_length": 246.13840103149414,
      "epoch": 0.09556142336225323,
      "grad_norm": 0.27228424645528715,
      "kl": 0.0003192424774169922,
      "learning_rate": 7.037037037037038e-08,
      "loss": 0.0,
      "reward": 1.778571479022503,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 570
    },
    {
      "completion_length": 240.89286422729492,
      "epoch": 0.0958967266021208,
      "grad_norm": 0.17697003476147488,
      "kl": 0.0003370046615600586,
      "learning_rate": 7.061728395061728e-08,
      "loss": 0.0,
      "reward": 1.8071429133415222,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8071428798139095,
      "rewards/format_reward_func": 1.0,
      "step": 572
    },
    {
      "completion_length": 245.89286613464355,
      "epoch": 0.09623202984198835,
      "grad_norm": 0.23976461206515592,
      "kl": 0.0003515481948852539,
      "learning_rate": 7.08641975308642e-08,
      "loss": 0.0,
      "reward": 1.7607143744826317,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.760714303702116,
      "rewards/format_reward_func": 1.0,
      "step": 574
    },
    {
      "completion_length": 257.33483505249023,
      "epoch": 0.0965673330818559,
      "grad_norm": 0.24968865023102368,
      "kl": 0.00032526254653930664,
      "learning_rate": 7.111111111111111e-08,
      "loss": 0.0,
      "reward": 1.7696429342031479,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7741071581840515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 576
    },
    {
      "completion_length": 236.85715579986572,
      "epoch": 0.09690263632172345,
      "grad_norm": 0.19104595177042635,
      "kl": 0.00033676624298095703,
      "learning_rate": 7.135802469135801e-08,
      "loss": 0.0,
      "reward": 1.8535714894533157,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8535714372992516,
      "rewards/format_reward_func": 1.0,
      "step": 578
    },
    {
      "completion_length": 250.72322750091553,
      "epoch": 0.09723793956159102,
      "grad_norm": 0.26928102942838894,
      "kl": 0.00033277273178100586,
      "learning_rate": 7.160493827160494e-08,
      "loss": 0.0,
      "reward": 1.7857143506407738,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 580
    },
    {
      "completion_length": 250.25001049041748,
      "epoch": 0.09757324280145857,
      "grad_norm": 0.2822786859772726,
      "kl": 0.0003406405448913574,
      "learning_rate": 7.185185185185184e-08,
      "loss": 0.0,
      "reward": 1.7553572058677673,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7598214596509933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 582
    },
    {
      "completion_length": 231.52679824829102,
      "epoch": 0.09790854604132612,
      "grad_norm": 0.3094974337963821,
      "kl": 0.00034546852111816406,
      "learning_rate": 7.209876543209877e-08,
      "loss": 0.0,
      "reward": 1.7392857745289803,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7482143137603998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 584
    },
    {
      "completion_length": 245.5134048461914,
      "epoch": 0.09824384928119367,
      "grad_norm": 0.29343542630310954,
      "kl": 0.0003381967544555664,
      "learning_rate": 7.234567901234567e-08,
      "loss": 0.0,
      "reward": 1.7714286372065544,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 586
    },
    {
      "completion_length": 246.6384038925171,
      "epoch": 0.09857915252106124,
      "grad_norm": 0.18806855527665864,
      "kl": 0.00032448768615722656,
      "learning_rate": 7.25925925925926e-08,
      "loss": 0.0,
      "reward": 1.7785715013742447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 588
    },
    {
      "completion_length": 251.34822940826416,
      "epoch": 0.09891445576092879,
      "grad_norm": 0.2674941278667219,
      "kl": 0.00032722949981689453,
      "learning_rate": 7.28395061728395e-08,
      "loss": 0.0,
      "reward": 1.667857214808464,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.6678571924567223,
      "rewards/format_reward_func": 1.0,
      "step": 590
    },
    {
      "completion_length": 239.42858123779297,
      "epoch": 0.09924975900079634,
      "grad_norm": 0.17254025141382712,
      "kl": 0.0003591179847717285,
      "learning_rate": 7.308641975308643e-08,
      "loss": 0.0,
      "reward": 1.7892857789993286,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857491970062,
      "rewards/format_reward_func": 1.0,
      "step": 592
    },
    {
      "completion_length": 247.8303689956665,
      "epoch": 0.0995850622406639,
      "grad_norm": 0.3699325611286679,
      "kl": 0.0003343820571899414,
      "learning_rate": 7.333333333333333e-08,
      "loss": 0.0,
      "reward": 1.7500000819563866,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 594
    },
    {
      "completion_length": 256.76340103149414,
      "epoch": 0.09992036548053146,
      "grad_norm": 0.3237781294586503,
      "kl": 0.00031179189682006836,
      "learning_rate": 7.358024691358024e-08,
      "loss": 0.0,
      "reward": 1.7428572252392769,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7428571823984385,
      "rewards/format_reward_func": 1.0,
      "step": 596
    },
    {
      "completion_length": 247.66965579986572,
      "epoch": 0.10025566872039901,
      "grad_norm": 0.3258600776488885,
      "kl": 0.0003464221954345703,
      "learning_rate": 7.382716049382716e-08,
      "loss": 0.0,
      "reward": 1.6928571984171867,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7017857357859612,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 598
    },
    {
      "completion_length": 249.6250114440918,
      "epoch": 0.10059097196026656,
      "grad_norm": 0.261689626399354,
      "kl": 0.0003151893615722656,
      "learning_rate": 7.407407407407407e-08,
      "loss": 0.0,
      "reward": 1.7535714954137802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714488476515,
      "rewards/format_reward_func": 1.0,
      "step": 600
    },
    {
      "completion_length": 242.17858123779297,
      "epoch": 0.10092627520013413,
      "grad_norm": 0.24581423709483807,
      "kl": 0.00032001733779907227,
      "learning_rate": 7.432098765432099e-08,
      "loss": 0.0,
      "reward": 1.7642857730388641,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 602
    },
    {
      "completion_length": 248.79912090301514,
      "epoch": 0.10126157844000168,
      "grad_norm": 0.3911909271722524,
      "kl": 0.0003205537796020508,
      "learning_rate": 7.456790123456789e-08,
      "loss": 0.0,
      "reward": 1.7214286774396896,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7303571663796902,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 604
    },
    {
      "completion_length": 240.64286994934082,
      "epoch": 0.10159688167986923,
      "grad_norm": 0.1934736455480886,
      "kl": 0.0003267526626586914,
      "learning_rate": 7.481481481481482e-08,
      "loss": 0.0,
      "reward": 1.7910714820027351,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7955357544124126,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 606
    },
    {
      "completion_length": 244.6160831451416,
      "epoch": 0.10193218491973678,
      "grad_norm": 0.22264172706402283,
      "kl": 0.00035625696182250977,
      "learning_rate": 7.506172839506172e-08,
      "loss": 0.0,
      "reward": 1.796428620815277,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 608
    },
    {
      "completion_length": 238.7232255935669,
      "epoch": 0.10226748815960435,
      "grad_norm": 0.1806141118501612,
      "kl": 0.00031882524490356445,
      "learning_rate": 7.530864197530865e-08,
      "loss": 0.0,
      "reward": 1.7785714864730835,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7875000238418579,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 610
    },
    {
      "completion_length": 239.7678689956665,
      "epoch": 0.1026027913994719,
      "grad_norm": 0.17213793905290295,
      "kl": 0.00036013126373291016,
      "learning_rate": 7.555555555555555e-08,
      "loss": 0.0,
      "reward": 1.6964286267757416,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.6964286155998707,
      "rewards/format_reward_func": 1.0,
      "step": 612
    },
    {
      "completion_length": 257.57590770721436,
      "epoch": 0.10293809463933945,
      "grad_norm": 0.3303583332562279,
      "kl": 0.0003082752227783203,
      "learning_rate": 7.580246913580246e-08,
      "loss": 0.0,
      "reward": 1.723214365541935,
      "reward_std": 0.09848987031728029,
      "rewards/equation_reward_func": 0.7276786044239998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 614
    },
    {
      "completion_length": 238.5312614440918,
      "epoch": 0.103273397879207,
      "grad_norm": 0.30225920527212524,
      "kl": 0.000339508056640625,
      "learning_rate": 7.604938271604938e-08,
      "loss": 0.0,
      "reward": 1.7571429386734962,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428716182709,
      "rewards/format_reward_func": 1.0,
      "step": 616
    },
    {
      "completion_length": 237.79911994934082,
      "epoch": 0.10360870111907457,
      "grad_norm": 0.1956188765585153,
      "kl": 0.00033271312713623047,
      "learning_rate": 7.629629629629629e-08,
      "loss": 0.0,
      "reward": 1.7821429148316383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 618
    },
    {
      "completion_length": 243.6651906967163,
      "epoch": 0.10394400435894212,
      "grad_norm": 0.21769077201626216,
      "kl": 0.0003572702407836914,
      "learning_rate": 7.654320987654321e-08,
      "loss": 0.0,
      "reward": 1.76071435213089,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143074274063,
      "rewards/format_reward_func": 1.0,
      "step": 620
    },
    {
      "completion_length": 252.90626335144043,
      "epoch": 0.10427930759880967,
      "grad_norm": 0.2545100315186332,
      "kl": 0.00031685829162597656,
      "learning_rate": 7.679012345679012e-08,
      "loss": 0.0,
      "reward": 1.7535715028643608,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7625000402331352,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 622
    },
    {
      "completion_length": 247.33929634094238,
      "epoch": 0.10461461083867724,
      "grad_norm": 0.3575182728529308,
      "kl": 0.0003485679626464844,
      "learning_rate": 7.703703703703704e-08,
      "loss": 0.0,
      "reward": 1.7625000774860382,
      "reward_std": 0.09343910776078701,
      "rewards/equation_reward_func": 0.766964316368103,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 624
    },
    {
      "completion_length": 256.81251525878906,
      "epoch": 0.10494991407854479,
      "grad_norm": 0.2494622615449293,
      "kl": 0.00032192468643188477,
      "learning_rate": 7.728395061728395e-08,
      "loss": 0.0,
      "reward": 1.7964286357164383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 626
    },
    {
      "completion_length": 245.0357265472412,
      "epoch": 0.10528521731841234,
      "grad_norm": 0.22256146612232233,
      "kl": 0.00034034252166748047,
      "learning_rate": 7.753086419753085e-08,
      "loss": 0.0,
      "reward": 1.6964286863803864,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.6964286006987095,
      "rewards/format_reward_func": 1.0,
      "step": 628
    },
    {
      "completion_length": 245.7634048461914,
      "epoch": 0.10562052055827989,
      "grad_norm": 0.2509199441030853,
      "kl": 0.00033855438232421875,
      "learning_rate": 7.777777777777778e-08,
      "loss": 0.0,
      "reward": 1.7732143476605415,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7776785977184772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 630
    },
    {
      "completion_length": 252.70090675354004,
      "epoch": 0.10595582379814746,
      "grad_norm": 0.2160039639235301,
      "kl": 0.0003249645233154297,
      "learning_rate": 7.802469135802468e-08,
      "loss": 0.0,
      "reward": 1.7446429207921028,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7491071820259094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 632
    },
    {
      "completion_length": 244.58929538726807,
      "epoch": 0.10629112703801501,
      "grad_norm": 0.2032292563867832,
      "kl": 0.00032579898834228516,
      "learning_rate": 7.82716049382716e-08,
      "loss": 0.0,
      "reward": 1.7160714864730835,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7205357402563095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 634
    },
    {
      "completion_length": 248.6026906967163,
      "epoch": 0.10662643027788256,
      "grad_norm": 0.27863712660799594,
      "kl": 0.00033843517303466797,
      "learning_rate": 7.851851851851851e-08,
      "loss": 0.0,
      "reward": 1.717857226729393,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.717857176437974,
      "rewards/format_reward_func": 1.0,
      "step": 636
    },
    {
      "completion_length": 245.8750114440918,
      "epoch": 0.10696173351775011,
      "grad_norm": 0.26748447053290475,
      "kl": 0.0003311634063720703,
      "learning_rate": 7.876543209876543e-08,
      "loss": 0.0,
      "reward": 1.8107143342494965,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8107143044471741,
      "rewards/format_reward_func": 1.0,
      "step": 638
    },
    {
      "completion_length": 237.75000953674316,
      "epoch": 0.10729703675761768,
      "grad_norm": 0.378121476129134,
      "kl": 0.00033354759216308594,
      "learning_rate": 7.901234567901234e-08,
      "loss": 0.0,
      "reward": 1.7196429446339607,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.724107176065445,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 640
    },
    {
      "completion_length": 254.9241189956665,
      "epoch": 0.10763233999748523,
      "grad_norm": 0.31986577676480604,
      "kl": 0.0003407001495361328,
      "learning_rate": 7.925925925925926e-08,
      "loss": 0.0,
      "reward": 1.7750000581145287,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7750000469386578,
      "rewards/format_reward_func": 1.0,
      "step": 642
    },
    {
      "completion_length": 251.55804824829102,
      "epoch": 0.10796764323735278,
      "grad_norm": 0.17222404932069518,
      "kl": 0.0003371238708496094,
      "learning_rate": 7.950617283950617e-08,
      "loss": 0.0,
      "reward": 1.7750000730156898,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7750000171363354,
      "rewards/format_reward_func": 1.0,
      "step": 644
    },
    {
      "completion_length": 238.65179920196533,
      "epoch": 0.10830294647722033,
      "grad_norm": 0.30833933519304063,
      "kl": 0.00034743547439575195,
      "learning_rate": 7.975308641975307e-08,
      "loss": 0.0,
      "reward": 1.7625000774860382,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7669643051922321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 646
    },
    {
      "completion_length": 255.2500123977661,
      "epoch": 0.1086382497170879,
      "grad_norm": 0.2769674980707656,
      "kl": 0.00034427642822265625,
      "learning_rate": 8e-08,
      "loss": 0.0,
      "reward": 1.7053572237491608,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7098214589059353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 648
    },
    {
      "completion_length": 240.78572750091553,
      "epoch": 0.10897355295695545,
      "grad_norm": 0.21938149745780547,
      "kl": 0.00034177303314208984,
      "learning_rate": 8.02469135802469e-08,
      "loss": 0.0,
      "reward": 1.7892857939004898,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 650
    },
    {
      "completion_length": 247.77233219146729,
      "epoch": 0.109308856196823,
      "grad_norm": 0.3524412693319839,
      "kl": 0.00033420324325561523,
      "learning_rate": 8.049382716049383e-08,
      "loss": 0.0,
      "reward": 1.7750000581145287,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7750000394880772,
      "rewards/format_reward_func": 1.0,
      "step": 652
    },
    {
      "completion_length": 239.45536994934082,
      "epoch": 0.10964415943669056,
      "grad_norm": 0.21691827702998728,
      "kl": 0.00033783912658691406,
      "learning_rate": 8.074074074074073e-08,
      "loss": 0.0,
      "reward": 1.762500062584877,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7669642977416515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 654
    },
    {
      "completion_length": 237.54018878936768,
      "epoch": 0.10997946267655812,
      "grad_norm": 0.36721379041357677,
      "kl": 0.00033032894134521484,
      "learning_rate": 8.098765432098766e-08,
      "loss": 0.0,
      "reward": 1.7375000566244125,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7419643104076385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 656
    },
    {
      "completion_length": 247.8303689956665,
      "epoch": 0.11031476591642567,
      "grad_norm": 0.22179464679431105,
      "kl": 0.0003440380096435547,
      "learning_rate": 8.123456790123456e-08,
      "loss": 0.0,
      "reward": 1.7375000715255737,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7419643178582191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 658
    },
    {
      "completion_length": 245.22322463989258,
      "epoch": 0.11065006915629322,
      "grad_norm": 0.28631903396643277,
      "kl": 0.0003451108932495117,
      "learning_rate": 8.148148148148149e-08,
      "loss": 0.0,
      "reward": 1.7678572088479996,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.776785746216774,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 660
    },
    {
      "completion_length": 239.6026906967163,
      "epoch": 0.11098537239616078,
      "grad_norm": 0.23815337312005594,
      "kl": 0.0003368854522705078,
      "learning_rate": 8.172839506172839e-08,
      "loss": 0.0,
      "reward": 1.7571429312229156,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 662
    },
    {
      "completion_length": 240.05804538726807,
      "epoch": 0.11132067563602834,
      "grad_norm": 0.00011184530104431835,
      "kl": 0.00032639503479003906,
      "learning_rate": 8.19753086419753e-08,
      "loss": 0.0,
      "reward": 1.7285715118050575,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7285714633762836,
      "rewards/format_reward_func": 1.0,
      "step": 664
    },
    {
      "completion_length": 238.4196548461914,
      "epoch": 0.11165597887589589,
      "grad_norm": 0.16343083171368733,
      "kl": 0.0003247857093811035,
      "learning_rate": 8.222222222222222e-08,
      "loss": 0.0,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857581377029,
      "rewards/format_reward_func": 1.0,
      "step": 666
    },
    {
      "completion_length": 234.1919755935669,
      "epoch": 0.11199128211576344,
      "grad_norm": 0.2718751293232413,
      "kl": 0.0003414750099182129,
      "learning_rate": 8.246913580246912e-08,
      "loss": 0.0,
      "reward": 1.7535714954137802,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7625000402331352,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 668
    },
    {
      "completion_length": 247.2634048461914,
      "epoch": 0.112326585355631,
      "grad_norm": 0.44564106362192857,
      "kl": 0.0003719329833984375,
      "learning_rate": 8.271604938271605e-08,
      "loss": 0.0,
      "reward": 1.7732143625617027,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7776786051690578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 670
    },
    {
      "completion_length": 233.86608123779297,
      "epoch": 0.11266188859549855,
      "grad_norm": 0.24836091057957915,
      "kl": 0.00035303831100463867,
      "learning_rate": 8.296296296296295e-08,
      "loss": 0.0,
      "reward": 1.7678571939468384,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 672
    },
    {
      "completion_length": 250.02679634094238,
      "epoch": 0.1129971918353661,
      "grad_norm": 0.33629052891667766,
      "kl": 0.0003447532653808594,
      "learning_rate": 8.320987654320988e-08,
      "loss": 0.0,
      "reward": 1.7428571805357933,
      "reward_std": 0.10101525206118822,
      "rewards/equation_reward_func": 0.7517857626080513,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 674
    },
    {
      "completion_length": 244.0669755935669,
      "epoch": 0.11333249507523366,
      "grad_norm": 0.22850472179793171,
      "kl": 0.00035583972930908203,
      "learning_rate": 8.345679012345678e-08,
      "loss": 0.0,
      "reward": 1.7214286476373672,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7214286103844643,
      "rewards/format_reward_func": 1.0,
      "step": 676
    },
    {
      "completion_length": 242.03572750091553,
      "epoch": 0.11366779831510122,
      "grad_norm": 0.23854294689657238,
      "kl": 0.00033485889434814453,
      "learning_rate": 8.37037037037037e-08,
      "loss": 0.0,
      "reward": 1.776785783469677,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.781250037252903,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 678
    },
    {
      "completion_length": 245.96429443359375,
      "epoch": 0.11400310155496877,
      "grad_norm": 0.16755888824586532,
      "kl": 0.0003756284713745117,
      "learning_rate": 8.395061728395061e-08,
      "loss": 0.0,
      "reward": 1.7571429163217545,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428883820772,
      "rewards/format_reward_func": 1.0,
      "step": 680
    },
    {
      "completion_length": 233.83929347991943,
      "epoch": 0.11433840479483633,
      "grad_norm": 0.24174291075874793,
      "kl": 0.0003483295440673828,
      "learning_rate": 8.419753086419753e-08,
      "loss": 0.0,
      "reward": 1.7196429297327995,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7241071742027998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 682
    },
    {
      "completion_length": 248.6160831451416,
      "epoch": 0.11467370803470389,
      "grad_norm": 0.22786130194654666,
      "kl": 0.00032138824462890625,
      "learning_rate": 8.444444444444444e-08,
      "loss": 0.0,
      "reward": 1.7500000670552254,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000223517418,
      "rewards/format_reward_func": 1.0,
      "step": 684
    },
    {
      "completion_length": 237.14733123779297,
      "epoch": 0.11500901127457144,
      "grad_norm": 0.20303655961874897,
      "kl": 0.00035130977630615234,
      "learning_rate": 8.469135802469136e-08,
      "loss": 0.0,
      "reward": 1.7642857804894447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857581377029,
      "rewards/format_reward_func": 1.0,
      "step": 686
    },
    {
      "completion_length": 243.2767972946167,
      "epoch": 0.115344314514439,
      "grad_norm": 0.19912653343272865,
      "kl": 0.0003395676612854004,
      "learning_rate": 8.493827160493827e-08,
      "loss": 0.0,
      "reward": 1.7232143729925156,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7366071753203869,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 688
    },
    {
      "completion_length": 249.47322463989258,
      "epoch": 0.11567961775430655,
      "grad_norm": 0.2278184337257273,
      "kl": 0.0003470182418823242,
      "learning_rate": 8.518518518518517e-08,
      "loss": 0.0,
      "reward": 1.7312500551342964,
      "reward_std": 0.06692260666750371,
      "rewards/equation_reward_func": 0.7473214492201805,
      "rewards/format_reward_func": 0.9839285835623741,
      "step": 690
    },
    {
      "completion_length": 238.8169755935669,
      "epoch": 0.11601492099417411,
      "grad_norm": 0.17911647827573768,
      "kl": 0.0003763437271118164,
      "learning_rate": 8.54320987654321e-08,
      "loss": 0.0,
      "reward": 1.7714286223053932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285850524902,
      "rewards/format_reward_func": 1.0,
      "step": 692
    },
    {
      "completion_length": 255.2366180419922,
      "epoch": 0.11635022423404166,
      "grad_norm": 0.2756423672346394,
      "kl": 0.0003293752670288086,
      "learning_rate": 8.5679012345679e-08,
      "loss": 0.0,
      "reward": 1.7410714998841286,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.7544643245637417,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 694
    },
    {
      "completion_length": 249.30804634094238,
      "epoch": 0.11668552747390921,
      "grad_norm": 0.2111679882859518,
      "kl": 0.00035190582275390625,
      "learning_rate": 8.592592592592592e-08,
      "loss": 0.0,
      "reward": 1.7196429073810577,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.7330357506871223,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 696
    },
    {
      "completion_length": 243.665189743042,
      "epoch": 0.11702083071377677,
      "grad_norm": 0.155534871523258,
      "kl": 0.00037992000579833984,
      "learning_rate": 8.617283950617283e-08,
      "loss": 0.0,
      "reward": 1.7196429446339607,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7241071686148643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 698
    },
    {
      "completion_length": 251.8348331451416,
      "epoch": 0.11735613395364433,
      "grad_norm": 0.26079539156736914,
      "kl": 0.0003502368927001953,
      "learning_rate": 8.641975308641975e-08,
      "loss": 0.0,
      "reward": 1.712500087916851,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7258928939700127,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 700
    },
    {
      "completion_length": 242.1071538925171,
      "epoch": 0.11769143719351188,
      "grad_norm": 0.2277414630854397,
      "kl": 0.0003610849380493164,
      "learning_rate": 8.666666666666666e-08,
      "loss": 0.0,
      "reward": 1.8321429193019867,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.832142885774374,
      "rewards/format_reward_func": 1.0,
      "step": 702
    },
    {
      "completion_length": 253.36608409881592,
      "epoch": 0.11802674043337943,
      "grad_norm": 0.30555546254853067,
      "kl": 0.0003439188003540039,
      "learning_rate": 8.691358024691358e-08,
      "loss": 0.0,
      "reward": 1.7464286386966705,
      "reward_std": 0.0858629634603858,
      "rewards/equation_reward_func": 0.746428607031703,
      "rewards/format_reward_func": 1.0,
      "step": 704
    },
    {
      "completion_length": 239.258939743042,
      "epoch": 0.11836204367324699,
      "grad_norm": 0.2848334485064057,
      "kl": 0.0003402233123779297,
      "learning_rate": 8.716049382716049e-08,
      "loss": 0.0,
      "reward": 1.771428644657135,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 706
    },
    {
      "completion_length": 239.32590293884277,
      "epoch": 0.11869734691311455,
      "grad_norm": 0.21435664986931335,
      "kl": 0.00037920475006103516,
      "learning_rate": 8.74074074074074e-08,
      "loss": 0.0,
      "reward": 1.7696429193019867,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7741071879863739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 708
    },
    {
      "completion_length": 243.0401906967163,
      "epoch": 0.1190326501529821,
      "grad_norm": 0.20084110899875415,
      "kl": 0.0003509521484375,
      "learning_rate": 8.765432098765432e-08,
      "loss": 0.0,
      "reward": 1.7660714909434319,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7705357521772385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 710
    },
    {
      "completion_length": 251.1339406967163,
      "epoch": 0.11936795339284965,
      "grad_norm": 0.6091294566896917,
      "kl": 0.00038111209869384766,
      "learning_rate": 8.790123456790124e-08,
      "loss": 0.0,
      "reward": 1.7428571954369545,
      "reward_std": 0.09091372694820166,
      "rewards/equation_reward_func": 0.7517857626080513,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 712
    },
    {
      "completion_length": 246.12947463989258,
      "epoch": 0.11970325663271722,
      "grad_norm": 0.24858780124660707,
      "kl": 0.0003471970558166504,
      "learning_rate": 8.814814814814814e-08,
      "loss": 0.0,
      "reward": 1.7714286595582962,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 714
    },
    {
      "completion_length": 260.6696557998657,
      "epoch": 0.12003855987258477,
      "grad_norm": 0.3235796854238328,
      "kl": 0.00035631656646728516,
      "learning_rate": 8.839506172839507e-08,
      "loss": 0.0,
      "reward": 1.680357240140438,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7026786040514708,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 716
    },
    {
      "completion_length": 244.77233219146729,
      "epoch": 0.12037386311245232,
      "grad_norm": 0.18663753240167508,
      "kl": 0.0003312826156616211,
      "learning_rate": 8.864197530864197e-08,
      "loss": 0.0,
      "reward": 1.8017857745289803,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8062500208616257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 718
    },
    {
      "completion_length": 244.47768688201904,
      "epoch": 0.12070916635231987,
      "grad_norm": 0.2831141797933221,
      "kl": 0.000345766544342041,
      "learning_rate": 8.888888888888888e-08,
      "loss": 0.0,
      "reward": 1.7343750670552254,
      "reward_std": 0.06250318652018905,
      "rewards/equation_reward_func": 0.7357143349945545,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 720
    },
    {
      "completion_length": 233.08483028411865,
      "epoch": 0.12104446959218744,
      "grad_norm": 0.2972022109726578,
      "kl": 0.0003477931022644043,
      "learning_rate": 8.91358024691358e-08,
      "loss": 0.0,
      "reward": 1.7500000596046448,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 722
    },
    {
      "completion_length": 238.62054634094238,
      "epoch": 0.12137977283205499,
      "grad_norm": 0.39101243176433587,
      "kl": 0.00035965442657470703,
      "learning_rate": 8.938271604938271e-08,
      "loss": 0.0,
      "reward": 1.7321429327130318,
      "reward_std": 0.0858629634603858,
      "rewards/equation_reward_func": 0.7321428917348385,
      "rewards/format_reward_func": 1.0,
      "step": 724
    },
    {
      "completion_length": 246.7053680419922,
      "epoch": 0.12171507607192254,
      "grad_norm": 0.19475704922237347,
      "kl": 0.0003685951232910156,
      "learning_rate": 8.962962962962963e-08,
      "loss": 0.0,
      "reward": 1.7625000402331352,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 726
    },
    {
      "completion_length": 235.30357837677002,
      "epoch": 0.1220503793117901,
      "grad_norm": 0.22615944528910542,
      "kl": 0.0003758668899536133,
      "learning_rate": 8.987654320987654e-08,
      "loss": 0.0,
      "reward": 1.735714353621006,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143312692642,
      "rewards/format_reward_func": 1.0,
      "step": 728
    },
    {
      "completion_length": 243.78572750091553,
      "epoch": 0.12238568255165766,
      "grad_norm": 0.20370458125847893,
      "kl": 0.00036966800689697266,
      "learning_rate": 9.012345679012346e-08,
      "loss": 0.0,
      "reward": 1.791071504354477,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7955357432365417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 730
    },
    {
      "completion_length": 241.9732255935669,
      "epoch": 0.12272098579152521,
      "grad_norm": 0.2427979605331861,
      "kl": 0.0003876686096191406,
      "learning_rate": 9.037037037037036e-08,
      "loss": 0.0,
      "reward": 1.8285714760422707,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8285714536905289,
      "rewards/format_reward_func": 1.0,
      "step": 732
    },
    {
      "completion_length": 242.00447273254395,
      "epoch": 0.12305628903139276,
      "grad_norm": 0.2729256204257734,
      "kl": 0.00039398670196533203,
      "learning_rate": 9.061728395061729e-08,
      "loss": 0.0,
      "reward": 1.8035714849829674,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 734
    },
    {
      "completion_length": 238.852689743042,
      "epoch": 0.12339159227126031,
      "grad_norm": 0.12436679177862824,
      "kl": 0.0003916025161743164,
      "learning_rate": 9.086419753086419e-08,
      "loss": 0.0,
      "reward": 1.7357143461704254,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143126428127,
      "rewards/format_reward_func": 1.0,
      "step": 736
    },
    {
      "completion_length": 238.6830472946167,
      "epoch": 0.12372689551112788,
      "grad_norm": 0.22044509282884264,
      "kl": 0.00040662288665771484,
      "learning_rate": 9.111111111111112e-08,
      "loss": 0.0,
      "reward": 1.7535715103149414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714730620384,
      "rewards/format_reward_func": 1.0,
      "step": 738
    },
    {
      "completion_length": 234.1473331451416,
      "epoch": 0.12406219875099543,
      "grad_norm": 0.10637798693708506,
      "kl": 0.00035452842712402344,
      "learning_rate": 9.135802469135802e-08,
      "loss": 0.0,
      "reward": 1.764285758137703,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857544124126,
      "rewards/format_reward_func": 1.0,
      "step": 740
    },
    {
      "completion_length": 247.75001049041748,
      "epoch": 0.12439750199086298,
      "grad_norm": 0.20902896319414058,
      "kl": 0.00039124488830566406,
      "learning_rate": 9.160493827160494e-08,
      "loss": 0.0,
      "reward": 1.7535714954137802,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714730620384,
      "rewards/format_reward_func": 1.0,
      "step": 742
    },
    {
      "completion_length": 247.42858028411865,
      "epoch": 0.12473280523073055,
      "grad_norm": 0.17218299822927055,
      "kl": 0.00037997961044311523,
      "learning_rate": 9.185185185185185e-08,
      "loss": 0.0,
      "reward": 1.7678572088479996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571604192257,
      "rewards/format_reward_func": 1.0,
      "step": 744
    },
    {
      "completion_length": 242.24554634094238,
      "epoch": 0.1250681084705981,
      "grad_norm": 0.3318051559590746,
      "kl": 0.0003941059112548828,
      "learning_rate": 9.209876543209875e-08,
      "loss": 0.0,
      "reward": 1.7178572043776512,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7178571745753288,
      "rewards/format_reward_func": 1.0,
      "step": 746
    },
    {
      "completion_length": 235.47322463989258,
      "epoch": 0.12540341171046565,
      "grad_norm": 0.31337923764354797,
      "kl": 0.0003949403762817383,
      "learning_rate": 9.234567901234568e-08,
      "loss": 0.0,
      "reward": 1.7839286401867867,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 748
    },
    {
      "completion_length": 243.5000114440918,
      "epoch": 0.1257387149503332,
      "grad_norm": 0.17612462070677215,
      "kl": 0.0003739595413208008,
      "learning_rate": 9.259259259259258e-08,
      "loss": 0.0,
      "reward": 1.7696429044008255,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7741071805357933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 750
    },
    {
      "completion_length": 251.64733028411865,
      "epoch": 0.12607401819020075,
      "grad_norm": 0.23943906846513874,
      "kl": 0.00038892030715942383,
      "learning_rate": 9.28395061728395e-08,
      "loss": 0.0,
      "reward": 1.74642863124609,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7464286126196384,
      "rewards/format_reward_func": 1.0,
      "step": 752
    },
    {
      "completion_length": 237.15625858306885,
      "epoch": 0.1264093214300683,
      "grad_norm": 0.2950467187688119,
      "kl": 0.00037169456481933594,
      "learning_rate": 9.308641975308641e-08,
      "loss": 0.0,
      "reward": 1.760714367032051,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.760714303702116,
      "rewards/format_reward_func": 1.0,
      "step": 754
    },
    {
      "completion_length": 251.42858409881592,
      "epoch": 0.12674462466993588,
      "grad_norm": 0.16807803968532695,
      "kl": 0.00037419795989990234,
      "learning_rate": 9.333333333333334e-08,
      "loss": 0.0,
      "reward": 1.7428572252392769,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7517857477068901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 756
    },
    {
      "completion_length": 243.0178680419922,
      "epoch": 0.12707992790980344,
      "grad_norm": 0.30999128943235993,
      "kl": 0.0003815889358520508,
      "learning_rate": 9.358024691358024e-08,
      "loss": 0.0,
      "reward": 1.8035715073347092,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 758
    },
    {
      "completion_length": 243.02233409881592,
      "epoch": 0.127415231149671,
      "grad_norm": 0.34858323722828805,
      "kl": 0.00039207935333251953,
      "learning_rate": 9.382716049382716e-08,
      "loss": 0.0,
      "reward": 1.7142857983708382,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857387661934,
      "rewards/format_reward_func": 1.0,
      "step": 760
    },
    {
      "completion_length": 241.68751049041748,
      "epoch": 0.12775053438953854,
      "grad_norm": 0.13222748191498046,
      "kl": 0.00041425228118896484,
      "learning_rate": 9.407407407407407e-08,
      "loss": 0.0,
      "reward": 1.7678572088479996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571864962578,
      "rewards/format_reward_func": 1.0,
      "step": 762
    },
    {
      "completion_length": 244.65626049041748,
      "epoch": 0.1280858376294061,
      "grad_norm": 0.30478521229008854,
      "kl": 0.0004159212112426758,
      "learning_rate": 9.432098765432098e-08,
      "loss": 0.0,
      "reward": 1.7089286670088768,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7133928947150707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 764
    },
    {
      "completion_length": 237.8125114440918,
      "epoch": 0.12842114086927364,
      "grad_norm": 0.21084985263179185,
      "kl": 0.00038814544677734375,
      "learning_rate": 9.45679012345679e-08,
      "loss": 0.0,
      "reward": 1.7928572073578835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.792857164517045,
      "rewards/format_reward_func": 1.0,
      "step": 766
    },
    {
      "completion_length": 234.95982933044434,
      "epoch": 0.1287564441091412,
      "grad_norm": 0.2433138284776026,
      "kl": 0.00036156177520751953,
      "learning_rate": 9.481481481481481e-08,
      "loss": 0.0,
      "reward": 1.7232143804430962,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7267857454717159,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 768
    },
    {
      "completion_length": 242.1875114440918,
      "epoch": 0.12909174734900877,
      "grad_norm": 0.15249408191828723,
      "kl": 0.00039446353912353516,
      "learning_rate": 9.506172839506173e-08,
      "loss": 0.0,
      "reward": 1.7392857745289803,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7392857633531094,
      "rewards/format_reward_func": 1.0,
      "step": 770
    },
    {
      "completion_length": 241.51786613464355,
      "epoch": 0.12942705058887632,
      "grad_norm": 0.19767611997197468,
      "kl": 0.0004150867462158203,
      "learning_rate": 9.530864197530864e-08,
      "loss": 0.0,
      "reward": 1.7767857611179352,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7812500260770321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 772
    },
    {
      "completion_length": 247.3259048461914,
      "epoch": 0.12976235382874388,
      "grad_norm": 0.2504453396389677,
      "kl": 0.00043642520904541016,
      "learning_rate": 9.555555555555556e-08,
      "loss": 0.0,
      "reward": 1.7517857998609543,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500350177288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 774
    },
    {
      "completion_length": 248.77679634094238,
      "epoch": 0.13009765706861143,
      "grad_norm": 0.1976896046009228,
      "kl": 0.0004209280014038086,
      "learning_rate": 9.580246913580246e-08,
      "loss": 0.0,
      "reward": 1.7642857730388641,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7732143141329288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 776
    },
    {
      "completion_length": 238.41965579986572,
      "epoch": 0.13043296030847898,
      "grad_norm": 0.3026664034458984,
      "kl": 0.0004271268844604492,
      "learning_rate": 9.604938271604938e-08,
      "loss": 0.0,
      "reward": 1.73392865806818,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7383928820490837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 778
    },
    {
      "completion_length": 242.62501049041748,
      "epoch": 0.13076826354834653,
      "grad_norm": 0.21444822804367636,
      "kl": 0.00046432018280029297,
      "learning_rate": 9.629629629629629e-08,
      "loss": 0.0,
      "reward": 1.7750000581145287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 780
    },
    {
      "completion_length": 247.27679538726807,
      "epoch": 0.13110356678821408,
      "grad_norm": 0.257440107889928,
      "kl": 0.0003908872604370117,
      "learning_rate": 9.65432098765432e-08,
      "loss": 0.0,
      "reward": 1.783928632736206,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 782
    },
    {
      "completion_length": 242.34822463989258,
      "epoch": 0.13143887002808163,
      "grad_norm": 0.2855112633919714,
      "kl": 0.0004401206970214844,
      "learning_rate": 9.679012345679012e-08,
      "loss": 0.0,
      "reward": 1.7821429297327995,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 784
    },
    {
      "completion_length": 252.64286613464355,
      "epoch": 0.1317741732679492,
      "grad_norm": 0.13781706493838075,
      "kl": 0.0004125833511352539,
      "learning_rate": 9.703703703703703e-08,
      "loss": 0.0,
      "reward": 1.687500074505806,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.6919643171131611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 786
    },
    {
      "completion_length": 241.74554538726807,
      "epoch": 0.13210947650781676,
      "grad_norm": 0.12654413814050075,
      "kl": 0.00040209293365478516,
      "learning_rate": 9.728395061728395e-08,
      "loss": 0.0,
      "reward": 1.7732143476605415,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7776786051690578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 788
    },
    {
      "completion_length": 243.47322750091553,
      "epoch": 0.13244477974768432,
      "grad_norm": 0.2837353459979307,
      "kl": 0.0004172325134277344,
      "learning_rate": 9.753086419753086e-08,
      "loss": 0.0,
      "reward": 1.728571504354477,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7285714503377676,
      "rewards/format_reward_func": 1.0,
      "step": 790
    },
    {
      "completion_length": 248.37054634094238,
      "epoch": 0.13278008298755187,
      "grad_norm": 0.22323115630506868,
      "kl": 0.00045418739318847656,
      "learning_rate": 9.777777777777778e-08,
      "loss": 0.0,
      "reward": 1.7357143834233284,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7357143126428127,
      "rewards/format_reward_func": 1.0,
      "step": 792
    },
    {
      "completion_length": 249.6071548461914,
      "epoch": 0.13311538622741942,
      "grad_norm": 0.24346910568279878,
      "kl": 0.000400543212890625,
      "learning_rate": 9.802469135802469e-08,
      "loss": 0.0,
      "reward": 1.7803572118282318,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 794
    },
    {
      "completion_length": 255.96875953674316,
      "epoch": 0.13345068946728697,
      "grad_norm": 0.3170654293138634,
      "kl": 0.00042128562927246094,
      "learning_rate": 9.82716049382716e-08,
      "loss": 0.0,
      "reward": 1.7089286372065544,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7133928816765547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 796
    },
    {
      "completion_length": 242.2053680419922,
      "epoch": 0.13378599270715452,
      "grad_norm": 0.2949914528878829,
      "kl": 0.0004379749298095703,
      "learning_rate": 9.851851851851852e-08,
      "loss": 0.0,
      "reward": 1.7357143685221672,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.735714316368103,
      "rewards/format_reward_func": 1.0,
      "step": 798
    },
    {
      "completion_length": 250.4375123977661,
      "epoch": 0.1341212959470221,
      "grad_norm": 0.21137036767136858,
      "kl": 0.00042176246643066406,
      "learning_rate": 9.876543209876542e-08,
      "loss": 0.0,
      "reward": 1.7678572162985802,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7678571734577417,
      "rewards/format_reward_func": 1.0,
      "step": 800
    },
    {
      "completion_length": 238.6919755935669,
      "epoch": 0.13445659918688965,
      "grad_norm": 0.2814166147627615,
      "kl": 0.00045418739318847656,
      "learning_rate": 9.901234567901235e-08,
      "loss": 0.0,
      "reward": 1.7857143580913544,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 802
    },
    {
      "completion_length": 242.8705472946167,
      "epoch": 0.1347919024267572,
      "grad_norm": 0.30188794758016163,
      "kl": 0.00044226646423339844,
      "learning_rate": 9.925925925925925e-08,
      "loss": 0.0,
      "reward": 1.7366072237491608,
      "reward_std": 0.05934646027162671,
      "rewards/equation_reward_func": 0.7383928894996643,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 804
    },
    {
      "completion_length": 257.3794775009155,
      "epoch": 0.13512720566662476,
      "grad_norm": 0.25469855228198224,
      "kl": 0.0004140138626098633,
      "learning_rate": 9.950617283950617e-08,
      "loss": 0.0,
      "reward": 1.7446429505944252,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 806
    },
    {
      "completion_length": 250.01787090301514,
      "epoch": 0.1354625089064923,
      "grad_norm": 0.27453581331030075,
      "kl": 0.0004793405532836914,
      "learning_rate": 9.975308641975308e-08,
      "loss": 0.0,
      "reward": 1.767857201397419,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7678571827709675,
      "rewards/format_reward_func": 1.0,
      "step": 808
    },
    {
      "completion_length": 244.21876049041748,
      "epoch": 0.13579781214635986,
      "grad_norm": 0.28017698150363773,
      "kl": 0.0004837512969970703,
      "learning_rate": 1e-07,
      "loss": 0.0,
      "reward": 1.7232143878936768,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7276785895228386,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 810
    },
    {
      "completion_length": 247.321439743042,
      "epoch": 0.1361331153862274,
      "grad_norm": 0.29506744159707904,
      "kl": 0.000514984130859375,
      "learning_rate": 1.0024691358024691e-07,
      "loss": 0.0,
      "reward": 1.737500086426735,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7419643122702837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 812
    },
    {
      "completion_length": 248.3794765472412,
      "epoch": 0.13646841862609496,
      "grad_norm": 0.45052300905816106,
      "kl": 0.0004858970642089844,
      "learning_rate": 1.0049382716049381e-07,
      "loss": 0.0,
      "reward": 1.807142935693264,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.8071428798139095,
      "rewards/format_reward_func": 1.0,
      "step": 814
    },
    {
      "completion_length": 247.02679443359375,
      "epoch": 0.13680372186596254,
      "grad_norm": 0.23890198941501573,
      "kl": 0.0005213022232055664,
      "learning_rate": 1.0074074074074074e-07,
      "loss": 0.0,
      "reward": 1.7535715028643608,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 816
    },
    {
      "completion_length": 255.1428680419922,
      "epoch": 0.1371390251058301,
      "grad_norm": 0.3321438645772538,
      "kl": 0.000493168830871582,
      "learning_rate": 1.0098765432098764e-07,
      "loss": 0.0,
      "reward": 1.7232143431901932,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7276786118745804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 818
    },
    {
      "completion_length": 244.43304824829102,
      "epoch": 0.13747432834569764,
      "grad_norm": 0.20810648283029934,
      "kl": 0.0005180835723876953,
      "learning_rate": 1.0123456790123457e-07,
      "loss": 0.0,
      "reward": 1.7392857894301414,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.739285746589303,
      "rewards/format_reward_func": 1.0,
      "step": 820
    },
    {
      "completion_length": 237.0134038925171,
      "epoch": 0.1378096315855652,
      "grad_norm": 0.22541616829027122,
      "kl": 0.0004540681838989258,
      "learning_rate": 1.0148148148148147e-07,
      "loss": 0.0,
      "reward": 1.8535714894533157,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.8625000156462193,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 822
    },
    {
      "completion_length": 242.77233219146729,
      "epoch": 0.13814493482543275,
      "grad_norm": 0.27520042618499535,
      "kl": 0.00048482418060302734,
      "learning_rate": 1.017283950617284e-07,
      "loss": 0.0,
      "reward": 1.789285771548748,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 824
    },
    {
      "completion_length": 248.8660831451416,
      "epoch": 0.1384802380653003,
      "grad_norm": 0.30717772584495134,
      "kl": 0.0005295276641845703,
      "learning_rate": 1.019753086419753e-07,
      "loss": 0.0,
      "reward": 1.7571429088711739,
      "reward_std": 0.06060914974659681,
      "rewards/equation_reward_func": 0.7571428753435612,
      "rewards/format_reward_func": 1.0,
      "step": 826
    },
    {
      "completion_length": 241.49554824829102,
      "epoch": 0.13881554130516785,
      "grad_norm": 0.16709679133115038,
      "kl": 0.0005042552947998047,
      "learning_rate": 1.0222222222222223e-07,
      "loss": 0.0,
      "reward": 1.7964286282658577,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 828
    },
    {
      "completion_length": 258.34376335144043,
      "epoch": 0.13915084454503543,
      "grad_norm": 0.19247229448577885,
      "kl": 0.0004944801330566406,
      "learning_rate": 1.0246913580246913e-07,
      "loss": 0.0,
      "reward": 1.755357213318348,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7598214484751225,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 830
    },
    {
      "completion_length": 236.6651906967163,
      "epoch": 0.13948614778490298,
      "grad_norm": 0.2414068343995031,
      "kl": 0.00048828125,
      "learning_rate": 1.0271604938271603e-07,
      "loss": 0.0,
      "reward": 1.7785714864730835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 832
    },
    {
      "completion_length": 245.133939743042,
      "epoch": 0.13982145102477053,
      "grad_norm": 0.328663766531583,
      "kl": 0.0005055665969848633,
      "learning_rate": 1.0296296296296296e-07,
      "loss": 0.0,
      "reward": 1.689285784959793,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.6892857402563095,
      "rewards/format_reward_func": 1.0,
      "step": 834
    },
    {
      "completion_length": 243.54018878936768,
      "epoch": 0.14015675426463808,
      "grad_norm": 0.4397604670140151,
      "kl": 0.00045621395111083984,
      "learning_rate": 1.0320987654320986e-07,
      "loss": 0.0,
      "reward": 1.7482143491506577,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7526785954833031,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 836
    },
    {
      "completion_length": 241.74108219146729,
      "epoch": 0.14049205750450564,
      "grad_norm": 0.29775824487765673,
      "kl": 0.00047719478607177734,
      "learning_rate": 1.0345679012345679e-07,
      "loss": 0.0,
      "reward": 1.7517857626080513,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7562500387430191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 838
    },
    {
      "completion_length": 239.21429538726807,
      "epoch": 0.1408273607443732,
      "grad_norm": 0.3551956976091781,
      "kl": 0.0004627704620361328,
      "learning_rate": 1.0370370370370369e-07,
      "loss": 0.0,
      "reward": 1.757589340209961,
      "reward_std": 0.09028238197788596,
      "rewards/equation_reward_func": 0.7633928917348385,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 840
    },
    {
      "completion_length": 249.44197368621826,
      "epoch": 0.14116266398424074,
      "grad_norm": 0.16079556819889082,
      "kl": 0.0005052089691162109,
      "learning_rate": 1.0395061728395062e-07,
      "loss": 0.0,
      "reward": 1.701785795390606,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7062500417232513,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 842
    },
    {
      "completion_length": 249.0178689956665,
      "epoch": 0.1414979672241083,
      "grad_norm": 0.43008578708661116,
      "kl": 0.0005563497543334961,
      "learning_rate": 1.0419753086419752e-07,
      "loss": 0.0,
      "reward": 1.7410714998841286,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7455357611179352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 844
    },
    {
      "completion_length": 240.696439743042,
      "epoch": 0.14183327046397587,
      "grad_norm": 0.19942106819097674,
      "kl": 0.0004343986511230469,
      "learning_rate": 1.0444444444444445e-07,
      "loss": 0.0,
      "reward": 1.7803571745753288,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 846
    },
    {
      "completion_length": 248.0937614440918,
      "epoch": 0.14216857370384342,
      "grad_norm": 0.32938655852109006,
      "kl": 0.0005087852478027344,
      "learning_rate": 1.0469135802469135e-07,
      "loss": 0.0,
      "reward": 1.7625000700354576,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7669643107801676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 848
    },
    {
      "completion_length": 241.70983123779297,
      "epoch": 0.14250387694371097,
      "grad_norm": 0.20099853956894784,
      "kl": 0.0004711151123046875,
      "learning_rate": 1.0493827160493827e-07,
      "loss": 0.0,
      "reward": 1.7571429014205933,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7571428921073675,
      "rewards/format_reward_func": 1.0,
      "step": 850
    },
    {
      "completion_length": 251.04911613464355,
      "epoch": 0.14283918018357852,
      "grad_norm": 0.26710235078942274,
      "kl": 0.0005549192428588867,
      "learning_rate": 1.0518518518518518e-07,
      "loss": 0.0,
      "reward": 1.7535715028643608,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 852
    },
    {
      "completion_length": 242.21875953674316,
      "epoch": 0.14317448342344608,
      "grad_norm": 0.24595597743321643,
      "kl": 0.000561833381652832,
      "learning_rate": 1.054320987654321e-07,
      "loss": 0.0,
      "reward": 1.7821429073810577,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7821428962051868,
      "rewards/format_reward_func": 1.0,
      "step": 854
    },
    {
      "completion_length": 239.6160831451416,
      "epoch": 0.14350978666331363,
      "grad_norm": 0.24191581359797343,
      "kl": 0.0005096197128295898,
      "learning_rate": 1.0567901234567901e-07,
      "loss": 0.0,
      "reward": 1.7125000730156898,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.716964315623045,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 856
    },
    {
      "completion_length": 242.3705472946167,
      "epoch": 0.14384508990318118,
      "grad_norm": 0.3070888032318429,
      "kl": 0.0004938840866088867,
      "learning_rate": 1.0592592592592592e-07,
      "loss": 0.0,
      "reward": 1.7250000685453415,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000275671482,
      "rewards/format_reward_func": 1.0,
      "step": 858
    },
    {
      "completion_length": 250.14733123779297,
      "epoch": 0.14418039314304876,
      "grad_norm": 0.2893268587254202,
      "kl": 0.0005345344543457031,
      "learning_rate": 1.0617283950617284e-07,
      "loss": 0.0,
      "reward": 1.7589286640286446,
      "reward_std": 0.09848987031728029,
      "rewards/equation_reward_func": 0.7633928768336773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 860
    },
    {
      "completion_length": 244.47322463989258,
      "epoch": 0.1445156963829163,
      "grad_norm": 0.19801472144759966,
      "kl": 0.0005357265472412109,
      "learning_rate": 1.0641975308641974e-07,
      "loss": 0.0,
      "reward": 1.7178572490811348,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7178571783006191,
      "rewards/format_reward_func": 1.0,
      "step": 862
    },
    {
      "completion_length": 241.60715293884277,
      "epoch": 0.14485099962278386,
      "grad_norm": 0.2844514690303181,
      "kl": 0.0005900859832763672,
      "learning_rate": 1.0666666666666667e-07,
      "loss": 0.0,
      "reward": 1.8107143342494965,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.8107143081724644,
      "rewards/format_reward_func": 1.0,
      "step": 864
    },
    {
      "completion_length": 246.8035831451416,
      "epoch": 0.1451863028626514,
      "grad_norm": 0.14321224353680495,
      "kl": 0.0005359649658203125,
      "learning_rate": 1.0691358024691357e-07,
      "loss": 0.0,
      "reward": 1.6946429386734962,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.6991071663796902,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 866
    },
    {
      "completion_length": 252.5446538925171,
      "epoch": 0.14552160610251896,
      "grad_norm": 0.2782425391415311,
      "kl": 0.00047397613525390625,
      "learning_rate": 1.0716049382716049e-07,
      "loss": 0.0,
      "reward": 1.778571479022503,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7875000424683094,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 868
    },
    {
      "completion_length": 240.04465293884277,
      "epoch": 0.14585690934238651,
      "grad_norm": 0.41490336823387663,
      "kl": 0.0005705356597900391,
      "learning_rate": 1.074074074074074e-07,
      "loss": 0.0,
      "reward": 1.7500000894069672,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7500000279396772,
      "rewards/format_reward_func": 1.0,
      "step": 870
    },
    {
      "completion_length": 249.63840675354004,
      "epoch": 0.14619221258225407,
      "grad_norm": 0.2734543943620819,
      "kl": 0.0005452632904052734,
      "learning_rate": 1.0765432098765431e-07,
      "loss": 0.0,
      "reward": 1.7714286372065544,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.771428594365716,
      "rewards/format_reward_func": 1.0,
      "step": 872
    },
    {
      "completion_length": 239.69197273254395,
      "epoch": 0.14652751582212162,
      "grad_norm": 0.2274695812309938,
      "kl": 0.000532984733581543,
      "learning_rate": 1.0790123456790123e-07,
      "loss": 0.0,
      "reward": 1.7750000581145287,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 874
    },
    {
      "completion_length": 238.9330472946167,
      "epoch": 0.1468628190619892,
      "grad_norm": 0.16239306753807847,
      "kl": 0.0005614757537841797,
      "learning_rate": 1.0814814814814814e-07,
      "loss": 0.0,
      "reward": 1.760714367032051,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.769642885774374,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 876
    },
    {
      "completion_length": 246.38393783569336,
      "epoch": 0.14719812230185675,
      "grad_norm": 0.25783637098150386,
      "kl": 0.0005737543106079102,
      "learning_rate": 1.0839506172839506e-07,
      "loss": 0.0,
      "reward": 1.7678572088479996,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7767857387661934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 878
    },
    {
      "completion_length": 241.51786708831787,
      "epoch": 0.1475334255417243,
      "grad_norm": 0.3677828611432142,
      "kl": 0.0005735158920288086,
      "learning_rate": 1.0864197530864197e-07,
      "loss": 0.0,
      "reward": 1.7125000730156898,
      "reward_std": 0.09343910776078701,
      "rewards/equation_reward_func": 0.7169643249362707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 880
    },
    {
      "completion_length": 242.76340675354004,
      "epoch": 0.14786872878159185,
      "grad_norm": 0.20185872070200894,
      "kl": 0.0005519390106201172,
      "learning_rate": 1.0888888888888888e-07,
      "loss": 0.0,
      "reward": 1.7696429267525673,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 882
    },
    {
      "completion_length": 239.31697368621826,
      "epoch": 0.1482040320214594,
      "grad_norm": 0.24556548501686268,
      "kl": 0.0005544424057006836,
      "learning_rate": 1.091358024691358e-07,
      "loss": 0.0,
      "reward": 1.7892857640981674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857510596514,
      "rewards/format_reward_func": 1.0,
      "step": 884
    },
    {
      "completion_length": 236.79465198516846,
      "epoch": 0.14853933526132695,
      "grad_norm": 0.19666918858541813,
      "kl": 0.0005842447280883789,
      "learning_rate": 1.093827160493827e-07,
      "loss": 0.0,
      "reward": 1.7142857909202576,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857201397419,
      "rewards/format_reward_func": 1.0,
      "step": 886
    },
    {
      "completion_length": 243.69644165039062,
      "epoch": 0.1488746385011945,
      "grad_norm": 0.2543890898105721,
      "kl": 0.0005837678909301758,
      "learning_rate": 1.0962962962962963e-07,
      "loss": 0.0,
      "reward": 1.7946429252624512,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.799107164144516,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 888
    },
    {
      "completion_length": 240.16518783569336,
      "epoch": 0.14920994174106209,
      "grad_norm": 0.2612662392595029,
      "kl": 0.000578761100769043,
      "learning_rate": 1.0987654320987653e-07,
      "loss": 0.0,
      "reward": 1.7535714879631996,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.753571443259716,
      "rewards/format_reward_func": 1.0,
      "step": 890
    },
    {
      "completion_length": 243.67412090301514,
      "epoch": 0.14954524498092964,
      "grad_norm": 0.2645548665031763,
      "kl": 0.0005819797515869141,
      "learning_rate": 1.1012345679012345e-07,
      "loss": 0.0,
      "reward": 1.760714367032051,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7696428671479225,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 892
    },
    {
      "completion_length": 240.39733123779297,
      "epoch": 0.1498805482207972,
      "grad_norm": 0.202545137842961,
      "kl": 0.0006400346755981445,
      "learning_rate": 1.1037037037037036e-07,
      "loss": 0.0,
      "reward": 1.7142857760190964,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857573926449,
      "rewards/format_reward_func": 1.0,
      "step": 894
    },
    {
      "completion_length": 244.81250953674316,
      "epoch": 0.15021585146066474,
      "grad_norm": 0.26342116615886524,
      "kl": 0.0005612373352050781,
      "learning_rate": 1.1061728395061728e-07,
      "loss": 0.0,
      "reward": 1.7250000834465027,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000387430191,
      "rewards/format_reward_func": 1.0,
      "step": 896
    },
    {
      "completion_length": 252.0714406967163,
      "epoch": 0.1505511547005323,
      "grad_norm": 0.19740268449770748,
      "kl": 0.0004432201385498047,
      "learning_rate": 1.108641975308642e-07,
      "loss": 0.0,
      "reward": 1.7375000715255737,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.741964302957058,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 898
    },
    {
      "completion_length": 244.9062623977661,
      "epoch": 0.15088645794039984,
      "grad_norm": 0.21749415281902099,
      "kl": 0.0005629062652587891,
      "learning_rate": 1.111111111111111e-07,
      "loss": 0.0,
      "reward": 1.755357213318348,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214708268642,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 900
    },
    {
      "completion_length": 238.94643783569336,
      "epoch": 0.1512217611802674,
      "grad_norm": 0.19636977379194281,
      "kl": 0.0005797147750854492,
      "learning_rate": 1.1135802469135802e-07,
      "loss": 0.0,
      "reward": 1.7910714745521545,
      "reward_std": 0.022728432901203632,
      "rewards/equation_reward_func": 0.7955357357859612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 902
    },
    {
      "completion_length": 237.40179634094238,
      "epoch": 0.15155706442013495,
      "grad_norm": 0.0812537734176348,
      "kl": 0.0005922317504882812,
      "learning_rate": 1.1160493827160493e-07,
      "loss": 0.0,
      "reward": 1.7500000819563866,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000111758709,
      "rewards/format_reward_func": 1.0,
      "step": 904
    },
    {
      "completion_length": 240.58482837677002,
      "epoch": 0.15189236766000253,
      "grad_norm": 0.31431448086070735,
      "kl": 0.0008060932159423828,
      "learning_rate": 1.1185185185185185e-07,
      "loss": 0.0,
      "reward": 1.7642857730388641,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7732143197208643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 906
    },
    {
      "completion_length": 234.2857265472412,
      "epoch": 0.15222767089987008,
      "grad_norm": 0.13392711374206998,
      "kl": 0.0005654096603393555,
      "learning_rate": 1.1209876543209876e-07,
      "loss": 0.0,
      "reward": 1.7714286297559738,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 908
    },
    {
      "completion_length": 244.5267972946167,
      "epoch": 0.15256297413973763,
      "grad_norm": 0.27221219675506475,
      "kl": 0.0005095005035400391,
      "learning_rate": 1.1234567901234568e-07,
      "loss": 0.0,
      "reward": 1.7267857789993286,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7312500383704901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 910
    },
    {
      "completion_length": 241.03126049041748,
      "epoch": 0.15289827737960518,
      "grad_norm": 0.31530224582644223,
      "kl": 0.0005794763565063477,
      "learning_rate": 1.1259259259259258e-07,
      "loss": 0.0,
      "reward": 1.7250000834465027,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000238418579,
      "rewards/format_reward_func": 1.0,
      "step": 912
    },
    {
      "completion_length": 246.7991180419922,
      "epoch": 0.15323358061947273,
      "grad_norm": 0.250379166539302,
      "kl": 0.0006912946701049805,
      "learning_rate": 1.1283950617283951e-07,
      "loss": 0.0,
      "reward": 1.7142858058214188,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7142857499420643,
      "rewards/format_reward_func": 1.0,
      "step": 914
    },
    {
      "completion_length": 238.97322463989258,
      "epoch": 0.15356888385934028,
      "grad_norm": 0.19226153813655147,
      "kl": 0.0006650686264038086,
      "learning_rate": 1.1308641975308641e-07,
      "loss": 0.0,
      "reward": 1.7267857640981674,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.740178594365716,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 916
    },
    {
      "completion_length": 236.57143688201904,
      "epoch": 0.15390418709920783,
      "grad_norm": 0.29729296307514946,
      "kl": 0.00064849853515625,
      "learning_rate": 1.1333333333333332e-07,
      "loss": 0.0,
      "reward": 1.7892857566475868,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7892857491970062,
      "rewards/format_reward_func": 1.0,
      "step": 918
    },
    {
      "completion_length": 241.696439743042,
      "epoch": 0.1542394903390754,
      "grad_norm": 0.23932282850941852,
      "kl": 0.0005295276641845703,
      "learning_rate": 1.1358024691358024e-07,
      "loss": 0.0,
      "reward": 1.7750000655651093,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 920
    },
    {
      "completion_length": 240.95090293884277,
      "epoch": 0.15457479357894297,
      "grad_norm": 0.1940264202902731,
      "kl": 0.0005745887756347656,
      "learning_rate": 1.1382716049382715e-07,
      "loss": 0.0,
      "reward": 1.769642911851406,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7741071619093418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 922
    },
    {
      "completion_length": 243.47322273254395,
      "epoch": 0.15491009681881052,
      "grad_norm": 0.21505952196582073,
      "kl": 0.0005426406860351562,
      "learning_rate": 1.1407407407407407e-07,
      "loss": 0.0,
      "reward": 1.7607143446803093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.760714303702116,
      "rewards/format_reward_func": 1.0,
      "step": 924
    },
    {
      "completion_length": 239.3616180419922,
      "epoch": 0.15524540005867807,
      "grad_norm": 0.2723087004689939,
      "kl": 0.0005807876586914062,
      "learning_rate": 1.1432098765432098e-07,
      "loss": 0.0,
      "reward": 1.8107143342494965,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8196429014205933,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 926
    },
    {
      "completion_length": 248.90179824829102,
      "epoch": 0.15558070329854562,
      "grad_norm": 0.4941249855932243,
      "kl": 0.0005967617034912109,
      "learning_rate": 1.145679012345679e-07,
      "loss": 0.0,
      "reward": 1.7767857536673546,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7812500186264515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 928
    },
    {
      "completion_length": 241.01340293884277,
      "epoch": 0.15591600653841317,
      "grad_norm": 0.1552928542536375,
      "kl": 0.0006102323532104492,
      "learning_rate": 1.148148148148148e-07,
      "loss": 0.0,
      "reward": 1.7500000894069672,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000223517418,
      "rewards/format_reward_func": 1.0,
      "step": 930
    },
    {
      "completion_length": 238.94197273254395,
      "epoch": 0.15625130977828072,
      "grad_norm": 0.418642637769558,
      "kl": 0.0006206035614013672,
      "learning_rate": 1.1506172839506173e-07,
      "loss": 0.0,
      "reward": 1.7428572177886963,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7428571749478579,
      "rewards/format_reward_func": 1.0,
      "step": 932
    },
    {
      "completion_length": 243.3437614440918,
      "epoch": 0.15658661301814827,
      "grad_norm": 0.223657964390005,
      "kl": 0.0006108283996582031,
      "learning_rate": 1.1530864197530863e-07,
      "loss": 0.0,
      "reward": 1.723214402794838,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7276786025613546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 934
    },
    {
      "completion_length": 238.477689743042,
      "epoch": 0.15692191625801585,
      "grad_norm": 0.19391093772147885,
      "kl": 0.0006520748138427734,
      "learning_rate": 1.1555555555555555e-07,
      "loss": 0.0,
      "reward": 1.7750000804662704,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 936
    },
    {
      "completion_length": 243.4955472946167,
      "epoch": 0.1572572194978834,
      "grad_norm": 0.29975605622207685,
      "kl": 0.000649571418762207,
      "learning_rate": 1.1580246913580246e-07,
      "loss": 0.0,
      "reward": 1.7625000849366188,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7669643014669418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 938
    },
    {
      "completion_length": 240.0625114440918,
      "epoch": 0.15759252273775096,
      "grad_norm": 0.1899270930718893,
      "kl": 0.0006060600280761719,
      "learning_rate": 1.1604938271604938e-07,
      "loss": 0.0,
      "reward": 1.7928571924567223,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571626543999,
      "rewards/format_reward_func": 1.0,
      "step": 940
    },
    {
      "completion_length": 246.10715293884277,
      "epoch": 0.1579278259776185,
      "grad_norm": 0.3378569868261778,
      "kl": 0.0005865097045898438,
      "learning_rate": 1.162962962962963e-07,
      "loss": 0.0,
      "reward": 1.7303572073578835,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7348214574158192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 942
    },
    {
      "completion_length": 238.98661613464355,
      "epoch": 0.15826312921748606,
      "grad_norm": 0.23510469699532882,
      "kl": 0.0005903244018554688,
      "learning_rate": 1.1654320987654321e-07,
      "loss": 0.0,
      "reward": 1.8107143267989159,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 944
    },
    {
      "completion_length": 239.17858219146729,
      "epoch": 0.1585984324573536,
      "grad_norm": 0.2531029628465875,
      "kl": 0.0005866289138793945,
      "learning_rate": 1.1679012345679012e-07,
      "loss": 0.0,
      "reward": 1.7553571909666061,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7598214522004128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 946
    },
    {
      "completion_length": 239.99108219146729,
      "epoch": 0.15893373569722116,
      "grad_norm": 0.1832711381666499,
      "kl": 0.0005750656127929688,
      "learning_rate": 1.1703703703703702e-07,
      "loss": 0.0,
      "reward": 1.8303571790456772,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.834821455180645,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 948
    },
    {
      "completion_length": 248.0223331451416,
      "epoch": 0.15926903893708874,
      "grad_norm": 0.21616249364705065,
      "kl": 0.0006320476531982422,
      "learning_rate": 1.1728395061728394e-07,
      "loss": 0.0,
      "reward": 1.7892857939004898,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 950
    },
    {
      "completion_length": 247.7053689956665,
      "epoch": 0.1596043421769563,
      "grad_norm": 0.2808309498061027,
      "kl": 0.0006732940673828125,
      "learning_rate": 1.1753086419753085e-07,
      "loss": 0.0,
      "reward": 1.732142947614193,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321428656578064,
      "rewards/format_reward_func": 1.0,
      "step": 952
    },
    {
      "completion_length": 247.93304634094238,
      "epoch": 0.15993964541682384,
      "grad_norm": 0.28244344156282136,
      "kl": 0.0006171464920043945,
      "learning_rate": 1.1777777777777777e-07,
      "loss": 0.0,
      "reward": 1.7678571939468384,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.776785746216774,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 954
    },
    {
      "completion_length": 249.13840293884277,
      "epoch": 0.1602749486566914,
      "grad_norm": 0.3471574848884779,
      "kl": 0.0006940364837646484,
      "learning_rate": 1.1802469135802468e-07,
      "loss": 0.0,
      "reward": 1.7535714879631996,
      "reward_std": 0.0858629634603858,
      "rewards/equation_reward_func": 0.7535714730620384,
      "rewards/format_reward_func": 1.0,
      "step": 956
    },
    {
      "completion_length": 237.1696548461914,
      "epoch": 0.16061025189655895,
      "grad_norm": 0.24213326359003345,
      "kl": 0.0006643533706665039,
      "learning_rate": 1.182716049382716e-07,
      "loss": 0.0,
      "reward": 1.7928571924567223,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 958
    },
    {
      "completion_length": 246.29019260406494,
      "epoch": 0.1609455551364265,
      "grad_norm": 0.25180774730216865,
      "kl": 0.0006979703903198242,
      "learning_rate": 1.1851851851851851e-07,
      "loss": 0.0,
      "reward": 1.7196429297327995,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.724107176065445,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 960
    },
    {
      "completion_length": 240.30804824829102,
      "epoch": 0.16128085837629405,
      "grad_norm": 0.27009721641478057,
      "kl": 0.0005599260330200195,
      "learning_rate": 1.1876543209876543e-07,
      "loss": 0.0,
      "reward": 1.7857143580913544,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 962
    },
    {
      "completion_length": 240.95090675354004,
      "epoch": 0.16161616161616163,
      "grad_norm": 0.20274197501049643,
      "kl": 0.0005586147308349609,
      "learning_rate": 1.1901234567901234e-07,
      "loss": 0.0,
      "reward": 1.7035715207457542,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.712500024586916,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 964
    },
    {
      "completion_length": 234.34376049041748,
      "epoch": 0.16195146485602918,
      "grad_norm": 0.2154479276321837,
      "kl": 0.0006890296936035156,
      "learning_rate": 1.1925925925925924e-07,
      "loss": 0.0,
      "reward": 1.7821429148316383,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7910714615136385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 966
    },
    {
      "completion_length": 246.4776906967163,
      "epoch": 0.16228676809589673,
      "grad_norm": 0.31223951314737436,
      "kl": 0.000695347785949707,
      "learning_rate": 1.1950617283950616e-07,
      "loss": 0.0,
      "reward": 1.74821437895298,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.752678606659174,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 968
    },
    {
      "completion_length": 247.47322463989258,
      "epoch": 0.16262207133576428,
      "grad_norm": 0.26589705178869344,
      "kl": 0.0006476640701293945,
      "learning_rate": 1.1975308641975307e-07,
      "loss": 0.0,
      "reward": 1.7379465252161026,
      "reward_std": 0.06755394977517426,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 970
    },
    {
      "completion_length": 243.9241180419922,
      "epoch": 0.16295737457563184,
      "grad_norm": 0.2610188794854556,
      "kl": 0.0007169246673583984,
      "learning_rate": 1.2e-07,
      "loss": 0.0,
      "reward": 1.733928695321083,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7383928894996643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 972
    },
    {
      "completion_length": 244.821439743042,
      "epoch": 0.1632926778154994,
      "grad_norm": 0.3007014229187292,
      "kl": 0.0007772445678710938,
      "learning_rate": 1.202469135802469e-07,
      "loss": 0.0,
      "reward": 1.7875000685453415,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7919643111526966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 974
    },
    {
      "completion_length": 244.1964406967163,
      "epoch": 0.16362798105536694,
      "grad_norm": 0.31254137964797346,
      "kl": 0.0007396936416625977,
      "learning_rate": 1.2049382716049382e-07,
      "loss": 0.0,
      "reward": 1.764285795390606,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7732143178582191,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 976
    },
    {
      "completion_length": 245.65180015563965,
      "epoch": 0.1639632842952345,
      "grad_norm": 0.12124514568359325,
      "kl": 0.0006595849990844727,
      "learning_rate": 1.2074074074074073e-07,
      "loss": 0.0,
      "reward": 1.7089286372065544,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.713392898440361,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 978
    },
    {
      "completion_length": 231.5982255935669,
      "epoch": 0.16429858753510207,
      "grad_norm": 0.30993064357758865,
      "kl": 0.0006895065307617188,
      "learning_rate": 1.2098765432098765e-07,
      "loss": 0.0,
      "reward": 1.757142923772335,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7651786059141159,
      "rewards/format_reward_func": 0.9919642880558968,
      "step": 980
    },
    {
      "completion_length": 240.02679538726807,
      "epoch": 0.16463389077496962,
      "grad_norm": 0.2581272471044061,
      "kl": 0.0006649494171142578,
      "learning_rate": 1.2123456790123456e-07,
      "loss": 0.0,
      "reward": 1.7482143566012383,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7526785992085934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 982
    },
    {
      "completion_length": 241.2678689956665,
      "epoch": 0.16496919401483717,
      "grad_norm": 0.36361643382301556,
      "kl": 0.00080108642578125,
      "learning_rate": 1.2148148148148148e-07,
      "loss": 0.0,
      "reward": 1.7607143446803093,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 984
    },
    {
      "completion_length": 250.83483505249023,
      "epoch": 0.16530449725470472,
      "grad_norm": 0.21378059263062124,
      "kl": 0.0007206201553344727,
      "learning_rate": 1.217283950617284e-07,
      "loss": 0.0,
      "reward": 1.7017857730388641,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7062500435858965,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 986
    },
    {
      "completion_length": 238.31697463989258,
      "epoch": 0.16563980049457228,
      "grad_norm": 0.32740835817943337,
      "kl": 0.0007141828536987305,
      "learning_rate": 1.219753086419753e-07,
      "loss": 0.0,
      "reward": 1.6821429505944252,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.6910714693367481,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 988
    },
    {
      "completion_length": 238.20983219146729,
      "epoch": 0.16597510373443983,
      "grad_norm": 0.19363083263718955,
      "kl": 0.0007278919219970703,
      "learning_rate": 1.2222222222222222e-07,
      "loss": 0.0,
      "reward": 1.7000000700354576,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7000000439584255,
      "rewards/format_reward_func": 1.0,
      "step": 990
    },
    {
      "completion_length": 234.18751049041748,
      "epoch": 0.16631040697430738,
      "grad_norm": 0.23389482243909598,
      "kl": 0.0006586313247680664,
      "learning_rate": 1.2246913580246914e-07,
      "loss": 0.0,
      "reward": 1.7593750730156898,
      "reward_std": 0.047350899782031775,
      "rewards/equation_reward_func": 0.7696428969502449,
      "rewards/format_reward_func": 0.9897321499884129,
      "step": 992
    },
    {
      "completion_length": 239.7009038925171,
      "epoch": 0.16664571021417496,
      "grad_norm": 0.2894204366418183,
      "kl": 0.0007368326187133789,
      "learning_rate": 1.2271604938271605e-07,
      "loss": 0.0,
      "reward": 1.7821429297327995,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7910714708268642,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 994
    },
    {
      "completion_length": 232.94643878936768,
      "epoch": 0.1669810134540425,
      "grad_norm": 0.24697120320275165,
      "kl": 0.0007069110870361328,
      "learning_rate": 1.2296296296296297e-07,
      "loss": 0.0,
      "reward": 1.7750000655651093,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 996
    },
    {
      "completion_length": 240.09376335144043,
      "epoch": 0.16731631669391006,
      "grad_norm": 0.26157424089241804,
      "kl": 0.0006704330444335938,
      "learning_rate": 1.2320987654320988e-07,
      "loss": 0.0,
      "reward": 1.764285795390606,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.764285733923316,
      "rewards/format_reward_func": 1.0,
      "step": 998
    },
    {
      "completion_length": 253.48215198516846,
      "epoch": 0.1676516199337776,
      "grad_norm": 0.2163902996279677,
      "kl": 0.0006573200225830078,
      "learning_rate": 1.2345679012345677e-07,
      "loss": 0.0,
      "reward": 1.7928572073578835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 1000
    },
    {
      "completion_length": 248.79018878936768,
      "epoch": 0.16798692317364516,
      "grad_norm": 0.12645581691936147,
      "kl": 0.0006426572799682617,
      "learning_rate": 1.237037037037037e-07,
      "loss": 0.0,
      "reward": 1.7178572341799736,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.71785718947649,
      "rewards/format_reward_func": 1.0,
      "step": 1002
    },
    {
      "completion_length": 229.32143878936768,
      "epoch": 0.16832222641351272,
      "grad_norm": 0.15066270893991937,
      "kl": 0.0006672143936157227,
      "learning_rate": 1.239506172839506e-07,
      "loss": 0.0,
      "reward": 1.7750000730156898,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 1004
    },
    {
      "completion_length": 243.65179538726807,
      "epoch": 0.16865752965338027,
      "grad_norm": 0.19863441041666846,
      "kl": 0.0008147954940795898,
      "learning_rate": 1.2419753086419751e-07,
      "loss": 0.0,
      "reward": 1.7517857924103737,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500275671482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1006
    },
    {
      "completion_length": 236.758939743042,
      "epoch": 0.16899283289324782,
      "grad_norm": 0.323289276257295,
      "kl": 0.0006589889526367188,
      "learning_rate": 1.2444444444444443e-07,
      "loss": 0.0,
      "reward": 1.816071480512619,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.8205357156693935,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1008
    },
    {
      "completion_length": 239.04465293884277,
      "epoch": 0.1693281361331154,
      "grad_norm": 0.21195653521421964,
      "kl": 0.0006917715072631836,
      "learning_rate": 1.2469135802469134e-07,
      "loss": 0.0,
      "reward": 1.7285715118050575,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7464285958558321,
      "rewards/format_reward_func": 0.9821428619325161,
      "step": 1010
    },
    {
      "completion_length": 234.57143878936768,
      "epoch": 0.16966343937298295,
      "grad_norm": 0.14904220568452134,
      "kl": 0.0006827116012573242,
      "learning_rate": 1.2493827160493826e-07,
      "loss": 0.0,
      "reward": 1.7477679178118706,
      "reward_std": 0.053664354141801596,
      "rewards/equation_reward_func": 0.7553571909666061,
      "rewards/format_reward_func": 0.9924107193946838,
      "step": 1012
    },
    {
      "completion_length": 239.071439743042,
      "epoch": 0.1699987426128505,
      "grad_norm": 0.32596573692825204,
      "kl": 0.0007929801940917969,
      "learning_rate": 1.2518518518518517e-07,
      "loss": 0.0,
      "reward": 1.7964286357164383,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7964286059141159,
      "rewards/format_reward_func": 1.0,
      "step": 1014
    },
    {
      "completion_length": 241.7053689956665,
      "epoch": 0.17033404585271805,
      "grad_norm": 0.22868795871007735,
      "kl": 0.0008907318115234375,
      "learning_rate": 1.254320987654321e-07,
      "loss": 0.0,
      "reward": 1.7107143700122833,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7196428813040257,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1016
    },
    {
      "completion_length": 242.83037185668945,
      "epoch": 0.1706693490925856,
      "grad_norm": 0.2450458836005228,
      "kl": 0.0008082389831542969,
      "learning_rate": 1.25679012345679e-07,
      "loss": 0.0,
      "reward": 1.764285795390606,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.7732143215835094,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1018
    },
    {
      "completion_length": 241.7321538925171,
      "epoch": 0.17100465233245316,
      "grad_norm": 0.22439507701355918,
      "kl": 0.0007718801498413086,
      "learning_rate": 1.2592592592592592e-07,
      "loss": 0.0,
      "reward": 1.7607143744826317,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143167406321,
      "rewards/format_reward_func": 1.0,
      "step": 1020
    },
    {
      "completion_length": 239.7857208251953,
      "epoch": 0.1713399555723207,
      "grad_norm": 0.275089302136405,
      "kl": 0.0007205009460449219,
      "learning_rate": 1.2617283950617283e-07,
      "loss": 0.0,
      "reward": 1.7464286461472511,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 1022
    },
    {
      "completion_length": 236.66518783569336,
      "epoch": 0.17167525881218829,
      "grad_norm": 0.31408574841301085,
      "kl": 0.0007870197296142578,
      "learning_rate": 1.2641975308641975e-07,
      "loss": 0.0,
      "reward": 1.6982143744826317,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7026785835623741,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1024
    },
    {
      "completion_length": 244.68751049041748,
      "epoch": 0.17201056205205584,
      "grad_norm": 0.23233340761476015,
      "kl": 0.0008363723754882812,
      "learning_rate": 1.2666666666666666e-07,
      "loss": 0.0,
      "reward": 1.717857226729393,
      "reward_std": 0.0656599160283804,
      "rewards/equation_reward_func": 0.7267857417464256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1026
    },
    {
      "completion_length": 243.33483409881592,
      "epoch": 0.1723458652919234,
      "grad_norm": 0.2969390274202134,
      "kl": 0.000970005989074707,
      "learning_rate": 1.2691358024691358e-07,
      "loss": 0.0,
      "reward": 1.7428572252392769,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 1028
    },
    {
      "completion_length": 232.19643783569336,
      "epoch": 0.17268116853179094,
      "grad_norm": 0.267704808940672,
      "kl": 0.0009148120880126953,
      "learning_rate": 1.271604938271605e-07,
      "loss": 0.0,
      "reward": 1.7482143491506577,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526785954833031,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1030
    },
    {
      "completion_length": 242.52679920196533,
      "epoch": 0.1730164717716585,
      "grad_norm": 0.41808152142365423,
      "kl": 0.0009038448333740234,
      "learning_rate": 1.274074074074074e-07,
      "loss": 0.0,
      "reward": 1.7160715162754059,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.720535745844245,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1032
    },
    {
      "completion_length": 237.8348331451416,
      "epoch": 0.17335177501152604,
      "grad_norm": 0.2222729619708794,
      "kl": 0.0007498264312744141,
      "learning_rate": 1.276543209876543e-07,
      "loss": 0.0,
      "reward": 1.716071493923664,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.720535745844245,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1034
    },
    {
      "completion_length": 239.33483219146729,
      "epoch": 0.1736870782513936,
      "grad_norm": 0.24150968139086826,
      "kl": 0.0009173154830932617,
      "learning_rate": 1.279012345679012e-07,
      "loss": 0.0,
      "reward": 1.7357143461704254,
      "reward_std": 0.06060914974659681,
      "rewards/equation_reward_func": 0.7446428909897804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1036
    },
    {
      "completion_length": 234.47322463989258,
      "epoch": 0.17402238149126115,
      "grad_norm": 0.24846206951979183,
      "kl": 0.0008068084716796875,
      "learning_rate": 1.2814814814814815e-07,
      "loss": 0.0,
      "reward": 1.766071505844593,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1038
    },
    {
      "completion_length": 237.48661708831787,
      "epoch": 0.17435768473112873,
      "grad_norm": 0.23149397808743663,
      "kl": 0.0008314847946166992,
      "learning_rate": 1.2839506172839507e-07,
      "loss": 0.0,
      "reward": 1.7964286357164383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 1040
    },
    {
      "completion_length": 237.63840293884277,
      "epoch": 0.17469298797099628,
      "grad_norm": 0.272804539190451,
      "kl": 0.0009119510650634766,
      "learning_rate": 1.2864197530864195e-07,
      "loss": 0.0,
      "reward": 1.759375087916851,
      "reward_std": 0.0473508988507092,
      "rewards/equation_reward_func": 0.760714303702116,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 1042
    },
    {
      "completion_length": 232.0491180419922,
      "epoch": 0.17502829121086383,
      "grad_norm": 0.26158107193461716,
      "kl": 0.0008912086486816406,
      "learning_rate": 1.2888888888888887e-07,
      "loss": 0.0,
      "reward": 1.705357238650322,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.709821455180645,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1044
    },
    {
      "completion_length": 233.5178680419922,
      "epoch": 0.17536359445073138,
      "grad_norm": 0.16438665975664443,
      "kl": 0.0008466243743896484,
      "learning_rate": 1.291358024691358e-07,
      "loss": 0.0,
      "reward": 1.799553632736206,
      "reward_std": 0.061240497045218945,
      "rewards/equation_reward_func": 0.8026785925030708,
      "rewards/format_reward_func": 0.9968750029802322,
      "step": 1046
    },
    {
      "completion_length": 232.6875123977661,
      "epoch": 0.17569889769059893,
      "grad_norm": 0.1428673378811846,
      "kl": 0.0009279251098632812,
      "learning_rate": 1.2938271604938273e-07,
      "loss": 0.0,
      "reward": 1.7607143595814705,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607142999768257,
      "rewards/format_reward_func": 1.0,
      "step": 1048
    },
    {
      "completion_length": 247.0312614440918,
      "epoch": 0.17603420093046648,
      "grad_norm": 0.24613124143913667,
      "kl": 0.0008966922760009766,
      "learning_rate": 1.2962962962962961e-07,
      "loss": 0.0,
      "reward": 1.760714367032051,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 1050
    },
    {
      "completion_length": 244.1875123977661,
      "epoch": 0.17636950417033403,
      "grad_norm": 0.20124636357271036,
      "kl": 0.00095367431640625,
      "learning_rate": 1.2987654320987653e-07,
      "loss": 0.0,
      "reward": 1.7178572341799736,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7178571708500385,
      "rewards/format_reward_func": 1.0,
      "step": 1052
    },
    {
      "completion_length": 243.13840675354004,
      "epoch": 0.17670480741020161,
      "grad_norm": 0.2653305938876837,
      "kl": 0.0008423328399658203,
      "learning_rate": 1.3012345679012347e-07,
      "loss": 0.0,
      "reward": 1.7285714969038963,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7285714633762836,
      "rewards/format_reward_func": 1.0,
      "step": 1054
    },
    {
      "completion_length": 238.96429824829102,
      "epoch": 0.17704011065006917,
      "grad_norm": 0.33954870085400096,
      "kl": 0.0009109973907470703,
      "learning_rate": 1.3037037037037036e-07,
      "loss": 0.0,
      "reward": 1.725000075995922,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7339286059141159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1056
    },
    {
      "completion_length": 255.60269451141357,
      "epoch": 0.17737541388993672,
      "grad_norm": 0.336539878900831,
      "kl": 0.0009992122650146484,
      "learning_rate": 1.3061728395061727e-07,
      "loss": 0.0,
      "reward": 1.703571505844593,
      "reward_std": 0.09091372601687908,
      "rewards/equation_reward_func": 0.7125000264495611,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1058
    },
    {
      "completion_length": 240.08929634094238,
      "epoch": 0.17771071712980427,
      "grad_norm": 0.26274674597932585,
      "kl": 0.0010219812393188477,
      "learning_rate": 1.308641975308642e-07,
      "loss": 0.0,
      "reward": 1.7392857745289803,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.748214315623045,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1060
    },
    {
      "completion_length": 238.883939743042,
      "epoch": 0.17804602036967182,
      "grad_norm": 0.29858341249542397,
      "kl": 0.000922083854675293,
      "learning_rate": 1.3111111111111113e-07,
      "loss": 0.0,
      "reward": 1.7035715132951736,
      "reward_std": 0.0858629634603858,
      "rewards/equation_reward_func": 0.7035714648663998,
      "rewards/format_reward_func": 1.0,
      "step": 1062
    },
    {
      "completion_length": 238.93751049041748,
      "epoch": 0.17838132360953937,
      "grad_norm": 0.26838733869333925,
      "kl": 0.0009055137634277344,
      "learning_rate": 1.3135802469135802e-07,
      "loss": 0.0,
      "reward": 1.7125000730156898,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7169643323868513,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1064
    },
    {
      "completion_length": 240.852689743042,
      "epoch": 0.17871662684940692,
      "grad_norm": 0.3392750138879499,
      "kl": 0.0009553432464599609,
      "learning_rate": 1.3160493827160493e-07,
      "loss": 0.0,
      "reward": 1.7785715088248253,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7875000238418579,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1066
    },
    {
      "completion_length": 235.77679824829102,
      "epoch": 0.17905193008927447,
      "grad_norm": 0.18746947628661784,
      "kl": 0.001000523567199707,
      "learning_rate": 1.3185185185185185e-07,
      "loss": 0.0,
      "reward": 1.778571479022503,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.778571454808116,
      "rewards/format_reward_func": 1.0,
      "step": 1068
    },
    {
      "completion_length": 241.3928680419922,
      "epoch": 0.17938723332914205,
      "grad_norm": 0.3135635456080588,
      "kl": 0.0010788440704345703,
      "learning_rate": 1.3209876543209874e-07,
      "loss": 0.0,
      "reward": 1.7571429535746574,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 1070
    },
    {
      "completion_length": 239.50001049041748,
      "epoch": 0.1797225365690096,
      "grad_norm": 0.15649356816838816,
      "kl": 0.0009417533874511719,
      "learning_rate": 1.3234567901234568e-07,
      "loss": 0.0,
      "reward": 1.7910714745521545,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7955357357859612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1072
    },
    {
      "completion_length": 242.91518878936768,
      "epoch": 0.18005783980887716,
      "grad_norm": 0.2412609119195414,
      "kl": 0.0009436607360839844,
      "learning_rate": 1.325925925925926e-07,
      "loss": 0.0,
      "reward": 1.775000050663948,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000264495611,
      "rewards/format_reward_func": 1.0,
      "step": 1074
    },
    {
      "completion_length": 230.6696538925171,
      "epoch": 0.1803931430487447,
      "grad_norm": 0.3883932440560228,
      "kl": 0.0012047290802001953,
      "learning_rate": 1.328395061728395e-07,
      "loss": 0.0,
      "reward": 1.7714286521077156,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 1076
    },
    {
      "completion_length": 243.3125114440918,
      "epoch": 0.18072844628861226,
      "grad_norm": 0.2765865793717639,
      "kl": 0.0011830329895019531,
      "learning_rate": 1.330864197530864e-07,
      "loss": 0.0,
      "reward": 1.7928572446107864,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 1078
    },
    {
      "completion_length": 242.84822463989258,
      "epoch": 0.1810637495284798,
      "grad_norm": 0.3631849365890873,
      "kl": 0.0010890960693359375,
      "learning_rate": 1.3333333333333334e-07,
      "loss": 0.0,
      "reward": 1.739285796880722,
      "reward_std": 0.09596448950469494,
      "rewards/equation_reward_func": 0.7482143230736256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1080
    },
    {
      "completion_length": 231.1919765472412,
      "epoch": 0.18139905276834736,
      "grad_norm": 0.12764449962903732,
      "kl": 0.0010178089141845703,
      "learning_rate": 1.3358024691358025e-07,
      "loss": 0.0,
      "reward": 1.757142923772335,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 1082
    },
    {
      "completion_length": 223.36161708831787,
      "epoch": 0.18173435600821494,
      "grad_norm": 0.223794147157261,
      "kl": 0.0009124279022216797,
      "learning_rate": 1.3382716049382717e-07,
      "loss": 0.0,
      "reward": 1.7535714954137802,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 1084
    },
    {
      "completion_length": 240.0357255935669,
      "epoch": 0.1820696592480825,
      "grad_norm": 0.21731252308037757,
      "kl": 0.0009878873825073242,
      "learning_rate": 1.3407407407407405e-07,
      "loss": 0.0,
      "reward": 1.775000087916851,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000394880772,
      "rewards/format_reward_func": 1.0,
      "step": 1086
    },
    {
      "completion_length": 238.94197463989258,
      "epoch": 0.18240496248795005,
      "grad_norm": 0.3074918487704999,
      "kl": 0.0010647773742675781,
      "learning_rate": 1.34320987654321e-07,
      "loss": 0.0,
      "reward": 1.7214286401867867,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.7303571701049805,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1088
    },
    {
      "completion_length": 239.14286708831787,
      "epoch": 0.1827402657278176,
      "grad_norm": 0.13563050504252466,
      "kl": 0.001009225845336914,
      "learning_rate": 1.345679012345679e-07,
      "loss": 0.0,
      "reward": 1.7839286252856255,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7973214648663998,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1090
    },
    {
      "completion_length": 244.07590293884277,
      "epoch": 0.18307556896768515,
      "grad_norm": 0.35995071786593574,
      "kl": 0.0010128021240234375,
      "learning_rate": 1.348148148148148e-07,
      "loss": 0.0,
      "reward": 1.764285795390606,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7732143141329288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1092
    },
    {
      "completion_length": 247.30358600616455,
      "epoch": 0.1834108722075527,
      "grad_norm": 0.20962785028909983,
      "kl": 0.00118255615234375,
      "learning_rate": 1.3506172839506171e-07,
      "loss": 0.0,
      "reward": 1.7321429327130318,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321428917348385,
      "rewards/format_reward_func": 1.0,
      "step": 1094
    },
    {
      "completion_length": 238.3571538925171,
      "epoch": 0.18374617544742025,
      "grad_norm": 0.27384299158814857,
      "kl": 0.0010139942169189453,
      "learning_rate": 1.3530864197530863e-07,
      "loss": 0.0,
      "reward": 1.7571429163217545,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428753435612,
      "rewards/format_reward_func": 1.0,
      "step": 1096
    },
    {
      "completion_length": 239.65179538726807,
      "epoch": 0.1840814786872878,
      "grad_norm": 0.28567982737530095,
      "kl": 0.0012295246124267578,
      "learning_rate": 1.3555555555555557e-07,
      "loss": 0.0,
      "reward": 1.750892922282219,
      "reward_std": 0.06944798585027456,
      "rewards/equation_reward_func": 0.7526785992085934,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 1098
    },
    {
      "completion_length": 237.58929634094238,
      "epoch": 0.18441678192715538,
      "grad_norm": 0.2796157535424371,
      "kl": 0.001096487045288086,
      "learning_rate": 1.3580246913580246e-07,
      "loss": 0.0,
      "reward": 1.7803571969270706,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214581608772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1100
    },
    {
      "completion_length": 236.9107265472412,
      "epoch": 0.18475208516702293,
      "grad_norm": 0.23820606951477952,
      "kl": 0.0011627674102783203,
      "learning_rate": 1.3604938271604937e-07,
      "loss": 0.0,
      "reward": 1.8071429207921028,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428835391998,
      "rewards/format_reward_func": 1.0,
      "step": 1102
    },
    {
      "completion_length": 238.8705472946167,
      "epoch": 0.18508738840689049,
      "grad_norm": 0.17300731372670122,
      "kl": 0.0011518001556396484,
      "learning_rate": 1.362962962962963e-07,
      "loss": 0.0,
      "reward": 1.717857226729393,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7178571671247482,
      "rewards/format_reward_func": 1.0,
      "step": 1104
    },
    {
      "completion_length": 241.66965293884277,
      "epoch": 0.18542269164675804,
      "grad_norm": 0.27020448765160554,
      "kl": 0.0010493993759155273,
      "learning_rate": 1.365432098765432e-07,
      "loss": 0.0,
      "reward": 1.784375049173832,
      "reward_std": 0.05240166233852506,
      "rewards/equation_reward_func": 0.7857143320143223,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 1106
    },
    {
      "completion_length": 243.57590579986572,
      "epoch": 0.1857579948866256,
      "grad_norm": 0.2713591769752031,
      "kl": 0.0012111663818359375,
      "learning_rate": 1.3679012345679012e-07,
      "loss": 0.0,
      "reward": 1.789285808801651,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.789285734295845,
      "rewards/format_reward_func": 1.0,
      "step": 1108
    },
    {
      "completion_length": 247.6339406967163,
      "epoch": 0.18609329812649314,
      "grad_norm": 0.3714389827370947,
      "kl": 0.0012345314025878906,
      "learning_rate": 1.3703703703703703e-07,
      "loss": 0.0,
      "reward": 1.7500000596046448,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 1110
    },
    {
      "completion_length": 241.4107255935669,
      "epoch": 0.1864286013663607,
      "grad_norm": 0.2053685320097415,
      "kl": 0.0010557174682617188,
      "learning_rate": 1.3728395061728395e-07,
      "loss": 0.0,
      "reward": 1.7500000819563866,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000223517418,
      "rewards/format_reward_func": 1.0,
      "step": 1112
    },
    {
      "completion_length": 233.26786994934082,
      "epoch": 0.18676390460622827,
      "grad_norm": 0.33413966119251975,
      "kl": 0.001161813735961914,
      "learning_rate": 1.3753086419753086e-07,
      "loss": 0.0,
      "reward": 1.7910714820027351,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7955357581377029,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1114
    },
    {
      "completion_length": 248.7053680419922,
      "epoch": 0.18709920784609582,
      "grad_norm": 0.19048515531842317,
      "kl": 0.0014069080352783203,
      "learning_rate": 1.3777777777777778e-07,
      "loss": 0.0,
      "reward": 1.7214286401867867,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7303571775555611,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1116
    },
    {
      "completion_length": 247.49108219146729,
      "epoch": 0.18743451108596337,
      "grad_norm": 0.23644455559250388,
      "kl": 0.0011017322540283203,
      "learning_rate": 1.380246913580247e-07,
      "loss": 0.0,
      "reward": 1.7392857819795609,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 1.0,
      "step": 1118
    },
    {
      "completion_length": 241.5982265472412,
      "epoch": 0.18776981432583092,
      "grad_norm": 0.20435446836020607,
      "kl": 0.0010347366333007812,
      "learning_rate": 1.3827160493827158e-07,
      "loss": 0.0,
      "reward": 1.733928643167019,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7383928894996643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1120
    },
    {
      "completion_length": 244.29018878936768,
      "epoch": 0.18810511756569848,
      "grad_norm": 0.2660093786588556,
      "kl": 0.001314401626586914,
      "learning_rate": 1.385185185185185e-07,
      "loss": 0.0,
      "reward": 1.7428572326898575,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7428571730852127,
      "rewards/format_reward_func": 1.0,
      "step": 1122
    },
    {
      "completion_length": 232.94643688201904,
      "epoch": 0.18844042080556603,
      "grad_norm": 0.2939042070750301,
      "kl": 0.0010373592376708984,
      "learning_rate": 1.3876543209876544e-07,
      "loss": 0.0,
      "reward": 1.7714286223053932,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7803571708500385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1124
    },
    {
      "completion_length": 232.59822463989258,
      "epoch": 0.18877572404543358,
      "grad_norm": 0.2232515295771991,
      "kl": 0.0011088848114013672,
      "learning_rate": 1.3901234567901235e-07,
      "loss": 0.0,
      "reward": 1.751785770058632,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500461935997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1126
    },
    {
      "completion_length": 235.14733123779297,
      "epoch": 0.18911102728530113,
      "grad_norm": 0.23194326544395175,
      "kl": 0.0011610984802246094,
      "learning_rate": 1.3925925925925924e-07,
      "loss": 0.0,
      "reward": 1.7428572177886963,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7517857402563095,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1128
    },
    {
      "completion_length": 235.05804634094238,
      "epoch": 0.1894463305251687,
      "grad_norm": 0.2536261881719766,
      "kl": 0.0012238025665283203,
      "learning_rate": 1.3950617283950615e-07,
      "loss": 0.0,
      "reward": 1.7553571984171867,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7598214447498322,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1130
    },
    {
      "completion_length": 224.6384038925171,
      "epoch": 0.18978163376503626,
      "grad_norm": 0.10219346905103746,
      "kl": 0.001035451889038086,
      "learning_rate": 1.397530864197531e-07,
      "loss": 0.0,
      "reward": 1.810714341700077,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143230736256,
      "rewards/format_reward_func": 1.0,
      "step": 1132
    },
    {
      "completion_length": 237.4776906967163,
      "epoch": 0.1901169370049038,
      "grad_norm": 0.20131262973323824,
      "kl": 0.0013401508331298828,
      "learning_rate": 1.4e-07,
      "loss": 0.0,
      "reward": 1.7607143446803093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 1134
    },
    {
      "completion_length": 242.58929443359375,
      "epoch": 0.19045224024477136,
      "grad_norm": 0.3540408384031364,
      "kl": 0.0012521743774414062,
      "learning_rate": 1.402469135802469e-07,
      "loss": 0.0,
      "reward": 1.7696429193019867,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7741071656346321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1136
    },
    {
      "completion_length": 244.79911613464355,
      "epoch": 0.19078754348463892,
      "grad_norm": 0.34346695563107493,
      "kl": 0.001157999038696289,
      "learning_rate": 1.404938271604938e-07,
      "loss": 0.0,
      "reward": 1.7196429371833801,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7241071611642838,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1138
    },
    {
      "completion_length": 241.477689743042,
      "epoch": 0.19112284672450647,
      "grad_norm": 0.20868917961928712,
      "kl": 0.0011320114135742188,
      "learning_rate": 1.4074074074074075e-07,
      "loss": 0.0,
      "reward": 1.7571429088711739,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 1140
    },
    {
      "completion_length": 244.14733409881592,
      "epoch": 0.19145814996437402,
      "grad_norm": 0.32124170225006765,
      "kl": 0.0014166831970214844,
      "learning_rate": 1.4098765432098764e-07,
      "loss": 0.0,
      "reward": 1.7107143476605415,
      "reward_std": 0.08586296439170837,
      "rewards/equation_reward_func": 0.7196428887546062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1142
    },
    {
      "completion_length": 236.6116180419922,
      "epoch": 0.1917934532042416,
      "grad_norm": 0.3877598831982432,
      "kl": 0.0013773441314697266,
      "learning_rate": 1.4123456790123456e-07,
      "loss": 0.0,
      "reward": 1.7232143878936768,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7276786006987095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1144
    },
    {
      "completion_length": 241.6562623977661,
      "epoch": 0.19212875644410915,
      "grad_norm": 0.2101486308792021,
      "kl": 0.0011439323425292969,
      "learning_rate": 1.4148148148148147e-07,
      "loss": 0.0,
      "reward": 1.760714367032051,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 1146
    },
    {
      "completion_length": 238.7857265472412,
      "epoch": 0.1924640596839767,
      "grad_norm": 0.24726688519808554,
      "kl": 0.0011260509490966797,
      "learning_rate": 1.417283950617284e-07,
      "loss": 0.0,
      "reward": 1.8035714849829674,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.8035714477300644,
      "rewards/format_reward_func": 1.0,
      "step": 1148
    },
    {
      "completion_length": 237.46429634094238,
      "epoch": 0.19279936292384425,
      "grad_norm": 0.24592047025377717,
      "kl": 0.001191854476928711,
      "learning_rate": 1.419753086419753e-07,
      "loss": 0.0,
      "reward": 1.7107143849134445,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7107143141329288,
      "rewards/format_reward_func": 1.0,
      "step": 1150
    },
    {
      "completion_length": 237.977689743042,
      "epoch": 0.1931346661637118,
      "grad_norm": 0.3134071733947118,
      "kl": 0.001195669174194336,
      "learning_rate": 1.4222222222222222e-07,
      "loss": 0.0,
      "reward": 1.8178571984171867,
      "reward_std": 0.08586296439170837,
      "rewards/equation_reward_func": 0.8267857357859612,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1152
    },
    {
      "completion_length": 251.6651906967163,
      "epoch": 0.19346996940357936,
      "grad_norm": 0.25882729527590953,
      "kl": 0.0014595985412597656,
      "learning_rate": 1.4246913580246913e-07,
      "loss": 0.0,
      "reward": 1.7571429163217545,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7660714462399483,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1154
    },
    {
      "completion_length": 239.1428689956665,
      "epoch": 0.1938052726434469,
      "grad_norm": 0.22827723141273715,
      "kl": 0.0012450218200683594,
      "learning_rate": 1.4271604938271602e-07,
      "loss": 0.0,
      "reward": 1.8142857626080513,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857290804386,
      "rewards/format_reward_func": 1.0,
      "step": 1156
    },
    {
      "completion_length": 227.54911613464355,
      "epoch": 0.19414057588331446,
      "grad_norm": 0.2404619261901875,
      "kl": 0.0012445449829101562,
      "learning_rate": 1.4296296296296296e-07,
      "loss": 0.0,
      "reward": 1.7464286535978317,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 1158
    },
    {
      "completion_length": 239.5759038925171,
      "epoch": 0.19447587912318204,
      "grad_norm": 0.22192693701997795,
      "kl": 0.0011388063430786133,
      "learning_rate": 1.4320987654320988e-07,
      "loss": 0.0,
      "reward": 1.7714286521077156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 1160
    },
    {
      "completion_length": 240.5357265472412,
      "epoch": 0.1948111823630496,
      "grad_norm": 0.29006089497662224,
      "kl": 0.0012791156768798828,
      "learning_rate": 1.434567901234568e-07,
      "loss": 0.0,
      "reward": 1.7678571939468384,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571809083223,
      "rewards/format_reward_func": 1.0,
      "step": 1162
    },
    {
      "completion_length": 232.29465293884277,
      "epoch": 0.19514648560291714,
      "grad_norm": 0.33240946056531395,
      "kl": 0.0012426376342773438,
      "learning_rate": 1.4370370370370368e-07,
      "loss": 0.0,
      "reward": 1.7428572103381157,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 1164
    },
    {
      "completion_length": 238.4196538925171,
      "epoch": 0.1954817888427847,
      "grad_norm": 0.27403505123708316,
      "kl": 0.0012884140014648438,
      "learning_rate": 1.4395061728395062e-07,
      "loss": 0.0,
      "reward": 1.7250000834465027,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7250000201165676,
      "rewards/format_reward_func": 1.0,
      "step": 1166
    },
    {
      "completion_length": 237.09822368621826,
      "epoch": 0.19581709208265224,
      "grad_norm": 0.4095224412410846,
      "kl": 0.0012373924255371094,
      "learning_rate": 1.4419753086419753e-07,
      "loss": 0.0,
      "reward": 1.8107143267989159,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8107143249362707,
      "rewards/format_reward_func": 1.0,
      "step": 1168
    },
    {
      "completion_length": 241.94197273254395,
      "epoch": 0.1961523953225198,
      "grad_norm": 0.4084487835326827,
      "kl": 0.0016338825225830078,
      "learning_rate": 1.4444444444444442e-07,
      "loss": 0.0,
      "reward": 1.7625000700354576,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7669643051922321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1170
    },
    {
      "completion_length": 230.22322368621826,
      "epoch": 0.19648769856238735,
      "grad_norm": 0.22951686862873494,
      "kl": 0.0014760494232177734,
      "learning_rate": 1.4469135802469134e-07,
      "loss": 0.0,
      "reward": 1.76071435213089,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143074274063,
      "rewards/format_reward_func": 1.0,
      "step": 1172
    },
    {
      "completion_length": 234.08483219146729,
      "epoch": 0.19682300180225493,
      "grad_norm": 0.4243391288107956,
      "kl": 0.001817464828491211,
      "learning_rate": 1.4493827160493828e-07,
      "loss": 0.0,
      "reward": 1.7357143759727478,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7446428872644901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1174
    },
    {
      "completion_length": 238.15625858306885,
      "epoch": 0.19715830504212248,
      "grad_norm": 0.40595547340881505,
      "kl": 0.0014379024505615234,
      "learning_rate": 1.451851851851852e-07,
      "loss": 0.0,
      "reward": 1.6946429386734962,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7080357484519482,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1176
    },
    {
      "completion_length": 236.58483123779297,
      "epoch": 0.19749360828199003,
      "grad_norm": 0.20405546121413234,
      "kl": 0.0012536048889160156,
      "learning_rate": 1.4543209876543208e-07,
      "loss": 0.0,
      "reward": 1.7214286774396896,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7214285992085934,
      "rewards/format_reward_func": 1.0,
      "step": 1178
    },
    {
      "completion_length": 241.3884048461914,
      "epoch": 0.19782891152185758,
      "grad_norm": 0.531077990728977,
      "kl": 0.001644134521484375,
      "learning_rate": 1.45679012345679e-07,
      "loss": 0.0,
      "reward": 1.7196429371833801,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.724107176065445,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1180
    },
    {
      "completion_length": 236.0535831451416,
      "epoch": 0.19816421476172513,
      "grad_norm": 0.16600747936281435,
      "kl": 0.001375436782836914,
      "learning_rate": 1.459259259259259e-07,
      "loss": 0.0,
      "reward": 1.7071429193019867,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7071428894996643,
      "rewards/format_reward_func": 1.0,
      "step": 1182
    },
    {
      "completion_length": 239.51786613464355,
      "epoch": 0.19849951800159268,
      "grad_norm": 0.18321459355862832,
      "kl": 0.0015826225280761719,
      "learning_rate": 1.4617283950617285e-07,
      "loss": 0.0,
      "reward": 1.76607146859169,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7705357521772385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1184
    },
    {
      "completion_length": 234.85268688201904,
      "epoch": 0.19883482124146024,
      "grad_norm": 0.2741705293832859,
      "kl": 0.002036571502685547,
      "learning_rate": 1.4641975308641974e-07,
      "loss": 0.0,
      "reward": 1.760714367032051,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7607143055647612,
      "rewards/format_reward_func": 1.0,
      "step": 1186
    },
    {
      "completion_length": 242.77679443359375,
      "epoch": 0.1991701244813278,
      "grad_norm": 0.2830046064019713,
      "kl": 0.0015716552734375,
      "learning_rate": 1.4666666666666666e-07,
      "loss": 0.0,
      "reward": 1.6839286535978317,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.6883928831666708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1188
    },
    {
      "completion_length": 245.2276906967163,
      "epoch": 0.19950542772119537,
      "grad_norm": 0.3715662769891078,
      "kl": 0.0017573833465576172,
      "learning_rate": 1.4691358024691357e-07,
      "loss": 0.0,
      "reward": 1.7375000789761543,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7419643141329288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1190
    },
    {
      "completion_length": 244.12054538726807,
      "epoch": 0.19984073096106292,
      "grad_norm": 0.27907119303296496,
      "kl": 0.0016562938690185547,
      "learning_rate": 1.4716049382716049e-07,
      "loss": 0.0,
      "reward": 1.8035714849829674,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8035714402794838,
      "rewards/format_reward_func": 1.0,
      "step": 1192
    },
    {
      "completion_length": 243.1205472946167,
      "epoch": 0.20017603420093047,
      "grad_norm": 0.45567982927097334,
      "kl": 0.0018799304962158203,
      "learning_rate": 1.474074074074074e-07,
      "loss": 0.0,
      "reward": 1.7321429401636124,
      "reward_std": 0.08586296439170837,
      "rewards/equation_reward_func": 0.7410714589059353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1194
    },
    {
      "completion_length": 232.75447368621826,
      "epoch": 0.20051133744079802,
      "grad_norm": 0.21608914943333704,
      "kl": 0.0015969276428222656,
      "learning_rate": 1.4765432098765432e-07,
      "loss": 0.0,
      "reward": 1.8000000417232513,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000193715096,
      "rewards/format_reward_func": 1.0,
      "step": 1196
    },
    {
      "completion_length": 241.44197368621826,
      "epoch": 0.20084664068066557,
      "grad_norm": 0.26748599126218425,
      "kl": 0.0018237829208374023,
      "learning_rate": 1.4790123456790123e-07,
      "loss": 0.0,
      "reward": 1.814285784959793,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857402563095,
      "rewards/format_reward_func": 1.0,
      "step": 1198
    },
    {
      "completion_length": 228.45983219146729,
      "epoch": 0.20118194392053312,
      "grad_norm": 0.28378913534322847,
      "kl": 0.0016875267028808594,
      "learning_rate": 1.4814814814814815e-07,
      "loss": 0.0,
      "reward": 1.7535714954137802,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 1200
    },
    {
      "completion_length": 241.91519165039062,
      "epoch": 0.20151724716040068,
      "grad_norm": 0.2504716674270821,
      "kl": 0.0017731189727783203,
      "learning_rate": 1.4839506172839506e-07,
      "loss": 0.0,
      "reward": 1.7714286297559738,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7803571745753288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1202
    },
    {
      "completion_length": 243.80804538726807,
      "epoch": 0.20185255040026825,
      "grad_norm": 0.33704098188644993,
      "kl": 0.002542257308959961,
      "learning_rate": 1.4864197530864197e-07,
      "loss": 0.0,
      "reward": 1.7053572237491608,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7098214663565159,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1204
    },
    {
      "completion_length": 234.44643878936768,
      "epoch": 0.2021878536401358,
      "grad_norm": 0.2747468585504003,
      "kl": 0.0015654563903808594,
      "learning_rate": 1.4888888888888886e-07,
      "loss": 0.0,
      "reward": 1.7964286357164383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7964285872876644,
      "rewards/format_reward_func": 1.0,
      "step": 1206
    },
    {
      "completion_length": 237.40625953674316,
      "epoch": 0.20252315688000336,
      "grad_norm": 0.24314632357287652,
      "kl": 0.001814126968383789,
      "learning_rate": 1.4913580246913578e-07,
      "loss": 0.0,
      "reward": 1.7107143551111221,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7107143215835094,
      "rewards/format_reward_func": 1.0,
      "step": 1208
    },
    {
      "completion_length": 239.18304824829102,
      "epoch": 0.2028584601198709,
      "grad_norm": 0.24318172658651124,
      "kl": 0.002092123031616211,
      "learning_rate": 1.4938271604938272e-07,
      "loss": 0.0,
      "reward": 1.828571505844593,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8285714499652386,
      "rewards/format_reward_func": 1.0,
      "step": 1210
    },
    {
      "completion_length": 238.27679824829102,
      "epoch": 0.20319376335973846,
      "grad_norm": 0.32728322145173633,
      "kl": 0.003085613250732422,
      "learning_rate": 1.4962962962962963e-07,
      "loss": 0.0,
      "reward": 1.744642935693264,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7491071671247482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1212
    },
    {
      "completion_length": 240.79465198516846,
      "epoch": 0.203529066599606,
      "grad_norm": 0.283836659998682,
      "kl": 0.0021562576293945312,
      "learning_rate": 1.4987654320987652e-07,
      "loss": 0.0,
      "reward": 1.716071493923664,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7294643130153418,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1214
    },
    {
      "completion_length": 244.0491180419922,
      "epoch": 0.20386436983947356,
      "grad_norm": 0.26081031140969885,
      "kl": 0.0022537708282470703,
      "learning_rate": 1.5012345679012344e-07,
      "loss": 0.0,
      "reward": 1.742857202887535,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.742857189849019,
      "rewards/format_reward_func": 1.0,
      "step": 1216
    },
    {
      "completion_length": 241.91072845458984,
      "epoch": 0.20419967307934114,
      "grad_norm": 0.30485410684658837,
      "kl": 0.0026428699493408203,
      "learning_rate": 1.5037037037037038e-07,
      "loss": 0.0,
      "reward": 1.7642857730388641,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7732143215835094,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1218
    },
    {
      "completion_length": 241.0669755935669,
      "epoch": 0.2045349763192087,
      "grad_norm": 0.2265319246913318,
      "kl": 0.0030205249786376953,
      "learning_rate": 1.506172839506173e-07,
      "loss": 0.0,
      "reward": 1.8232143446803093,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.8276785835623741,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1220
    },
    {
      "completion_length": 238.30358028411865,
      "epoch": 0.20487027955907625,
      "grad_norm": 0.2382537179736265,
      "kl": 0.0023665428161621094,
      "learning_rate": 1.5086419753086418e-07,
      "loss": 0.0,
      "reward": 1.8017857819795609,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.8062500357627869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1222
    },
    {
      "completion_length": 245.17411708831787,
      "epoch": 0.2052055827989438,
      "grad_norm": 0.3324516681199202,
      "kl": 0.0020515918731689453,
      "learning_rate": 1.511111111111111e-07,
      "loss": 0.0,
      "reward": 1.7857143357396126,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143245637417,
      "rewards/format_reward_func": 1.0,
      "step": 1224
    },
    {
      "completion_length": 239.4196548461914,
      "epoch": 0.20554088603881135,
      "grad_norm": 0.25251488896253566,
      "kl": 0.0025861263275146484,
      "learning_rate": 1.5135802469135804e-07,
      "loss": 0.0,
      "reward": 1.7250000983476639,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000424683094,
      "rewards/format_reward_func": 1.0,
      "step": 1226
    },
    {
      "completion_length": 236.84822463989258,
      "epoch": 0.2058761892786789,
      "grad_norm": 0.23428687641725407,
      "kl": 0.0025599002838134766,
      "learning_rate": 1.5160493827160493e-07,
      "loss": 0.0,
      "reward": 1.6964286267757416,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.6964286118745804,
      "rewards/format_reward_func": 1.0,
      "step": 1228
    },
    {
      "completion_length": 252.34375953674316,
      "epoch": 0.20621149251854645,
      "grad_norm": 0.32242002681200926,
      "kl": 0.0028755664825439453,
      "learning_rate": 1.5185185185185184e-07,
      "loss": 0.0,
      "reward": 1.7125000804662704,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7169643137603998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1230
    },
    {
      "completion_length": 242.9107265472412,
      "epoch": 0.206546795758414,
      "grad_norm": 0.24004483575380084,
      "kl": 0.002190113067626953,
      "learning_rate": 1.5209876543209876e-07,
      "loss": 0.0,
      "reward": 1.7250000834465027,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7339285984635353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1232
    },
    {
      "completion_length": 237.33036708831787,
      "epoch": 0.20688209899828158,
      "grad_norm": 0.24368559931981323,
      "kl": 0.0017604827880859375,
      "learning_rate": 1.523456790123457e-07,
      "loss": 0.0,
      "reward": 1.7660715132951736,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1234
    },
    {
      "completion_length": 231.47322368621826,
      "epoch": 0.20721740223814913,
      "grad_norm": 0.5328530490006232,
      "kl": 0.0025615692138671875,
      "learning_rate": 1.5259259259259259e-07,
      "loss": 0.0,
      "reward": 1.7785715162754059,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 1236
    },
    {
      "completion_length": 240.35268878936768,
      "epoch": 0.20755270547801669,
      "grad_norm": 0.18652124710992135,
      "kl": 0.002346515655517578,
      "learning_rate": 1.528395061728395e-07,
      "loss": 0.0,
      "reward": 1.776785783469677,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7812500223517418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1238
    },
    {
      "completion_length": 236.0446548461914,
      "epoch": 0.20788800871788424,
      "grad_norm": 0.30917406808952935,
      "kl": 0.0021347999572753906,
      "learning_rate": 1.5308641975308642e-07,
      "loss": 0.0,
      "reward": 1.7455357909202576,
      "reward_std": 0.07702413015067577,
      "rewards/equation_reward_func": 0.7517857402563095,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 1240
    },
    {
      "completion_length": 246.4285831451416,
      "epoch": 0.2082233119577518,
      "grad_norm": 0.24049563158557588,
      "kl": 0.004585981369018555,
      "learning_rate": 1.533333333333333e-07,
      "loss": 0.0,
      "reward": 1.7035715207457542,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7035714723169804,
      "rewards/format_reward_func": 1.0,
      "step": 1242
    },
    {
      "completion_length": 252.42858123779297,
      "epoch": 0.20855861519761934,
      "grad_norm": 0.28232304155666915,
      "kl": 0.004839181900024414,
      "learning_rate": 1.5358024691358024e-07,
      "loss": 0.0,
      "reward": 1.8035715073347092,
      "reward_std": 0.09596449043601751,
      "rewards/equation_reward_func": 0.8125000149011612,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1244
    },
    {
      "completion_length": 237.1562614440918,
      "epoch": 0.2088939184374869,
      "grad_norm": 0.252566996772104,
      "kl": 0.0032820701599121094,
      "learning_rate": 1.5382716049382716e-07,
      "loss": 0.0,
      "reward": 1.8035714626312256,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 1246
    },
    {
      "completion_length": 224.24554538726807,
      "epoch": 0.20922922167735447,
      "grad_norm": 0.4379151122153357,
      "kl": 0.0019059181213378906,
      "learning_rate": 1.5407407407407407e-07,
      "loss": 0.0,
      "reward": 1.7571429461240768,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 1248
    },
    {
      "completion_length": 243.2366189956665,
      "epoch": 0.20956452491722202,
      "grad_norm": 0.2435262045125475,
      "kl": 0.0021789073944091797,
      "learning_rate": 1.5432098765432096e-07,
      "loss": 0.0,
      "reward": 1.7750000730156898,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7839285843074322,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1250
    },
    {
      "completion_length": 240.821439743042,
      "epoch": 0.20989982815708957,
      "grad_norm": 0.3508141244807139,
      "kl": 0.002946138381958008,
      "learning_rate": 1.545679012345679e-07,
      "loss": 0.0,
      "reward": 1.6964286416769028,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.6964286155998707,
      "rewards/format_reward_func": 1.0,
      "step": 1252
    },
    {
      "completion_length": 247.0669755935669,
      "epoch": 0.21023513139695713,
      "grad_norm": 0.24180904017268992,
      "kl": 0.002747058868408203,
      "learning_rate": 1.5481481481481482e-07,
      "loss": 0.0,
      "reward": 1.7857143431901932,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7946428805589676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1254
    },
    {
      "completion_length": 229.96875858306885,
      "epoch": 0.21057043463682468,
      "grad_norm": 0.2411987768476652,
      "kl": 0.002772808074951172,
      "learning_rate": 1.550617283950617e-07,
      "loss": 0.0,
      "reward": 1.7464286461472511,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286033064127,
      "rewards/format_reward_func": 1.0,
      "step": 1256
    },
    {
      "completion_length": 236.8303680419922,
      "epoch": 0.21090573787669223,
      "grad_norm": 0.32063347431687766,
      "kl": 0.002824068069458008,
      "learning_rate": 1.5530864197530862e-07,
      "loss": 0.0,
      "reward": 1.7892857789993286,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 1258
    },
    {
      "completion_length": 241.53125953674316,
      "epoch": 0.21124104111655978,
      "grad_norm": 0.31144432521616555,
      "kl": 0.004393339157104492,
      "learning_rate": 1.5555555555555556e-07,
      "loss": 0.0,
      "reward": 1.782142885029316,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821429073810577,
      "rewards/format_reward_func": 1.0,
      "step": 1260
    },
    {
      "completion_length": 238.9642972946167,
      "epoch": 0.21157634435642733,
      "grad_norm": 0.29715450473241195,
      "kl": 0.004702568054199219,
      "learning_rate": 1.5580246913580248e-07,
      "loss": 0.0,
      "reward": 1.7357143610715866,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7357143312692642,
      "rewards/format_reward_func": 1.0,
      "step": 1262
    },
    {
      "completion_length": 241.18751049041748,
      "epoch": 0.2119116475962949,
      "grad_norm": 0.1951487271047744,
      "kl": 0.0032672882080078125,
      "learning_rate": 1.5604938271604937e-07,
      "loss": 0.0,
      "reward": 1.7500000596046448,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 1264
    },
    {
      "completion_length": 241.1384038925171,
      "epoch": 0.21224695083616246,
      "grad_norm": 0.24012888678173097,
      "kl": 0.0034089088439941406,
      "learning_rate": 1.5629629629629628e-07,
      "loss": 0.0,
      "reward": 1.7678572088479996,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 1266
    },
    {
      "completion_length": 232.91965293884277,
      "epoch": 0.21258225407603001,
      "grad_norm": 0.25773928795116036,
      "kl": 0.003378152847290039,
      "learning_rate": 1.565432098765432e-07,
      "loss": 0.0,
      "reward": 1.7964286357164383,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.8053571693599224,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1268
    },
    {
      "completion_length": 243.12054920196533,
      "epoch": 0.21291755731589757,
      "grad_norm": 0.17985039137077788,
      "kl": 0.003942012786865234,
      "learning_rate": 1.5679012345679014e-07,
      "loss": 0.0,
      "reward": 1.7482143491506577,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526786010712385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1270
    },
    {
      "completion_length": 236.90626049041748,
      "epoch": 0.21325286055576512,
      "grad_norm": 0.1548271078866053,
      "kl": 0.005479335784912109,
      "learning_rate": 1.5703703703703703e-07,
      "loss": 0.0,
      "reward": 1.7303572297096252,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7348214648663998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1272
    },
    {
      "completion_length": 241.95536708831787,
      "epoch": 0.21358816379563267,
      "grad_norm": 0.27982895939363556,
      "kl": 0.004017353057861328,
      "learning_rate": 1.5728395061728394e-07,
      "loss": 0.0,
      "reward": 1.79642865806818,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 1274
    },
    {
      "completion_length": 246.4598331451416,
      "epoch": 0.21392346703550022,
      "grad_norm": 0.24055112645480836,
      "kl": 0.00449061393737793,
      "learning_rate": 1.5753086419753086e-07,
      "loss": 0.0,
      "reward": 1.7892857789993286,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 1276
    },
    {
      "completion_length": 242.5312623977661,
      "epoch": 0.2142587702753678,
      "grad_norm": 0.19247495138177573,
      "kl": 0.002553701400756836,
      "learning_rate": 1.5777777777777777e-07,
      "loss": 0.0,
      "reward": 1.8089286163449287,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8133928887546062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1278
    },
    {
      "completion_length": 242.32144260406494,
      "epoch": 0.21459407351523535,
      "grad_norm": 0.1627487076419372,
      "kl": 0.0036182403564453125,
      "learning_rate": 1.5802469135802468e-07,
      "loss": 0.0,
      "reward": 1.7607143372297287,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 1280
    },
    {
      "completion_length": 240.27233123779297,
      "epoch": 0.2149293767551029,
      "grad_norm": 0.22177047448633702,
      "kl": 0.005349874496459961,
      "learning_rate": 1.582716049382716e-07,
      "loss": 0.0,
      "reward": 1.7321429252624512,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428880095482,
      "rewards/format_reward_func": 1.0,
      "step": 1282
    },
    {
      "completion_length": 234.53572463989258,
      "epoch": 0.21526467999497045,
      "grad_norm": 0.2958328217477393,
      "kl": 0.004651069641113281,
      "learning_rate": 1.5851851851851851e-07,
      "loss": 0.0,
      "reward": 1.7214286476373672,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.721428606659174,
      "rewards/format_reward_func": 1.0,
      "step": 1284
    },
    {
      "completion_length": 241.8080472946167,
      "epoch": 0.215599983234838,
      "grad_norm": 0.33299459227609995,
      "kl": 0.004356861114501953,
      "learning_rate": 1.5876543209876543e-07,
      "loss": 0.0,
      "reward": 1.7285714969038963,
      "reward_std": 0.09091372694820166,
      "rewards/equation_reward_func": 0.7375000417232513,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1286
    },
    {
      "completion_length": 235.60268878936768,
      "epoch": 0.21593528647470556,
      "grad_norm": 0.35271283163499656,
      "kl": 0.005709171295166016,
      "learning_rate": 1.5901234567901234e-07,
      "loss": 0.0,
      "reward": 1.7089286595582962,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7133928686380386,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1288
    },
    {
      "completion_length": 244.56251049041748,
      "epoch": 0.2162705897145731,
      "grad_norm": 0.1859424366173882,
      "kl": 0.003217935562133789,
      "learning_rate": 1.5925925925925926e-07,
      "loss": 0.0,
      "reward": 1.7625000476837158,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7669642996042967,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1290
    },
    {
      "completion_length": 240.76786708831787,
      "epoch": 0.21660589295444066,
      "grad_norm": 0.24532934625705524,
      "kl": 0.0031633377075195312,
      "learning_rate": 1.5950617283950615e-07,
      "loss": 0.0,
      "reward": 1.6625000908970833,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.6669643167406321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1292
    },
    {
      "completion_length": 243.30358505249023,
      "epoch": 0.21694119619430824,
      "grad_norm": 0.2586537390088073,
      "kl": 0.002529144287109375,
      "learning_rate": 1.5975308641975306e-07,
      "loss": 0.0,
      "reward": 1.776785783469677,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7812500260770321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1294
    },
    {
      "completion_length": 234.05357933044434,
      "epoch": 0.2172764994341758,
      "grad_norm": 0.20464408868137368,
      "kl": 0.006105184555053711,
      "learning_rate": 1.6e-07,
      "loss": 0.0,
      "reward": 1.7696429193019867,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.774107176810503,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1296
    },
    {
      "completion_length": 243.42858123779297,
      "epoch": 0.21761180267404334,
      "grad_norm": 0.4311301072230461,
      "kl": 0.004184722900390625,
      "learning_rate": 1.6024691358024692e-07,
      "loss": 0.0,
      "reward": 1.7160714864730835,
      "reward_std": 0.11364215798676014,
      "rewards/equation_reward_func": 0.7294643148779869,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1298
    },
    {
      "completion_length": 233.65626049041748,
      "epoch": 0.2179471059139109,
      "grad_norm": 0.5118573409819533,
      "kl": 0.006073713302612305,
      "learning_rate": 1.604938271604938e-07,
      "loss": 0.0,
      "reward": 1.7464286535978317,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7553571686148643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1300
    },
    {
      "completion_length": 242.00894165039062,
      "epoch": 0.21828240915377844,
      "grad_norm": 0.18901941957665191,
      "kl": 0.004821300506591797,
      "learning_rate": 1.6074074074074072e-07,
      "loss": 0.0,
      "reward": 1.7214286476373672,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7214285992085934,
      "rewards/format_reward_func": 1.0,
      "step": 1302
    },
    {
      "completion_length": 243.61608123779297,
      "epoch": 0.218617712393646,
      "grad_norm": 0.2868402947325543,
      "kl": 0.004178762435913086,
      "learning_rate": 1.6098765432098766e-07,
      "loss": 0.0,
      "reward": 1.7732143253087997,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7776785865426064,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1304
    },
    {
      "completion_length": 231.63840293884277,
      "epoch": 0.21895301563351355,
      "grad_norm": 0.28362586224310077,
      "kl": 0.0027713775634765625,
      "learning_rate": 1.6123456790123455e-07,
      "loss": 0.0,
      "reward": 1.7857143357396126,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7857143357396126,
      "rewards/format_reward_func": 1.0,
      "step": 1306
    },
    {
      "completion_length": 244.44643783569336,
      "epoch": 0.21928831887338113,
      "grad_norm": 0.2076709051554259,
      "kl": 0.0029850006103515625,
      "learning_rate": 1.6148148148148147e-07,
      "loss": 0.0,
      "reward": 1.703571505844593,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7125000357627869,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1308
    },
    {
      "completion_length": 233.55358219146729,
      "epoch": 0.21962362211324868,
      "grad_norm": 0.25018701924401665,
      "kl": 0.0067141056060791016,
      "learning_rate": 1.6172839506172838e-07,
      "loss": 0.0,
      "reward": 1.782142922282219,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 1310
    },
    {
      "completion_length": 239.6026906967163,
      "epoch": 0.21995892535311623,
      "grad_norm": 0.3511161464206752,
      "kl": 0.00603485107421875,
      "learning_rate": 1.6197530864197532e-07,
      "loss": 0.0,
      "reward": 1.7714286297559738,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 1312
    },
    {
      "completion_length": 240.23215579986572,
      "epoch": 0.22029422859298378,
      "grad_norm": 0.28800201598541564,
      "kl": 0.003075122833251953,
      "learning_rate": 1.622222222222222e-07,
      "loss": 0.0,
      "reward": 1.7714286670088768,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.7803571633994579,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1314
    },
    {
      "completion_length": 234.04019165039062,
      "epoch": 0.22062953183285133,
      "grad_norm": 0.2676481126389581,
      "kl": 0.004632472991943359,
      "learning_rate": 1.6246913580246912e-07,
      "loss": 0.0,
      "reward": 1.7339286282658577,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7383929006755352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1316
    },
    {
      "completion_length": 244.27679443359375,
      "epoch": 0.22096483507271888,
      "grad_norm": 0.3087973565894431,
      "kl": 0.010606765747070312,
      "learning_rate": 1.6271604938271604e-07,
      "loss": 0.0,
      "reward": 1.6928572058677673,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7017857395112514,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1318
    },
    {
      "completion_length": 241.70983123779297,
      "epoch": 0.22130013831258644,
      "grad_norm": 0.25034205125370274,
      "kl": 0.00440216064453125,
      "learning_rate": 1.6296296296296298e-07,
      "loss": 0.0,
      "reward": 1.7589286267757416,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7723214589059353,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1320
    },
    {
      "completion_length": 234.6785831451416,
      "epoch": 0.221635441552454,
      "grad_norm": 0.41038167641507267,
      "kl": 0.00315093994140625,
      "learning_rate": 1.6320987654320987e-07,
      "loss": 0.0,
      "reward": 1.7392857819795609,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 1.0,
      "step": 1322
    },
    {
      "completion_length": 245.37947368621826,
      "epoch": 0.22197074479232157,
      "grad_norm": 0.806420438279851,
      "kl": 0.009297370910644531,
      "learning_rate": 1.6345679012345678e-07,
      "loss": 0.0,
      "reward": 1.7625000551342964,
      "reward_std": 0.0732360603287816,
      "rewards/equation_reward_func": 0.7758928723633289,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1324
    },
    {
      "completion_length": 229.21429538726807,
      "epoch": 0.22230604803218912,
      "grad_norm": 0.29786161917524756,
      "kl": 0.004057168960571289,
      "learning_rate": 1.637037037037037e-07,
      "loss": 0.0,
      "reward": 1.7857143580913544,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7857143059372902,
      "rewards/format_reward_func": 1.0,
      "step": 1326
    },
    {
      "completion_length": 241.70090293884277,
      "epoch": 0.22264135127205667,
      "grad_norm": 0.2705780895983189,
      "kl": 0.0032672882080078125,
      "learning_rate": 1.639506172839506e-07,
      "loss": 0.0,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 1328
    },
    {
      "completion_length": 244.21429538726807,
      "epoch": 0.22297665451192422,
      "grad_norm": 0.24169684934286853,
      "kl": 0.003974437713623047,
      "learning_rate": 1.6419753086419753e-07,
      "loss": 0.0,
      "reward": 1.771428644657135,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 1330
    },
    {
      "completion_length": 236.27679634094238,
      "epoch": 0.22331195775179177,
      "grad_norm": 0.2513951306851172,
      "kl": 0.003792285919189453,
      "learning_rate": 1.6444444444444444e-07,
      "loss": 0.0,
      "reward": 1.7232143506407738,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7276785969734192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1332
    },
    {
      "completion_length": 239.6830472946167,
      "epoch": 0.22364726099165932,
      "grad_norm": 0.26017730409226486,
      "kl": 0.006929159164428711,
      "learning_rate": 1.6469135802469136e-07,
      "loss": 0.0,
      "reward": 1.8214286267757416,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8214285969734192,
      "rewards/format_reward_func": 1.0,
      "step": 1334
    },
    {
      "completion_length": 240.4062614440918,
      "epoch": 0.22398256423152688,
      "grad_norm": 0.33342545268150153,
      "kl": 0.007687568664550781,
      "learning_rate": 1.6493827160493825e-07,
      "loss": 0.0,
      "reward": 1.765178643167019,
      "reward_std": 0.03914341004565358,
      "rewards/equation_reward_func": 0.7669643089175224,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 1336
    },
    {
      "completion_length": 238.1294755935669,
      "epoch": 0.22431786747139446,
      "grad_norm": 0.28918367802683476,
      "kl": 0.003994941711425781,
      "learning_rate": 1.651851851851852e-07,
      "loss": 0.0,
      "reward": 1.7035715281963348,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7035714536905289,
      "rewards/format_reward_func": 1.0,
      "step": 1338
    },
    {
      "completion_length": 237.70090103149414,
      "epoch": 0.224653170711262,
      "grad_norm": 0.2494069780734928,
      "kl": 0.011424541473388672,
      "learning_rate": 1.654320987654321e-07,
      "loss": 0.0,
      "reward": 1.741071492433548,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7455357518047094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1340
    },
    {
      "completion_length": 237.7991180419922,
      "epoch": 0.22498847395112956,
      "grad_norm": 0.303123570931485,
      "kl": 0.003299713134765625,
      "learning_rate": 1.65679012345679e-07,
      "loss": 0.0,
      "reward": 1.6857143640518188,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.6946428939700127,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1342
    },
    {
      "completion_length": 245.04911708831787,
      "epoch": 0.2253237771909971,
      "grad_norm": 0.2523171541215057,
      "kl": 0.005613803863525391,
      "learning_rate": 1.659259259259259e-07,
      "loss": 0.0,
      "reward": 1.7571429088711739,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7660714723169804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1344
    },
    {
      "completion_length": 243.0937614440918,
      "epoch": 0.22565908043086466,
      "grad_norm": 0.14018283087264063,
      "kl": 0.0050811767578125,
      "learning_rate": 1.6617283950617285e-07,
      "loss": 0.0,
      "reward": 1.7250000685453415,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000443309546,
      "rewards/format_reward_func": 1.0,
      "step": 1346
    },
    {
      "completion_length": 241.71429443359375,
      "epoch": 0.2259943836707322,
      "grad_norm": 0.41487774404342265,
      "kl": 0.005957126617431641,
      "learning_rate": 1.6641975308641976e-07,
      "loss": 0.0,
      "reward": 1.7410714998841286,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7455357573926449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1348
    },
    {
      "completion_length": 230.00893878936768,
      "epoch": 0.22632968691059976,
      "grad_norm": 0.2806775670690932,
      "kl": 0.00467681884765625,
      "learning_rate": 1.6666666666666665e-07,
      "loss": 0.0,
      "reward": 1.7714286372065544,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7803571708500385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1350
    },
    {
      "completion_length": 231.69643878936768,
      "epoch": 0.22666499015046732,
      "grad_norm": 0.24691130694143,
      "kl": 0.005158901214599609,
      "learning_rate": 1.6691358024691357e-07,
      "loss": 0.0,
      "reward": 1.7321429401636124,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7321428805589676,
      "rewards/format_reward_func": 1.0,
      "step": 1352
    },
    {
      "completion_length": 240.7857255935669,
      "epoch": 0.2270002933903349,
      "grad_norm": 0.24994952871151663,
      "kl": 0.005618095397949219,
      "learning_rate": 1.6716049382716048e-07,
      "loss": 0.0,
      "reward": 1.7571429163217545,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428921073675,
      "rewards/format_reward_func": 1.0,
      "step": 1354
    },
    {
      "completion_length": 248.33929824829102,
      "epoch": 0.22733559663020245,
      "grad_norm": 0.09872595517896748,
      "kl": 0.009372234344482422,
      "learning_rate": 1.674074074074074e-07,
      "loss": 0.0,
      "reward": 1.717857226729393,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7267857305705547,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1356
    },
    {
      "completion_length": 245.62054634094238,
      "epoch": 0.22767089987007,
      "grad_norm": 0.2373904924245633,
      "kl": 0.008755207061767578,
      "learning_rate": 1.676543209876543e-07,
      "loss": 0.0,
      "reward": 1.710714377462864,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7107143122702837,
      "rewards/format_reward_func": 1.0,
      "step": 1358
    },
    {
      "completion_length": 234.75893878936768,
      "epoch": 0.22800620310993755,
      "grad_norm": 0.20854539089599378,
      "kl": 0.007511138916015625,
      "learning_rate": 1.6790123456790122e-07,
      "loss": 0.0,
      "reward": 1.7500000670552254,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 1360
    },
    {
      "completion_length": 234.74108028411865,
      "epoch": 0.2283415063498051,
      "grad_norm": 0.28694424946821906,
      "kl": 0.0039539337158203125,
      "learning_rate": 1.6814814814814814e-07,
      "loss": 0.0,
      "reward": 1.7000000700354576,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7000000551342964,
      "rewards/format_reward_func": 1.0,
      "step": 1362
    },
    {
      "completion_length": 241.7366189956665,
      "epoch": 0.22867680958967265,
      "grad_norm": 0.3141697765365759,
      "kl": 0.010674476623535156,
      "learning_rate": 1.6839506172839505e-07,
      "loss": 0.0,
      "reward": 1.8071429431438446,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.8071428835391998,
      "rewards/format_reward_func": 1.0,
      "step": 1364
    },
    {
      "completion_length": 239.72322368621826,
      "epoch": 0.2290121128295402,
      "grad_norm": 0.2671441737251063,
      "kl": 0.005108356475830078,
      "learning_rate": 1.6864197530864197e-07,
      "loss": 0.0,
      "reward": 1.7696429267525673,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1366
    },
    {
      "completion_length": 228.58483219146729,
      "epoch": 0.22934741606940778,
      "grad_norm": 0.21440951279482529,
      "kl": 0.003632068634033203,
      "learning_rate": 1.6888888888888888e-07,
      "loss": 0.0,
      "reward": 1.74642863124609,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464286126196384,
      "rewards/format_reward_func": 1.0,
      "step": 1368
    },
    {
      "completion_length": 235.47322463989258,
      "epoch": 0.22968271930927533,
      "grad_norm": 0.20954095826447186,
      "kl": 0.008241653442382812,
      "learning_rate": 1.691358024691358e-07,
      "loss": 0.0,
      "reward": 1.7107143700122833,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.710714315995574,
      "rewards/format_reward_func": 1.0,
      "step": 1370
    },
    {
      "completion_length": 237.51786994934082,
      "epoch": 0.2300180225491429,
      "grad_norm": 0.17949433295497624,
      "kl": 0.007944583892822266,
      "learning_rate": 1.693827160493827e-07,
      "loss": 0.0,
      "reward": 1.7250000536441803,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.72500004991889,
      "rewards/format_reward_func": 1.0,
      "step": 1372
    },
    {
      "completion_length": 239.4687614440918,
      "epoch": 0.23035332578901044,
      "grad_norm": 0.3259964424478922,
      "kl": 0.004845619201660156,
      "learning_rate": 1.6962962962962963e-07,
      "loss": 0.0,
      "reward": 1.7285715192556381,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7285714671015739,
      "rewards/format_reward_func": 1.0,
      "step": 1374
    },
    {
      "completion_length": 241.49108409881592,
      "epoch": 0.230688629028878,
      "grad_norm": 0.27244331118122284,
      "kl": 0.009942054748535156,
      "learning_rate": 1.6987654320987654e-07,
      "loss": 0.0,
      "reward": 1.7714286595582962,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7714285925030708,
      "rewards/format_reward_func": 1.0,
      "step": 1376
    },
    {
      "completion_length": 236.321439743042,
      "epoch": 0.23102393226874554,
      "grad_norm": 0.07781785904774369,
      "kl": 0.0042362213134765625,
      "learning_rate": 1.7012345679012343e-07,
      "loss": 0.0,
      "reward": 1.7178572192788124,
      "reward_std": 0.017677669413387775,
      "rewards/equation_reward_func": 0.7258928790688515,
      "rewards/format_reward_func": 0.9919642955064774,
      "step": 1378
    },
    {
      "completion_length": 237.6562623977661,
      "epoch": 0.2313592355086131,
      "grad_norm": 0.3306974061678042,
      "kl": 0.02078533172607422,
      "learning_rate": 1.7037037037037035e-07,
      "loss": 0.0,
      "reward": 1.7660714909434319,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7705357484519482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1380
    },
    {
      "completion_length": 234.9107265472412,
      "epoch": 0.23169453874848064,
      "grad_norm": 0.15438550905304013,
      "kl": 0.010261058807373047,
      "learning_rate": 1.706172839506173e-07,
      "loss": 0.0,
      "reward": 1.7000000476837158,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7089286036789417,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1382
    },
    {
      "completion_length": 243.08483505249023,
      "epoch": 0.23202984198834822,
      "grad_norm": 0.26681977267282586,
      "kl": 0.017525196075439453,
      "learning_rate": 1.708641975308642e-07,
      "loss": 0.0,
      "reward": 1.7392857894301414,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857633531094,
      "rewards/format_reward_func": 1.0,
      "step": 1384
    },
    {
      "completion_length": 235.85715103149414,
      "epoch": 0.23236514522821577,
      "grad_norm": 0.2000816579266356,
      "kl": 0.01715850830078125,
      "learning_rate": 1.711111111111111e-07,
      "loss": 0.0,
      "reward": 1.7321429327130318,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321428991854191,
      "rewards/format_reward_func": 1.0,
      "step": 1386
    },
    {
      "completion_length": 242.9017972946167,
      "epoch": 0.23270044846808333,
      "grad_norm": 0.22182251115978738,
      "kl": 0.006072044372558594,
      "learning_rate": 1.71358024691358e-07,
      "loss": 0.0,
      "reward": 1.7125000655651093,
      "reward_std": 0.09343911055475473,
      "rewards/equation_reward_func": 0.7258928790688515,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1388
    },
    {
      "completion_length": 237.6830472946167,
      "epoch": 0.23303575170795088,
      "grad_norm": 0.14505985813023767,
      "kl": 0.006579399108886719,
      "learning_rate": 1.7160493827160495e-07,
      "loss": 0.0,
      "reward": 1.830357201397419,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8348214477300644,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1390
    },
    {
      "completion_length": 221.95536518096924,
      "epoch": 0.23337105494781843,
      "grad_norm": 0.3887991399283552,
      "kl": 0.009667396545410156,
      "learning_rate": 1.7185185185185183e-07,
      "loss": 0.0,
      "reward": 1.7843750938773155,
      "reward_std": 0.06250318652018905,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 1392
    },
    {
      "completion_length": 239.72322750091553,
      "epoch": 0.23370635818768598,
      "grad_norm": 0.31497940932668206,
      "kl": 0.00586700439453125,
      "learning_rate": 1.7209876543209875e-07,
      "loss": 0.0,
      "reward": 1.778571493923664,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7785714417695999,
      "rewards/format_reward_func": 1.0,
      "step": 1394
    },
    {
      "completion_length": 237.57143783569336,
      "epoch": 0.23404166142755353,
      "grad_norm": 0.1722442961054919,
      "kl": 0.01059722900390625,
      "learning_rate": 1.7234567901234566e-07,
      "loss": 0.0,
      "reward": 1.7535715028643608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 1396
    },
    {
      "completion_length": 241.54018878936768,
      "epoch": 0.2343769646674211,
      "grad_norm": 0.23289390607293414,
      "kl": 0.016029834747314453,
      "learning_rate": 1.725925925925926e-07,
      "loss": 0.0,
      "reward": 1.69821435213089,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7116071581840515,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1398
    },
    {
      "completion_length": 240.33036613464355,
      "epoch": 0.23471226790728866,
      "grad_norm": 0.3716827959785001,
      "kl": 0.014810562133789062,
      "learning_rate": 1.728395061728395e-07,
      "loss": 0.0,
      "reward": 1.6892858073115349,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.6982143186032772,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1400
    },
    {
      "completion_length": 235.44197463989258,
      "epoch": 0.23504757114715621,
      "grad_norm": 0.15371024300134672,
      "kl": 0.017034530639648438,
      "learning_rate": 1.730864197530864e-07,
      "loss": 0.0,
      "reward": 1.7178572118282318,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7178571820259094,
      "rewards/format_reward_func": 1.0,
      "step": 1402
    },
    {
      "completion_length": 228.0089406967163,
      "epoch": 0.23538287438702377,
      "grad_norm": 0.280897369010513,
      "kl": 0.011105060577392578,
      "learning_rate": 1.7333333333333332e-07,
      "loss": 0.0,
      "reward": 1.7964286655187607,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.796428594738245,
      "rewards/format_reward_func": 1.0,
      "step": 1404
    },
    {
      "completion_length": 242.0759048461914,
      "epoch": 0.23571817762689132,
      "grad_norm": 0.4060917647274439,
      "kl": 0.014769554138183594,
      "learning_rate": 1.7358024691358027e-07,
      "loss": 0.0,
      "reward": 1.6803572252392769,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.684821454808116,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1406
    },
    {
      "completion_length": 234.25000953674316,
      "epoch": 0.23605348086675887,
      "grad_norm": 0.29418746031538556,
      "kl": 0.007956981658935547,
      "learning_rate": 1.7382716049382715e-07,
      "loss": 0.0,
      "reward": 1.6821429431438446,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.6821429021656513,
      "rewards/format_reward_func": 1.0,
      "step": 1408
    },
    {
      "completion_length": 238.62501049041748,
      "epoch": 0.23638878410662642,
      "grad_norm": 0.5412517362596901,
      "kl": 0.016735076904296875,
      "learning_rate": 1.7407407407407407e-07,
      "loss": 0.0,
      "reward": 1.7750000432133675,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 1410
    },
    {
      "completion_length": 244.31697273254395,
      "epoch": 0.23672408734649397,
      "grad_norm": 0.2775285412382821,
      "kl": 0.013233184814453125,
      "learning_rate": 1.7432098765432098e-07,
      "loss": 0.0,
      "reward": 1.751785784959793,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7562500387430191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1412
    },
    {
      "completion_length": 230.88393878936768,
      "epoch": 0.23705939058636155,
      "grad_norm": 0.30478659922904655,
      "kl": 0.005757331848144531,
      "learning_rate": 1.7456790123456787e-07,
      "loss": 0.0,
      "reward": 1.8285714760422707,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8285714387893677,
      "rewards/format_reward_func": 1.0,
      "step": 1414
    },
    {
      "completion_length": 230.1384038925171,
      "epoch": 0.2373946938262291,
      "grad_norm": 0.159625589618849,
      "kl": 0.018886566162109375,
      "learning_rate": 1.748148148148148e-07,
      "loss": 0.0,
      "reward": 1.7750000730156898,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 1416
    },
    {
      "completion_length": 233.52233409881592,
      "epoch": 0.23772999706609665,
      "grad_norm": 0.3840361324210306,
      "kl": 0.018432140350341797,
      "learning_rate": 1.7506172839506173e-07,
      "loss": 0.0,
      "reward": 1.7607143372297287,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7696428969502449,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1418
    },
    {
      "completion_length": 233.1428689956665,
      "epoch": 0.2380653003059642,
      "grad_norm": 0.31642404943545666,
      "kl": 0.019349098205566406,
      "learning_rate": 1.7530864197530864e-07,
      "loss": 0.0,
      "reward": 1.7107143551111221,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7107143178582191,
      "rewards/format_reward_func": 1.0,
      "step": 1420
    },
    {
      "completion_length": 245.95983219146729,
      "epoch": 0.23840060354583176,
      "grad_norm": 0.1705597345975343,
      "kl": 0.008008956909179688,
      "learning_rate": 1.7555555555555553e-07,
      "loss": 0.0,
      "reward": 1.76071435213089,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143223285675,
      "rewards/format_reward_func": 1.0,
      "step": 1422
    },
    {
      "completion_length": 232.80358123779297,
      "epoch": 0.2387359067856993,
      "grad_norm": 0.6085086681901897,
      "kl": 0.05005693435668945,
      "learning_rate": 1.7580246913580247e-07,
      "loss": 0.0001,
      "reward": 1.7196429297327995,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.724107176065445,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1424
    },
    {
      "completion_length": 237.4107255935669,
      "epoch": 0.23907121002556686,
      "grad_norm": 0.22994413881569736,
      "kl": 0.02605152130126953,
      "learning_rate": 1.7604938271604939e-07,
      "loss": 0.0,
      "reward": 1.7214286550879478,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.721428606659174,
      "rewards/format_reward_func": 1.0,
      "step": 1426
    },
    {
      "completion_length": 241.17411994934082,
      "epoch": 0.23940651326543444,
      "grad_norm": 0.4663555822879698,
      "kl": 0.04881477355957031,
      "learning_rate": 1.7629629629629627e-07,
      "loss": 0.0,
      "reward": 1.6642858311533928,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.6732143200933933,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1428
    },
    {
      "completion_length": 229.7678680419922,
      "epoch": 0.239741816505302,
      "grad_norm": 0.08922893400986967,
      "kl": 0.012132644653320312,
      "learning_rate": 1.765432098765432e-07,
      "loss": 0.0,
      "reward": 1.7000000923871994,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7000000365078449,
      "rewards/format_reward_func": 1.0,
      "step": 1430
    },
    {
      "completion_length": 235.18304443359375,
      "epoch": 0.24007711974516954,
      "grad_norm": 0.2590381600205453,
      "kl": 0.028981685638427734,
      "learning_rate": 1.7679012345679013e-07,
      "loss": 0.0,
      "reward": 1.7321429401636124,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7410714440047741,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1432
    },
    {
      "completion_length": 235.06697368621826,
      "epoch": 0.2404124229850371,
      "grad_norm": 0.21589017259195808,
      "kl": 0.09281253814697266,
      "learning_rate": 1.7703703703703705e-07,
      "loss": 0.0001,
      "reward": 1.7142857909202576,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7142857536673546,
      "rewards/format_reward_func": 1.0,
      "step": 1434
    },
    {
      "completion_length": 235.43304634094238,
      "epoch": 0.24074772622490465,
      "grad_norm": 0.11790548457588271,
      "kl": 0.08521080017089844,
      "learning_rate": 1.7728395061728393e-07,
      "loss": 0.0001,
      "reward": 1.7964286357164383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 1436
    },
    {
      "completion_length": 234.1696538925171,
      "epoch": 0.2410830294647722,
      "grad_norm": 0.23563333190513872,
      "kl": 0.02545928955078125,
      "learning_rate": 1.7753086419753085e-07,
      "loss": 0.0,
      "reward": 1.7071429416537285,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7071428894996643,
      "rewards/format_reward_func": 1.0,
      "step": 1438
    },
    {
      "completion_length": 230.19197463989258,
      "epoch": 0.24141833270463975,
      "grad_norm": 0.2772527679123808,
      "kl": 0.06699180603027344,
      "learning_rate": 1.7777777777777776e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 1440
    },
    {
      "completion_length": 233.6294755935669,
      "epoch": 0.2417536359445073,
      "grad_norm": 0.10134875392539826,
      "kl": 0.036830902099609375,
      "learning_rate": 1.7802469135802468e-07,
      "loss": 0.0,
      "reward": 1.725000061094761,
      "reward_std": 0.005050762556493282,
      "rewards/equation_reward_func": 0.7250000275671482,
      "rewards/format_reward_func": 1.0,
      "step": 1442
    },
    {
      "completion_length": 236.0178680419922,
      "epoch": 0.24208893918437488,
      "grad_norm": 0.7041311897794371,
      "kl": 0.15368270874023438,
      "learning_rate": 1.782716049382716e-07,
      "loss": 0.0002,
      "reward": 1.7839286252856255,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7883928865194321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1444
    },
    {
      "completion_length": 231.64733219146729,
      "epoch": 0.24242424242424243,
      "grad_norm": 0.26936496812526695,
      "kl": 0.08620834350585938,
      "learning_rate": 1.785185185185185e-07,
      "loss": 0.0001,
      "reward": 1.7125000804662704,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7169643267989159,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1446
    },
    {
      "completion_length": 234.21875858306885,
      "epoch": 0.24275954566410998,
      "grad_norm": 0.29822283393676635,
      "kl": 0.137451171875,
      "learning_rate": 1.7876543209876542e-07,
      "loss": 0.0001,
      "reward": 1.721428632736206,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7214286271482706,
      "rewards/format_reward_func": 1.0,
      "step": 1448
    },
    {
      "completion_length": 233.08483123779297,
      "epoch": 0.24309484890397753,
      "grad_norm": 0.13332326691597654,
      "kl": 0.16257810592651367,
      "learning_rate": 1.7901234567901234e-07,
      "loss": 0.0002,
      "reward": 1.7089286521077156,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7133928947150707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1450
    },
    {
      "completion_length": 234.89733219146729,
      "epoch": 0.24343015214384509,
      "grad_norm": 0.17016165008545617,
      "kl": 0.09032154083251953,
      "learning_rate": 1.7925925925925925e-07,
      "loss": 0.0001,
      "reward": 1.762500062584877,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643089175224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1452
    },
    {
      "completion_length": 237.5848331451416,
      "epoch": 0.24376545538371264,
      "grad_norm": 0.21223775500209358,
      "kl": 0.168792724609375,
      "learning_rate": 1.7950617283950617e-07,
      "loss": 0.0002,
      "reward": 1.7071429565548897,
      "reward_std": 0.03030457627028227,
      "rewards/equation_reward_func": 0.716071454808116,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1454
    },
    {
      "completion_length": 236.383939743042,
      "epoch": 0.2441007586235802,
      "grad_norm": 0.22677331825081817,
      "kl": 0.09221458435058594,
      "learning_rate": 1.7975308641975308e-07,
      "loss": 0.0001,
      "reward": 1.734375074505806,
      "reward_std": 0.05240166210569441,
      "rewards/equation_reward_func": 0.735714316368103,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 1456
    },
    {
      "completion_length": 236.5357265472412,
      "epoch": 0.24443606186344777,
      "grad_norm": 0.3695446121649956,
      "kl": 0.1760234832763672,
      "learning_rate": 1.8e-07,
      "loss": 0.0002,
      "reward": 1.7285715341567993,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7375000342726707,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1458
    },
    {
      "completion_length": 235.0535831451416,
      "epoch": 0.24477136510331532,
      "grad_norm": 0.19789398802687613,
      "kl": 0.11786079406738281,
      "learning_rate": 1.802469135802469e-07,
      "loss": 0.0001,
      "reward": 1.751785784959793,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.7651786003261805,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1460
    },
    {
      "completion_length": 231.3928680419922,
      "epoch": 0.24510666834318287,
      "grad_norm": 0.49119529018607455,
      "kl": 0.14634132385253906,
      "learning_rate": 1.8049382716049383e-07,
      "loss": 0.0001,
      "reward": 1.7946429029107094,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7991071715950966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1462
    },
    {
      "completion_length": 237.61608409881592,
      "epoch": 0.24544197158305042,
      "grad_norm": 0.34020786263175756,
      "kl": 0.2747945785522461,
      "learning_rate": 1.8074074074074072e-07,
      "loss": 0.0003,
      "reward": 1.778571479022503,
      "reward_std": 0.1010152529925108,
      "rewards/equation_reward_func": 0.7875000312924385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1464
    },
    {
      "completion_length": 238.42858219146729,
      "epoch": 0.24577727482291797,
      "grad_norm": 0.2181514729696241,
      "kl": 0.48163700103759766,
      "learning_rate": 1.8098765432098763e-07,
      "loss": 0.0005,
      "reward": 1.717857226729393,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.7267857454717159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1466
    },
    {
      "completion_length": 228.74108028411865,
      "epoch": 0.24611257806278553,
      "grad_norm": 0.22860592494644247,
      "kl": 0.4790782928466797,
      "learning_rate": 1.8123456790123457e-07,
      "loss": 0.0005,
      "reward": 1.7303571850061417,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7348214704543352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1468
    },
    {
      "completion_length": 234.67411994934082,
      "epoch": 0.24644788130265308,
      "grad_norm": 0.24797354188614557,
      "kl": 0.008512496948242188,
      "learning_rate": 1.8148148148148149e-07,
      "loss": 0.0,
      "reward": 1.7625000700354576,
      "reward_std": 0.09343910869210958,
      "rewards/equation_reward_func": 0.7669643200933933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1470
    },
    {
      "completion_length": 243.23215770721436,
      "epoch": 0.24678318454252063,
      "grad_norm": 0.20385499798706525,
      "kl": 0.18136978149414062,
      "learning_rate": 1.8172839506172837e-07,
      "loss": 0.0002,
      "reward": 1.7107143476605415,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7107143308967352,
      "rewards/format_reward_func": 1.0,
      "step": 1472
    },
    {
      "completion_length": 230.69197463989258,
      "epoch": 0.2471184877823882,
      "grad_norm": 0.4924804061728603,
      "kl": 0.16703414916992188,
      "learning_rate": 1.819753086419753e-07,
      "loss": 0.0002,
      "reward": 1.723214365541935,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7276786044239998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1474
    },
    {
      "completion_length": 230.3750123977661,
      "epoch": 0.24745379102225576,
      "grad_norm": 0.41683170206161424,
      "kl": 0.630040168762207,
      "learning_rate": 1.8222222222222223e-07,
      "loss": 0.0006,
      "reward": 1.669642947614193,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.6741071697324514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1476
    },
    {
      "completion_length": 235.13840103149414,
      "epoch": 0.2477890942621233,
      "grad_norm": 0.31188651409494367,
      "kl": 0.8284721374511719,
      "learning_rate": 1.8246913580246912e-07,
      "loss": 0.0008,
      "reward": 1.7071429193019867,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7160714641213417,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1478
    },
    {
      "completion_length": 234.34375858306885,
      "epoch": 0.24812439750199086,
      "grad_norm": 0.3133969512879245,
      "kl": 0.012765884399414062,
      "learning_rate": 1.8271604938271603e-07,
      "loss": 0.0,
      "reward": 1.7303572222590446,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.734821455553174,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1480
    },
    {
      "completion_length": 236.3259038925171,
      "epoch": 0.2484597007418584,
      "grad_norm": 0.20803832575885528,
      "kl": 0.3232555389404297,
      "learning_rate": 1.8296296296296295e-07,
      "loss": 0.0003,
      "reward": 1.7571428939700127,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 1482
    },
    {
      "completion_length": 241.29465293884277,
      "epoch": 0.24879500398172597,
      "grad_norm": 0.32787663426007396,
      "kl": 0.4119834899902344,
      "learning_rate": 1.832098765432099e-07,
      "loss": 0.0004,
      "reward": 1.7107143849134445,
      "reward_std": 0.07576144114136696,
      "rewards/equation_reward_func": 0.719642885029316,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1484
    },
    {
      "completion_length": 234.43304538726807,
      "epoch": 0.24913030722159352,
      "grad_norm": 0.2691108188962023,
      "kl": 0.06020641326904297,
      "learning_rate": 1.8345679012345678e-07,
      "loss": 0.0001,
      "reward": 1.7767857685685158,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7812500260770321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1486
    },
    {
      "completion_length": 235.89286994934082,
      "epoch": 0.2494656104614611,
      "grad_norm": 0.34431069891908556,
      "kl": 0.8705596923828125,
      "learning_rate": 1.837037037037037e-07,
      "loss": 0.0009,
      "reward": 1.775000050663948,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000301748514,
      "rewards/format_reward_func": 1.0,
      "step": 1488
    },
    {
      "completion_length": 227.37500953674316,
      "epoch": 0.24980091370132865,
      "grad_norm": 0.6695050569787687,
      "kl": 0.19437503814697266,
      "learning_rate": 1.839506172839506e-07,
      "loss": 0.0002,
      "reward": 1.7946429252624512,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7991071715950966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1490
    },
    {
      "completion_length": 228.99108123779297,
      "epoch": 0.2501362169411962,
      "grad_norm": 0.30169039119381347,
      "kl": 0.5019960403442383,
      "learning_rate": 1.841975308641975e-07,
      "loss": 0.0005,
      "reward": 1.744642935693264,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7491071671247482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1492
    },
    {
      "completion_length": 225.2142972946167,
      "epoch": 0.2504715201810637,
      "grad_norm": 0.14293763277183175,
      "kl": 0.21941280364990234,
      "learning_rate": 1.8444444444444444e-07,
      "loss": 0.0002,
      "reward": 1.7214286550879478,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7214286029338837,
      "rewards/format_reward_func": 1.0,
      "step": 1494
    },
    {
      "completion_length": 242.78126049041748,
      "epoch": 0.2508068234209313,
      "grad_norm": 0.19395245164727634,
      "kl": 0.25534629821777344,
      "learning_rate": 1.8469135802469135e-07,
      "loss": 0.0003,
      "reward": 1.7732143551111221,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.777678593993187,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1496
    },
    {
      "completion_length": 225.95536708831787,
      "epoch": 0.2511421266607989,
      "grad_norm": 0.21436022346581826,
      "kl": 0.3440971374511719,
      "learning_rate": 1.8493827160493827e-07,
      "loss": 0.0003,
      "reward": 1.8075893446803093,
      "reward_std": 0.0460882093757391,
      "rewards/equation_reward_func": 0.813392873853445,
      "rewards/format_reward_func": 0.994196429848671,
      "step": 1498
    },
    {
      "completion_length": 233.38393878936768,
      "epoch": 0.2514774299006664,
      "grad_norm": 0.3264524194551336,
      "kl": 0.22805309295654297,
      "learning_rate": 1.8518518518518516e-07,
      "loss": 0.0002,
      "reward": 1.6928572431206703,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7017857395112514,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1500
    },
    {
      "completion_length": 229.2991189956665,
      "epoch": 0.251812733140534,
      "grad_norm": 0.20478533160273893,
      "kl": 0.14220809936523438,
      "learning_rate": 1.854320987654321e-07,
      "loss": 0.0001,
      "reward": 1.7910714745521545,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7955357432365417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1502
    },
    {
      "completion_length": 237.68750762939453,
      "epoch": 0.2521480363804015,
      "grad_norm": 0.22934017891844902,
      "kl": 0.11223888397216797,
      "learning_rate": 1.85679012345679e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.810714315623045,
      "rewards/format_reward_func": 1.0,
      "step": 1504
    },
    {
      "completion_length": 235.20536708831787,
      "epoch": 0.2524833396202691,
      "grad_norm": 0.4432363920330559,
      "kl": 0.08673858642578125,
      "learning_rate": 1.8592592592592593e-07,
      "loss": 0.0001,
      "reward": 1.7696429342031479,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7741071581840515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1506
    },
    {
      "completion_length": 230.18304443359375,
      "epoch": 0.2528186428601366,
      "grad_norm": 0.21735637485812978,
      "kl": 0.16140174865722656,
      "learning_rate": 1.8617283950617281e-07,
      "loss": 0.0002,
      "reward": 1.7714286297559738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 1508
    },
    {
      "completion_length": 232.8884038925171,
      "epoch": 0.2531539461000042,
      "grad_norm": 0.2751427765130532,
      "kl": 0.03536796569824219,
      "learning_rate": 1.8641975308641976e-07,
      "loss": 0.0,
      "reward": 1.8071428909897804,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8071428760886192,
      "rewards/format_reward_func": 1.0,
      "step": 1510
    },
    {
      "completion_length": 241.68304634094238,
      "epoch": 0.25348924933987177,
      "grad_norm": 0.3222525911957361,
      "kl": 0.14169883728027344,
      "learning_rate": 1.8666666666666667e-07,
      "loss": 0.0001,
      "reward": 1.6553572192788124,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.6687500290572643,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1512
    },
    {
      "completion_length": 235.7053689956665,
      "epoch": 0.2538245525797393,
      "grad_norm": 0.6563885167540162,
      "kl": 0.09424400329589844,
      "learning_rate": 1.8691358024691356e-07,
      "loss": 0.0001,
      "reward": 1.7357143685221672,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7357143349945545,
      "rewards/format_reward_func": 1.0,
      "step": 1514
    },
    {
      "completion_length": 233.75447368621826,
      "epoch": 0.2541598558196069,
      "grad_norm": 0.5746414100535419,
      "kl": 0.06946945190429688,
      "learning_rate": 1.8716049382716047e-07,
      "loss": 0.0001,
      "reward": 1.7678572311997414,
      "reward_std": 0.07576144114136696,
      "rewards/equation_reward_func": 0.7767857387661934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1516
    },
    {
      "completion_length": 223.63393783569336,
      "epoch": 0.2544951590594744,
      "grad_norm": 0.1957711124441575,
      "kl": 0.10572624206542969,
      "learning_rate": 1.8740740740740742e-07,
      "loss": 0.0001,
      "reward": 1.7928571924567223,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.8017857410013676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1518
    },
    {
      "completion_length": 236.38393783569336,
      "epoch": 0.254830462299342,
      "grad_norm": 0.30273470214807835,
      "kl": 0.15691757202148438,
      "learning_rate": 1.8765432098765433e-07,
      "loss": 0.0002,
      "reward": 1.7214286476373672,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.721428606659174,
      "rewards/format_reward_func": 1.0,
      "step": 1520
    },
    {
      "completion_length": 226.78572463989258,
      "epoch": 0.2551657655392095,
      "grad_norm": 0.3477983467378772,
      "kl": 0.047286033630371094,
      "learning_rate": 1.8790123456790122e-07,
      "loss": 0.0,
      "reward": 1.7339286357164383,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.73839289881289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1522
    },
    {
      "completion_length": 225.79465198516846,
      "epoch": 0.2555010687790771,
      "grad_norm": 0.21323643527682343,
      "kl": 0.05094718933105469,
      "learning_rate": 1.8814814814814813e-07,
      "loss": 0.0001,
      "reward": 1.7392858192324638,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857372760773,
      "rewards/format_reward_func": 1.0,
      "step": 1524
    },
    {
      "completion_length": 237.12947845458984,
      "epoch": 0.25583637201894466,
      "grad_norm": 0.2633799298401754,
      "kl": 0.10648727416992188,
      "learning_rate": 1.8839506172839505e-07,
      "loss": 0.0001,
      "reward": 1.6892857775092125,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6892857439815998,
      "rewards/format_reward_func": 1.0,
      "step": 1526
    },
    {
      "completion_length": 233.39286613464355,
      "epoch": 0.2561716752588122,
      "grad_norm": 0.3611704280068289,
      "kl": 0.039473533630371094,
      "learning_rate": 1.8864197530864196e-07,
      "loss": 0.0,
      "reward": 1.7428572177886963,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571842610836,
      "rewards/format_reward_func": 1.0,
      "step": 1528
    },
    {
      "completion_length": 238.5580472946167,
      "epoch": 0.25650697849867976,
      "grad_norm": 0.32581177303136893,
      "kl": 0.12656688690185547,
      "learning_rate": 1.8888888888888888e-07,
      "loss": 0.0001,
      "reward": 1.7446429505944252,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7491071633994579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1530
    },
    {
      "completion_length": 227.6741180419922,
      "epoch": 0.2568422817385473,
      "grad_norm": 0.3034539012000064,
      "kl": 0.046042442321777344,
      "learning_rate": 1.891358024691358e-07,
      "loss": 0.0,
      "reward": 1.7821428999304771,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 1532
    },
    {
      "completion_length": 239.6250114440918,
      "epoch": 0.25717758497841486,
      "grad_norm": 0.48368909019447376,
      "kl": 0.06012153625488281,
      "learning_rate": 1.893827160493827e-07,
      "loss": 0.0001,
      "reward": 1.737500086426735,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7419643141329288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1534
    },
    {
      "completion_length": 237.06251049041748,
      "epoch": 0.2575128882182824,
      "grad_norm": 0.26842429522669164,
      "kl": 0.06570243835449219,
      "learning_rate": 1.8962962962962962e-07,
      "loss": 0.0001,
      "reward": 1.7267857789993286,
      "reward_std": 0.10354063380509615,
      "rewards/equation_reward_func": 0.7401786055415869,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1536
    },
    {
      "completion_length": 228.790189743042,
      "epoch": 0.25784819145814997,
      "grad_norm": 0.4018896461427612,
      "kl": 0.035400390625,
      "learning_rate": 1.8987654320987654e-07,
      "loss": 0.0,
      "reward": 1.737500086426735,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7419643178582191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1538
    },
    {
      "completion_length": 229.99554634094238,
      "epoch": 0.25818349469801755,
      "grad_norm": 0.2506558632879474,
      "kl": 0.0076465606689453125,
      "learning_rate": 1.9012345679012345e-07,
      "loss": 0.0,
      "reward": 1.805357187986374,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8187500257045031,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1540
    },
    {
      "completion_length": 230.18304347991943,
      "epoch": 0.25851879793788507,
      "grad_norm": 0.29934272366305864,
      "kl": 0.06483650207519531,
      "learning_rate": 1.9037037037037037e-07,
      "loss": 0.0001,
      "reward": 1.7214286625385284,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7303571719676256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1542
    },
    {
      "completion_length": 224.05358219146729,
      "epoch": 0.25885410117775265,
      "grad_norm": 0.3433338629396455,
      "kl": 0.02986907958984375,
      "learning_rate": 1.9061728395061728e-07,
      "loss": 0.0,
      "reward": 1.7750000730156898,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 1544
    },
    {
      "completion_length": 239.09375953674316,
      "epoch": 0.2591894044176202,
      "grad_norm": 0.31108987070476973,
      "kl": 0.058963775634765625,
      "learning_rate": 1.908641975308642e-07,
      "loss": 0.0001,
      "reward": 1.7410715073347092,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.745535746216774,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1546
    },
    {
      "completion_length": 228.30358123779297,
      "epoch": 0.25952470765748775,
      "grad_norm": 0.2713275566948536,
      "kl": 0.023523330688476562,
      "learning_rate": 1.911111111111111e-07,
      "loss": 0.0,
      "reward": 1.766071505844593,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7705357521772385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1548
    },
    {
      "completion_length": 224.51340293884277,
      "epoch": 0.2598600108973553,
      "grad_norm": 0.31686450102462,
      "kl": 0.010550498962402344,
      "learning_rate": 1.91358024691358e-07,
      "loss": 0.0,
      "reward": 1.7750000581145287,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7839285973459482,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1550
    },
    {
      "completion_length": 225.01786708831787,
      "epoch": 0.26019531413722286,
      "grad_norm": 0.515107201500274,
      "kl": 0.0798177719116211,
      "learning_rate": 1.9160493827160491e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7875000238418579,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1552
    },
    {
      "completion_length": 228.12947463989258,
      "epoch": 0.2605306173770904,
      "grad_norm": 0.3542582192758546,
      "kl": 0.040058135986328125,
      "learning_rate": 1.9185185185185186e-07,
      "loss": 0.0,
      "reward": 1.7946429029107094,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7991071753203869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1554
    },
    {
      "completion_length": 236.19644165039062,
      "epoch": 0.26086592061695796,
      "grad_norm": 0.31969628871868205,
      "kl": 0.06807518005371094,
      "learning_rate": 1.9209876543209877e-07,
      "loss": 0.0001,
      "reward": 1.7142858132719994,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7142857536673546,
      "rewards/format_reward_func": 1.0,
      "step": 1556
    },
    {
      "completion_length": 240.19643878936768,
      "epoch": 0.26120122385682554,
      "grad_norm": 0.41985963092768086,
      "kl": 0.11923789978027344,
      "learning_rate": 1.9234567901234566e-07,
      "loss": 0.0001,
      "reward": 1.6767858043313026,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.689285745844245,
      "rewards/format_reward_func": 0.9875000044703484,
      "step": 1558
    },
    {
      "completion_length": 238.20537185668945,
      "epoch": 0.26153652709669306,
      "grad_norm": 0.26763216571838205,
      "kl": 0.036202430725097656,
      "learning_rate": 1.9259259259259257e-07,
      "loss": 0.0,
      "reward": 1.8017857670783997,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.806250024586916,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1560
    },
    {
      "completion_length": 241.16072368621826,
      "epoch": 0.26187183033656064,
      "grad_norm": 0.7795636460003993,
      "kl": 0.08544540405273438,
      "learning_rate": 1.9283950617283951e-07,
      "loss": 0.0001,
      "reward": 1.7285714894533157,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7464286163449287,
      "rewards/format_reward_func": 0.9821428619325161,
      "step": 1562
    },
    {
      "completion_length": 226.32590198516846,
      "epoch": 0.26220713357642816,
      "grad_norm": 0.26431885626858,
      "kl": 0.031281471252441406,
      "learning_rate": 1.930864197530864e-07,
      "loss": 0.0,
      "reward": 1.7482143566012383,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7526785954833031,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1564
    },
    {
      "completion_length": 230.01786708831787,
      "epoch": 0.26254243681629574,
      "grad_norm": 0.25896091143440436,
      "kl": 0.19019317626953125,
      "learning_rate": 1.9333333333333332e-07,
      "loss": 0.0002,
      "reward": 1.778571479022503,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7875000275671482,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1566
    },
    {
      "completion_length": 228.42858219146729,
      "epoch": 0.26287774005616327,
      "grad_norm": 0.22123026075224217,
      "kl": 0.11663246154785156,
      "learning_rate": 1.9358024691358023e-07,
      "loss": 0.0001,
      "reward": 1.741071492433548,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357499420643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1568
    },
    {
      "completion_length": 223.83929634094238,
      "epoch": 0.26321304329603085,
      "grad_norm": 0.34891901610585063,
      "kl": 0.0828256607055664,
      "learning_rate": 1.9382716049382717e-07,
      "loss": 0.0001,
      "reward": 1.7392857745289803,
      "reward_std": 0.07576143741607666,
      "rewards/equation_reward_func": 0.7482143193483353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1570
    },
    {
      "completion_length": 234.86607933044434,
      "epoch": 0.2635483465358984,
      "grad_norm": 0.7646794550333904,
      "kl": 0.0537567138671875,
      "learning_rate": 1.9407407407407406e-07,
      "loss": 0.0001,
      "reward": 1.7732143476605415,
      "reward_std": 0.0883883461356163,
      "rewards/equation_reward_func": 0.777678582817316,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1572
    },
    {
      "completion_length": 231.30804538726807,
      "epoch": 0.26388364977576595,
      "grad_norm": 0.20786625219993726,
      "kl": 0.06808948516845703,
      "learning_rate": 1.9432098765432098e-07,
      "loss": 0.0001,
      "reward": 1.7410714849829674,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7446428909897804,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 1574
    },
    {
      "completion_length": 234.66072463989258,
      "epoch": 0.26421895301563353,
      "grad_norm": 0.17919898669534326,
      "kl": 0.19156932830810547,
      "learning_rate": 1.945679012345679e-07,
      "loss": 0.0002,
      "reward": 1.8160715103149414,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.820535734295845,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1576
    },
    {
      "completion_length": 234.99107933044434,
      "epoch": 0.26455425625550105,
      "grad_norm": 0.25375846103409333,
      "kl": 0.1639862060546875,
      "learning_rate": 1.9481481481481478e-07,
      "loss": 0.0002,
      "reward": 1.7392857819795609,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7392857447266579,
      "rewards/format_reward_func": 1.0,
      "step": 1578
    },
    {
      "completion_length": 232.89733028411865,
      "epoch": 0.26488955949536863,
      "grad_norm": 0.20598495484765386,
      "kl": 0.32021331787109375,
      "learning_rate": 1.9506172839506172e-07,
      "loss": 0.0003,
      "reward": 1.7625000849366188,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7758928909897804,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1580
    },
    {
      "completion_length": 229.95090293884277,
      "epoch": 0.26522486273523616,
      "grad_norm": 0.23689001841514462,
      "kl": 0.028433799743652344,
      "learning_rate": 1.9530864197530864e-07,
      "loss": 0.0,
      "reward": 1.7750000581145287,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7839285992085934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1582
    },
    {
      "completion_length": 237.3973331451416,
      "epoch": 0.26556016597510373,
      "grad_norm": 0.30083193411151415,
      "kl": 0.0551910400390625,
      "learning_rate": 1.9555555555555555e-07,
      "loss": 0.0001,
      "reward": 1.7236607745289803,
      "reward_std": 0.0776554741896689,
      "rewards/equation_reward_func": 0.7250000350177288,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 1584
    },
    {
      "completion_length": 233.04018783569336,
      "epoch": 0.2658954692149713,
      "grad_norm": 0.28273465688517146,
      "kl": 0.12781333923339844,
      "learning_rate": 1.9580246913580244e-07,
      "loss": 0.0001,
      "reward": 1.7196429297327995,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7241071742027998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1586
    },
    {
      "completion_length": 236.07143878936768,
      "epoch": 0.26623077245483884,
      "grad_norm": 0.13755953595441434,
      "kl": 0.26348114013671875,
      "learning_rate": 1.9604938271604938e-07,
      "loss": 0.0003,
      "reward": 1.7428572252392769,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571674972773,
      "rewards/format_reward_func": 1.0,
      "step": 1588
    },
    {
      "completion_length": 231.6205472946167,
      "epoch": 0.2665660756947064,
      "grad_norm": 0.24383190621635564,
      "kl": 0.07479667663574219,
      "learning_rate": 1.962962962962963e-07,
      "loss": 0.0001,
      "reward": 1.741071492433548,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7455357536673546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1590
    },
    {
      "completion_length": 239.19197368621826,
      "epoch": 0.26690137893457394,
      "grad_norm": 0.2082010962256202,
      "kl": 0.21643352508544922,
      "learning_rate": 1.965432098765432e-07,
      "loss": 0.0002,
      "reward": 1.7642857655882835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 1592
    },
    {
      "completion_length": 220.26786613464355,
      "epoch": 0.2672366821744415,
      "grad_norm": 0.33452171117357393,
      "kl": 0.12536048889160156,
      "learning_rate": 1.967901234567901e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.7410714514553547,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1594
    },
    {
      "completion_length": 226.27233219146729,
      "epoch": 0.26757198541430904,
      "grad_norm": 0.3662231272162297,
      "kl": 0.12168693542480469,
      "learning_rate": 1.9703703703703704e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571626543999,
      "rewards/format_reward_func": 1.0,
      "step": 1596
    },
    {
      "completion_length": 223.64286708831787,
      "epoch": 0.2679072886541766,
      "grad_norm": 0.45793697052291993,
      "kl": 0.15146446228027344,
      "learning_rate": 1.9728395061728395e-07,
      "loss": 0.0002,
      "reward": 1.8000000268220901,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.8089286014437675,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1598
    },
    {
      "completion_length": 227.0178689956665,
      "epoch": 0.2682425918940442,
      "grad_norm": 0.5879741708402264,
      "kl": 0.11877250671386719,
      "learning_rate": 1.9753086419753084e-07,
      "loss": 0.0001,
      "reward": 1.7714286670088768,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 1600
    },
    {
      "completion_length": 225.34822463989258,
      "epoch": 0.2685778951339117,
      "grad_norm": 0.3172965825133875,
      "kl": 0.1242218017578125,
      "learning_rate": 1.9777777777777776e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 1602
    },
    {
      "completion_length": 225.34376049041748,
      "epoch": 0.2689131983737793,
      "grad_norm": 0.19786567305881758,
      "kl": 0.041957855224609375,
      "learning_rate": 1.980246913580247e-07,
      "loss": 0.0,
      "reward": 1.7821429297327995,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 1604
    },
    {
      "completion_length": 231.17411613464355,
      "epoch": 0.26924850161364683,
      "grad_norm": 0.199203557943275,
      "kl": 0.2041006088256836,
      "learning_rate": 1.9827160493827161e-07,
      "loss": 0.0002,
      "reward": 1.807142935693264,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428835391998,
      "rewards/format_reward_func": 1.0,
      "step": 1606
    },
    {
      "completion_length": 244.96429824829102,
      "epoch": 0.2695838048535144,
      "grad_norm": 0.5953925603348911,
      "kl": 0.39769744873046875,
      "learning_rate": 1.985185185185185e-07,
      "loss": 0.0004,
      "reward": 1.721428632736206,
      "reward_std": 0.0909137288108468,
      "rewards/equation_reward_func": 0.7392857391387224,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 1608
    },
    {
      "completion_length": 227.01340198516846,
      "epoch": 0.26991910809338193,
      "grad_norm": 0.6255436231056462,
      "kl": 0.01708221435546875,
      "learning_rate": 1.9876543209876542e-07,
      "loss": 0.0,
      "reward": 1.6892858073115349,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6892857495695353,
      "rewards/format_reward_func": 1.0,
      "step": 1610
    },
    {
      "completion_length": 227.75001049041748,
      "epoch": 0.2702544113332495,
      "grad_norm": 0.23378469972731863,
      "kl": 0.2529106140136719,
      "learning_rate": 1.9901234567901233e-07,
      "loss": 0.0003,
      "reward": 1.7500000670552254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 1612
    },
    {
      "completion_length": 241.0134048461914,
      "epoch": 0.27058971457311704,
      "grad_norm": 0.2597360936259886,
      "kl": 0.17315101623535156,
      "learning_rate": 1.9925925925925925e-07,
      "loss": 0.0002,
      "reward": 1.717857226729393,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7178571783006191,
      "rewards/format_reward_func": 1.0,
      "step": 1614
    },
    {
      "completion_length": 231.7946548461914,
      "epoch": 0.2709250178129846,
      "grad_norm": 0.46129278988052735,
      "kl": 0.10466766357421875,
      "learning_rate": 1.9950617283950616e-07,
      "loss": 0.0001,
      "reward": 1.7446428909897804,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7491071820259094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1616
    },
    {
      "completion_length": 224.696439743042,
      "epoch": 0.2712603210528522,
      "grad_norm": 0.4255237430650551,
      "kl": 0.17496871948242188,
      "learning_rate": 1.9975308641975308e-07,
      "loss": 0.0002,
      "reward": 1.7714286521077156,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 1618
    },
    {
      "completion_length": 229.60715198516846,
      "epoch": 0.2715956242927197,
      "grad_norm": 0.189806473011921,
      "kl": 0.1701374053955078,
      "learning_rate": 2e-07,
      "loss": 0.0002,
      "reward": 1.7821429297327995,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428906172514,
      "rewards/format_reward_func": 1.0,
      "step": 1620
    },
    {
      "completion_length": 223.0089406967163,
      "epoch": 0.2719309275325873,
      "grad_norm": 0.2910744729086953,
      "kl": 0.29755210876464844,
      "learning_rate": 2.002469135802469e-07,
      "loss": 0.0003,
      "reward": 1.6928572282195091,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7017857581377029,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1622
    },
    {
      "completion_length": 221.98215007781982,
      "epoch": 0.2722662307724548,
      "grad_norm": 0.22473947784807705,
      "kl": 0.17630863189697266,
      "learning_rate": 2.0049382716049382e-07,
      "loss": 0.0002,
      "reward": 1.7428571805357933,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7428571842610836,
      "rewards/format_reward_func": 1.0,
      "step": 1624
    },
    {
      "completion_length": 244.11608219146729,
      "epoch": 0.2726015340123224,
      "grad_norm": 0.40284671702968305,
      "kl": 0.11990642547607422,
      "learning_rate": 2.0074074074074074e-07,
      "loss": 0.0001,
      "reward": 1.7321429178118706,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7410714589059353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1626
    },
    {
      "completion_length": 227.2187614440918,
      "epoch": 0.2729368372521899,
      "grad_norm": 0.26856909090880804,
      "kl": 0.08802032470703125,
      "learning_rate": 2.0098765432098762e-07,
      "loss": 0.0001,
      "reward": 1.6857143640518188,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.6857143305242062,
      "rewards/format_reward_func": 1.0,
      "step": 1628
    },
    {
      "completion_length": 220.85715103149414,
      "epoch": 0.2732721404920575,
      "grad_norm": 0.20613533704375775,
      "kl": 0.019829750061035156,
      "learning_rate": 2.0123456790123457e-07,
      "loss": 0.0,
      "reward": 1.7928572073578835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571626543999,
      "rewards/format_reward_func": 1.0,
      "step": 1630
    },
    {
      "completion_length": 227.7321548461914,
      "epoch": 0.2736074437319251,
      "grad_norm": 0.20721345610666672,
      "kl": 0.10821723937988281,
      "learning_rate": 2.0148148148148148e-07,
      "loss": 0.0001,
      "reward": 1.7071429267525673,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7160714566707611,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1632
    },
    {
      "completion_length": 224.10268783569336,
      "epoch": 0.2739427469717926,
      "grad_norm": 0.2862899956793819,
      "kl": 0.10094642639160156,
      "learning_rate": 2.017283950617284e-07,
      "loss": 0.0001,
      "reward": 1.7678572237491608,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 1634
    },
    {
      "completion_length": 221.59375953674316,
      "epoch": 0.2742780502116602,
      "grad_norm": 0.3526294409199377,
      "kl": 0.10612869262695312,
      "learning_rate": 2.0197530864197528e-07,
      "loss": 0.0001,
      "reward": 1.8129464909434319,
      "reward_std": 0.04230013629421592,
      "rewards/equation_reward_func": 0.8142857439815998,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 1636
    },
    {
      "completion_length": 240.34822368621826,
      "epoch": 0.2746133534515277,
      "grad_norm": 0.22284661939811562,
      "kl": 0.19942092895507812,
      "learning_rate": 2.022222222222222e-07,
      "loss": 0.0002,
      "reward": 1.7035714983940125,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7035714723169804,
      "rewards/format_reward_func": 1.0,
      "step": 1638
    },
    {
      "completion_length": 228.54465103149414,
      "epoch": 0.2749486566913953,
      "grad_norm": 0.23164365281361313,
      "kl": 0.035366058349609375,
      "learning_rate": 2.0246913580246914e-07,
      "loss": 0.0,
      "reward": 1.700000062584877,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7000000216066837,
      "rewards/format_reward_func": 1.0,
      "step": 1640
    },
    {
      "completion_length": 224.1071538925171,
      "epoch": 0.2752839599312628,
      "grad_norm": 0.3284594210277132,
      "kl": 0.09695816040039062,
      "learning_rate": 2.0271604938271605e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 1642
    },
    {
      "completion_length": 229.30804634094238,
      "epoch": 0.2756192631711304,
      "grad_norm": 0.383168344007737,
      "kl": 0.11481475830078125,
      "learning_rate": 2.0296296296296294e-07,
      "loss": 0.0001,
      "reward": 1.7125000804662704,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.7258928753435612,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1644
    },
    {
      "completion_length": 228.91518783569336,
      "epoch": 0.27595456641099797,
      "grad_norm": 0.4223823876902598,
      "kl": 0.1330738067626953,
      "learning_rate": 2.0320987654320986e-07,
      "loss": 0.0001,
      "reward": 1.7196429371833801,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.724107189103961,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1646
    },
    {
      "completion_length": 230.92858219146729,
      "epoch": 0.2762898696508655,
      "grad_norm": 0.19741134994974258,
      "kl": 0.24378204345703125,
      "learning_rate": 2.034567901234568e-07,
      "loss": 0.0002,
      "reward": 1.7535714730620384,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 1648
    },
    {
      "completion_length": 238.7812623977661,
      "epoch": 0.2766251728907331,
      "grad_norm": 0.23204192482524613,
      "kl": 0.17847061157226562,
      "learning_rate": 2.0370370370370369e-07,
      "loss": 0.0002,
      "reward": 1.7232143431901932,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7276786044239998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1650
    },
    {
      "completion_length": 227.9196538925171,
      "epoch": 0.2769604761306006,
      "grad_norm": 0.43032013537420893,
      "kl": 0.3296394348144531,
      "learning_rate": 2.039506172839506e-07,
      "loss": 0.0003,
      "reward": 1.7875000536441803,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7919643186032772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1652
    },
    {
      "completion_length": 237.1696538925171,
      "epoch": 0.2772957793704682,
      "grad_norm": 0.2077137808223099,
      "kl": 0.31763267517089844,
      "learning_rate": 2.0419753086419752e-07,
      "loss": 0.0003,
      "reward": 1.7214286550879478,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7214286103844643,
      "rewards/format_reward_func": 1.0,
      "step": 1654
    },
    {
      "completion_length": 231.7098331451416,
      "epoch": 0.2776310826103357,
      "grad_norm": 0.08385124065366012,
      "kl": 0.20266056060791016,
      "learning_rate": 2.0444444444444446e-07,
      "loss": 0.0002,
      "reward": 1.725000075995922,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7339286096394062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1656
    },
    {
      "completion_length": 238.44643878936768,
      "epoch": 0.2779663858502033,
      "grad_norm": 0.22718182123890435,
      "kl": 0.027462005615234375,
      "learning_rate": 2.0469135802469135e-07,
      "loss": 0.0,
      "reward": 1.7660714909434319,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7705357484519482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1658
    },
    {
      "completion_length": 229.26340198516846,
      "epoch": 0.27830168909007086,
      "grad_norm": 0.30742028326270204,
      "kl": 0.30859375,
      "learning_rate": 2.0493827160493826e-07,
      "loss": 0.0003,
      "reward": 1.7589286267757416,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7633928954601288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1660
    },
    {
      "completion_length": 229.0580472946167,
      "epoch": 0.2786369923299384,
      "grad_norm": 0.44218696016693715,
      "kl": 0.05316638946533203,
      "learning_rate": 2.0518518518518518e-07,
      "loss": 0.0001,
      "reward": 1.7598214894533157,
      "reward_std": 0.07702413015067577,
      "rewards/equation_reward_func": 0.7660714574158192,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 1662
    },
    {
      "completion_length": 230.37947463989258,
      "epoch": 0.27897229556980596,
      "grad_norm": 0.1850784449678431,
      "kl": 0.06500816345214844,
      "learning_rate": 2.0543209876543206e-07,
      "loss": 0.0001,
      "reward": 1.7750000730156898,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 1664
    },
    {
      "completion_length": 222.53126049041748,
      "epoch": 0.2793075988096735,
      "grad_norm": 0.279684266768164,
      "kl": 0.011688232421875,
      "learning_rate": 2.05679012345679e-07,
      "loss": 0.0,
      "reward": 1.766964353621006,
      "reward_std": 0.06692260596901178,
      "rewards/equation_reward_func": 0.7741071730852127,
      "rewards/format_reward_func": 0.9928571507334709,
      "step": 1666
    },
    {
      "completion_length": 223.3303689956665,
      "epoch": 0.27964290204954106,
      "grad_norm": 0.2295089233084164,
      "kl": 0.034732818603515625,
      "learning_rate": 2.0592592592592592e-07,
      "loss": 0.0,
      "reward": 1.7821429371833801,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 1668
    },
    {
      "completion_length": 224.62501049041748,
      "epoch": 0.2799782052894086,
      "grad_norm": 0.16455998588464119,
      "kl": 0.03418731689453125,
      "learning_rate": 2.0617283950617283e-07,
      "loss": 0.0,
      "reward": 1.7607143595814705,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 1670
    },
    {
      "completion_length": 228.66965103149414,
      "epoch": 0.28031350852927617,
      "grad_norm": 0.16323662515693607,
      "kl": 0.1743144989013672,
      "learning_rate": 2.0641975308641972e-07,
      "loss": 0.0002,
      "reward": 1.753571480512619,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 1672
    },
    {
      "completion_length": 231.9107265472412,
      "epoch": 0.28064881176914375,
      "grad_norm": 0.17971824743419645,
      "kl": 0.07829761505126953,
      "learning_rate": 2.0666666666666666e-07,
      "loss": 0.0001,
      "reward": 1.7553572207689285,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7598214540630579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1674
    },
    {
      "completion_length": 236.86162090301514,
      "epoch": 0.28098411500901127,
      "grad_norm": 0.18232481891962177,
      "kl": 0.1730976104736328,
      "learning_rate": 2.0691358024691358e-07,
      "loss": 0.0002,
      "reward": 1.755357213318348,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7687500342726707,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1676
    },
    {
      "completion_length": 233.86161613464355,
      "epoch": 0.28131941824887885,
      "grad_norm": 0.2875084708059203,
      "kl": 0.08128738403320312,
      "learning_rate": 2.0716049382716047e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 1678
    },
    {
      "completion_length": 230.69644260406494,
      "epoch": 0.2816547214887464,
      "grad_norm": 0.23955868610264447,
      "kl": 0.12451171875,
      "learning_rate": 2.0740740740740738e-07,
      "loss": 0.0001,
      "reward": 1.7160715088248253,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7205357421189547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1680
    },
    {
      "completion_length": 225.41965293884277,
      "epoch": 0.28199002472861395,
      "grad_norm": 0.24345243540818787,
      "kl": 0.27367496490478516,
      "learning_rate": 2.0765432098765432e-07,
      "loss": 0.0003,
      "reward": 1.776785783469677,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7812500223517418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1682
    },
    {
      "completion_length": 223.46875953674316,
      "epoch": 0.2823253279684815,
      "grad_norm": 0.25014435964613124,
      "kl": 0.030881881713867188,
      "learning_rate": 2.0790123456790124e-07,
      "loss": 0.0,
      "reward": 1.7571429163217545,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 1684
    },
    {
      "completion_length": 224.60268783569336,
      "epoch": 0.28266063120834906,
      "grad_norm": 0.2666289593787631,
      "kl": 0.03217506408691406,
      "learning_rate": 2.0814814814814813e-07,
      "loss": 0.0,
      "reward": 1.7571429163217545,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571429051458836,
      "rewards/format_reward_func": 1.0,
      "step": 1686
    },
    {
      "completion_length": 224.8214406967163,
      "epoch": 0.2829959344482166,
      "grad_norm": 0.2344435869553199,
      "kl": 0.014842987060546875,
      "learning_rate": 2.0839506172839504e-07,
      "loss": 0.0,
      "reward": 1.7553572207689285,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7598214633762836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1688
    },
    {
      "completion_length": 222.98661613464355,
      "epoch": 0.28333123768808416,
      "grad_norm": 0.2974750754107423,
      "kl": 0.10511493682861328,
      "learning_rate": 2.0864197530864198e-07,
      "loss": 0.0001,
      "reward": 1.7678572311997414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 1690
    },
    {
      "completion_length": 225.80804538726807,
      "epoch": 0.28366654092795174,
      "grad_norm": 0.40492073155295566,
      "kl": 0.19135475158691406,
      "learning_rate": 2.088888888888889e-07,
      "loss": 0.0002,
      "reward": 1.7625000551342964,
      "reward_std": 0.0833375845104456,
      "rewards/equation_reward_func": 0.7758928760886192,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1692
    },
    {
      "completion_length": 225.75000858306885,
      "epoch": 0.28400184416781926,
      "grad_norm": 0.21306977700182334,
      "kl": 0.2654247283935547,
      "learning_rate": 2.0913580246913579e-07,
      "loss": 0.0003,
      "reward": 1.6964286789298058,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.6964285988360643,
      "rewards/format_reward_func": 1.0,
      "step": 1694
    },
    {
      "completion_length": 224.08036708831787,
      "epoch": 0.28433714740768684,
      "grad_norm": 0.25962073008852654,
      "kl": 0.022909164428710938,
      "learning_rate": 2.093827160493827e-07,
      "loss": 0.0,
      "reward": 1.7464286163449287,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286163449287,
      "rewards/format_reward_func": 1.0,
      "step": 1696
    },
    {
      "completion_length": 220.73661708831787,
      "epoch": 0.28467245064755436,
      "grad_norm": 0.2998414003107778,
      "kl": 0.34929847717285156,
      "learning_rate": 2.0962962962962962e-07,
      "loss": 0.0003,
      "reward": 1.7428572326898575,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 1698
    },
    {
      "completion_length": 222.66965198516846,
      "epoch": 0.28500775388742194,
      "grad_norm": 0.21281974812319984,
      "kl": 0.18515777587890625,
      "learning_rate": 2.0987654320987653e-07,
      "loss": 0.0002,
      "reward": 1.775000050663948,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 1700
    },
    {
      "completion_length": 220.08036613464355,
      "epoch": 0.28534305712728947,
      "grad_norm": 0.3371442250969937,
      "kl": 0.10428047180175781,
      "learning_rate": 2.1012345679012345e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7678571864962578,
      "rewards/format_reward_func": 1.0,
      "step": 1702
    },
    {
      "completion_length": 222.62054538726807,
      "epoch": 0.28567836036715705,
      "grad_norm": 0.5105200926308263,
      "kl": 0.22168350219726562,
      "learning_rate": 2.1037037037037036e-07,
      "loss": 0.0002,
      "reward": 1.7428572252392769,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571842610836,
      "rewards/format_reward_func": 1.0,
      "step": 1704
    },
    {
      "completion_length": 226.08929538726807,
      "epoch": 0.2860136636070246,
      "grad_norm": 0.4370022389501131,
      "kl": 0.1873798370361328,
      "learning_rate": 2.1061728395061727e-07,
      "loss": 0.0002,
      "reward": 1.773214340209961,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.777678593993187,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1706
    },
    {
      "completion_length": 226.66518878936768,
      "epoch": 0.28634896684689215,
      "grad_norm": 0.11375834492782687,
      "kl": 0.1496267318725586,
      "learning_rate": 2.108641975308642e-07,
      "loss": 0.0002,
      "reward": 1.8160714656114578,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.8205357566475868,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1708
    },
    {
      "completion_length": 230.83483123779297,
      "epoch": 0.28668427008675973,
      "grad_norm": 0.46898913636215384,
      "kl": 0.1600666046142578,
      "learning_rate": 2.111111111111111e-07,
      "loss": 0.0002,
      "reward": 1.7767858058214188,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7812500149011612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1710
    },
    {
      "completion_length": 219.47768783569336,
      "epoch": 0.28701957332662725,
      "grad_norm": 0.0017106252395304078,
      "kl": 0.03565216064453125,
      "learning_rate": 2.1135802469135802e-07,
      "loss": 0.0,
      "reward": 1.7357143685221672,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7357143107801676,
      "rewards/format_reward_func": 1.0,
      "step": 1712
    },
    {
      "completion_length": 221.4509048461914,
      "epoch": 0.28735487656649483,
      "grad_norm": 0.29632758745427196,
      "kl": 0.10880851745605469,
      "learning_rate": 2.116049382716049e-07,
      "loss": 0.0001,
      "reward": 1.8000000417232513,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 1714
    },
    {
      "completion_length": 236.94197463989258,
      "epoch": 0.28769017980636236,
      "grad_norm": 0.2917454687041177,
      "kl": 0.2630786895751953,
      "learning_rate": 2.1185185185185185e-07,
      "loss": 0.0003,
      "reward": 1.787500061094761,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7919643074274063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1716
    },
    {
      "completion_length": 222.91965293884277,
      "epoch": 0.28802548304622994,
      "grad_norm": 0.3137815271928074,
      "kl": 0.04746055603027344,
      "learning_rate": 2.1209876543209876e-07,
      "loss": 0.0,
      "reward": 1.7642857804894447,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 1718
    },
    {
      "completion_length": 231.45983123779297,
      "epoch": 0.2883607862860975,
      "grad_norm": 0.28325392986376813,
      "kl": 0.1917266845703125,
      "learning_rate": 2.1234567901234568e-07,
      "loss": 0.0002,
      "reward": 1.8071429133415222,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428686380386,
      "rewards/format_reward_func": 1.0,
      "step": 1720
    },
    {
      "completion_length": 226.11608219146729,
      "epoch": 0.28869608952596504,
      "grad_norm": 0.6358652431461438,
      "kl": 0.370849609375,
      "learning_rate": 2.1259259259259257e-07,
      "loss": 0.0004,
      "reward": 1.7017857879400253,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7062500268220901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1722
    },
    {
      "completion_length": 232.03572463989258,
      "epoch": 0.2890313927658326,
      "grad_norm": 0.1300999799664777,
      "kl": 0.3420524597167969,
      "learning_rate": 2.1283950617283948e-07,
      "loss": 0.0003,
      "reward": 1.7178572341799736,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7178571783006191,
      "rewards/format_reward_func": 1.0,
      "step": 1724
    },
    {
      "completion_length": 223.0357255935669,
      "epoch": 0.28936669600570014,
      "grad_norm": 0.269776246641907,
      "kl": 0.7043094635009766,
      "learning_rate": 2.1308641975308642e-07,
      "loss": 0.0007,
      "reward": 1.7642857804894447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 1726
    },
    {
      "completion_length": 229.25447177886963,
      "epoch": 0.2897019992455677,
      "grad_norm": 0.26928261439673645,
      "kl": 0.1688404083251953,
      "learning_rate": 2.1333333333333334e-07,
      "loss": 0.0002,
      "reward": 1.725000075995922,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000461935997,
      "rewards/format_reward_func": 1.0,
      "step": 1728
    },
    {
      "completion_length": 227.66518688201904,
      "epoch": 0.29003730248543524,
      "grad_norm": 0.25291738156225757,
      "kl": 0.019290924072265625,
      "learning_rate": 2.1358024691358023e-07,
      "loss": 0.0,
      "reward": 1.782142922282219,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 1730
    },
    {
      "completion_length": 232.74108123779297,
      "epoch": 0.2903726057253028,
      "grad_norm": 0.3125680880835837,
      "kl": 0.11983871459960938,
      "learning_rate": 2.1382716049382714e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7571428958326578,
      "rewards/format_reward_func": 1.0,
      "step": 1732
    },
    {
      "completion_length": 234.12501049041748,
      "epoch": 0.2907079089651704,
      "grad_norm": 0.16799823386356563,
      "kl": 0.32921409606933594,
      "learning_rate": 2.1407407407407408e-07,
      "loss": 0.0003,
      "reward": 1.7500000670552254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 1734
    },
    {
      "completion_length": 229.38840293884277,
      "epoch": 0.2910432122050379,
      "grad_norm": 0.2370698650557852,
      "kl": 0.29618072509765625,
      "learning_rate": 2.1432098765432097e-07,
      "loss": 0.0003,
      "reward": 1.767857201397419,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 1736
    },
    {
      "completion_length": 233.07590198516846,
      "epoch": 0.2913785154449055,
      "grad_norm": 0.31089682021611725,
      "kl": 0.3679676055908203,
      "learning_rate": 2.1456790123456789e-07,
      "loss": 0.0004,
      "reward": 1.746428668498993,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 1738
    },
    {
      "completion_length": 220.51786518096924,
      "epoch": 0.29171381868477303,
      "grad_norm": 0.3056727159750165,
      "kl": 0.5169277191162109,
      "learning_rate": 2.148148148148148e-07,
      "loss": 0.0005,
      "reward": 1.7142857983708382,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857443541288,
      "rewards/format_reward_func": 1.0,
      "step": 1740
    },
    {
      "completion_length": 229.81250953674316,
      "epoch": 0.2920491219246406,
      "grad_norm": 0.1519463725257348,
      "kl": 0.013641357421875,
      "learning_rate": 2.1506172839506174e-07,
      "loss": 0.0,
      "reward": 1.707142911851406,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7071428913623095,
      "rewards/format_reward_func": 1.0,
      "step": 1742
    },
    {
      "completion_length": 233.602689743042,
      "epoch": 0.29238442516450813,
      "grad_norm": 0.3022476612112489,
      "kl": 0.06960678100585938,
      "learning_rate": 2.1530864197530863e-07,
      "loss": 0.0001,
      "reward": 1.70089291036129,
      "reward_std": 0.069447988178581,
      "rewards/equation_reward_func": 0.7205357626080513,
      "rewards/format_reward_func": 0.9803571552038193,
      "step": 1744
    },
    {
      "completion_length": 225.93304443359375,
      "epoch": 0.2927197284043757,
      "grad_norm": 0.16061238412841028,
      "kl": 0.06715774536132812,
      "learning_rate": 2.1555555555555554e-07,
      "loss": 0.0001,
      "reward": 1.7910714745521545,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7955357432365417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1746
    },
    {
      "completion_length": 231.93304634094238,
      "epoch": 0.29305503164424324,
      "grad_norm": 0.1863992188614141,
      "kl": 0.30138206481933594,
      "learning_rate": 2.1580246913580246e-07,
      "loss": 0.0003,
      "reward": 1.7571429014205933,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7660714574158192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1748
    },
    {
      "completion_length": 234.3259038925171,
      "epoch": 0.2933903348841108,
      "grad_norm": 0.4137435993933847,
      "kl": 0.1421070098876953,
      "learning_rate": 2.1604938271604935e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7928571812808514,
      "rewards/format_reward_func": 1.0,
      "step": 1750
    },
    {
      "completion_length": 225.38840579986572,
      "epoch": 0.2937256381239784,
      "grad_norm": 0.24968596600314152,
      "kl": 0.10309982299804688,
      "learning_rate": 2.162962962962963e-07,
      "loss": 0.0001,
      "reward": 1.7160715162754059,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7205357551574707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1752
    },
    {
      "completion_length": 228.83036708831787,
      "epoch": 0.2940609413638459,
      "grad_norm": 0.3084047530337667,
      "kl": 0.11126136779785156,
      "learning_rate": 2.165432098765432e-07,
      "loss": 0.0001,
      "reward": 1.687500074505806,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.6919643208384514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1754
    },
    {
      "completion_length": 241.81697368621826,
      "epoch": 0.2943962446037135,
      "grad_norm": 0.18964625385824094,
      "kl": 0.3287525177001953,
      "learning_rate": 2.1679012345679012e-07,
      "loss": 0.0003,
      "reward": 1.650000087916851,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.65892861969769,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1756
    },
    {
      "completion_length": 236.86608219146729,
      "epoch": 0.294731547843581,
      "grad_norm": 0.3619400638075198,
      "kl": 0.15472793579101562,
      "learning_rate": 2.17037037037037e-07,
      "loss": 0.0002,
      "reward": 1.773214340209961,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7776785921305418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1758
    },
    {
      "completion_length": 235.95983123779297,
      "epoch": 0.2950668510834486,
      "grad_norm": 0.3397716552384549,
      "kl": 0.6170005798339844,
      "learning_rate": 2.1728395061728395e-07,
      "loss": 0.0006,
      "reward": 1.7392857894301414,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 1.0,
      "step": 1760
    },
    {
      "completion_length": 233.0759048461914,
      "epoch": 0.2954021543233161,
      "grad_norm": 0.23654675030018038,
      "kl": 0.013632774353027344,
      "learning_rate": 2.1753086419753086e-07,
      "loss": 0.0,
      "reward": 1.7642857804894447,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 1762
    },
    {
      "completion_length": 226.56250953674316,
      "epoch": 0.2957374575631837,
      "grad_norm": 0.288814634064109,
      "kl": 0.12615585327148438,
      "learning_rate": 2.1777777777777775e-07,
      "loss": 0.0001,
      "reward": 1.725000075995922,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000163912773,
      "rewards/format_reward_func": 1.0,
      "step": 1764
    },
    {
      "completion_length": 229.02679824829102,
      "epoch": 0.2960727608030513,
      "grad_norm": 0.24413996834108162,
      "kl": 0.1447772979736328,
      "learning_rate": 2.1802469135802467e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7607143260538578,
      "rewards/format_reward_func": 1.0,
      "step": 1766
    },
    {
      "completion_length": 231.7634038925171,
      "epoch": 0.2964080640429188,
      "grad_norm": 0.4000458770990593,
      "kl": 0.15684127807617188,
      "learning_rate": 2.182716049382716e-07,
      "loss": 0.0002,
      "reward": 1.7267858013510704,
      "reward_std": 0.09343910869210958,
      "rewards/equation_reward_func": 0.7312500234693289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1768
    },
    {
      "completion_length": 223.84375953674316,
      "epoch": 0.2967433672827864,
      "grad_norm": 0.2284878268099746,
      "kl": 0.13471412658691406,
      "learning_rate": 2.1851851851851852e-07,
      "loss": 0.0001,
      "reward": 1.767857238650322,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.767857164144516,
      "rewards/format_reward_func": 1.0,
      "step": 1770
    },
    {
      "completion_length": 215.60715293884277,
      "epoch": 0.2970786705226539,
      "grad_norm": 0.19678904064037697,
      "kl": 0.05077934265136719,
      "learning_rate": 2.187654320987654e-07,
      "loss": 0.0001,
      "reward": 1.742857202887535,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571730852127,
      "rewards/format_reward_func": 1.0,
      "step": 1772
    },
    {
      "completion_length": 227.2946538925171,
      "epoch": 0.2974139737625215,
      "grad_norm": 0.33814303171679666,
      "kl": 0.11888504028320312,
      "learning_rate": 2.1901234567901233e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 1774
    },
    {
      "completion_length": 223.72322463989258,
      "epoch": 0.297749277002389,
      "grad_norm": 0.19717208048606372,
      "kl": 0.24188232421875,
      "learning_rate": 2.1925925925925927e-07,
      "loss": 0.0002,
      "reward": 1.7928572222590446,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 1776
    },
    {
      "completion_length": 228.0625123977661,
      "epoch": 0.2980845802422566,
      "grad_norm": 0.271144369694773,
      "kl": 0.2233419418334961,
      "learning_rate": 2.1950617283950618e-07,
      "loss": 0.0002,
      "reward": 1.7839286178350449,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.788392897695303,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1778
    },
    {
      "completion_length": 237.03572368621826,
      "epoch": 0.29841988348212417,
      "grad_norm": 0.2724767955336668,
      "kl": 0.021409988403320312,
      "learning_rate": 2.1975308641975307e-07,
      "loss": 0.0,
      "reward": 1.775000087916851,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 1780
    },
    {
      "completion_length": 230.90179538726807,
      "epoch": 0.2987551867219917,
      "grad_norm": 0.1573001174468948,
      "kl": 0.029628753662109375,
      "learning_rate": 2.1999999999999998e-07,
      "loss": 0.0,
      "reward": 1.7125000953674316,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7250000350177288,
      "rewards/format_reward_func": 0.987500011920929,
      "step": 1782
    },
    {
      "completion_length": 230.42411994934082,
      "epoch": 0.2990904899618593,
      "grad_norm": 0.29308188881949576,
      "kl": 0.06266975402832031,
      "learning_rate": 2.202469135802469e-07,
      "loss": 0.0001,
      "reward": 1.7500000521540642,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 1784
    },
    {
      "completion_length": 225.75447368621826,
      "epoch": 0.2994257932017268,
      "grad_norm": 0.37598447598224566,
      "kl": 0.06233024597167969,
      "learning_rate": 2.2049382716049381e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714693367481,
      "rewards/format_reward_func": 1.0,
      "step": 1786
    },
    {
      "completion_length": 239.06697368621826,
      "epoch": 0.2997610964415944,
      "grad_norm": 0.2410658800415634,
      "kl": 0.08719062805175781,
      "learning_rate": 2.2074074074074073e-07,
      "loss": 0.0001,
      "reward": 1.7142857909202576,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7232143227010965,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1788
    },
    {
      "completion_length": 237.25001430511475,
      "epoch": 0.3000963996814619,
      "grad_norm": 0.27428234718509725,
      "kl": 0.0132598876953125,
      "learning_rate": 2.2098765432098764e-07,
      "loss": 0.0,
      "reward": 1.760714367032051,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7607143074274063,
      "rewards/format_reward_func": 1.0,
      "step": 1790
    },
    {
      "completion_length": 230.80804538726807,
      "epoch": 0.3004317029213295,
      "grad_norm": 0.20216357930418644,
      "kl": 0.03976631164550781,
      "learning_rate": 2.2123456790123456e-07,
      "loss": 0.0,
      "reward": 1.7232143729925156,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7276785839349031,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1792
    },
    {
      "completion_length": 230.75893783569336,
      "epoch": 0.30076700616119706,
      "grad_norm": 0.3042072979752627,
      "kl": 0.09014701843261719,
      "learning_rate": 2.2148148148148147e-07,
      "loss": 0.0001,
      "reward": 1.7464286461472511,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 1794
    },
    {
      "completion_length": 227.02679443359375,
      "epoch": 0.3011023094010646,
      "grad_norm": 0.27795542549745866,
      "kl": 0.04588603973388672,
      "learning_rate": 2.217283950617284e-07,
      "loss": 0.0,
      "reward": 1.7678571939468384,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571827709675,
      "rewards/format_reward_func": 1.0,
      "step": 1796
    },
    {
      "completion_length": 236.84822750091553,
      "epoch": 0.30143761264093216,
      "grad_norm": 0.4820397237602499,
      "kl": 0.4211463928222656,
      "learning_rate": 2.219753086419753e-07,
      "loss": 0.0004,
      "reward": 1.7625000402331352,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.7758928947150707,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1798
    },
    {
      "completion_length": 226.65179538726807,
      "epoch": 0.3017729158807997,
      "grad_norm": 0.2887272445792341,
      "kl": 0.016399383544921875,
      "learning_rate": 2.222222222222222e-07,
      "loss": 0.0,
      "reward": 1.7678571939468384,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571660071611,
      "rewards/format_reward_func": 1.0,
      "step": 1800
    },
    {
      "completion_length": 235.01340293884277,
      "epoch": 0.30210821912066727,
      "grad_norm": 0.6643608729826055,
      "kl": 0.1521320343017578,
      "learning_rate": 2.2246913580246913e-07,
      "loss": 0.0002,
      "reward": 1.7410714998841286,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.745535746216774,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1802
    },
    {
      "completion_length": 225.6384038925171,
      "epoch": 0.3024435223605348,
      "grad_norm": 0.26640352513056065,
      "kl": 0.03233146667480469,
      "learning_rate": 2.2271604938271605e-07,
      "loss": 0.0,
      "reward": 1.7500000596046448,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 1804
    },
    {
      "completion_length": 228.18750953674316,
      "epoch": 0.30277882560040237,
      "grad_norm": 0.14731187386849665,
      "kl": 0.027124404907226562,
      "learning_rate": 2.2296296296296296e-07,
      "loss": 0.0,
      "reward": 1.7089286521077156,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7133928928524256,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1806
    },
    {
      "completion_length": 239.9464406967163,
      "epoch": 0.3031141288402699,
      "grad_norm": 0.29234930155500477,
      "kl": 0.020626068115234375,
      "learning_rate": 2.2320987654320985e-07,
      "loss": 0.0,
      "reward": 1.7607143372297287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143223285675,
      "rewards/format_reward_func": 1.0,
      "step": 1808
    },
    {
      "completion_length": 235.46875858306885,
      "epoch": 0.30344943208013747,
      "grad_norm": 0.1040229140664818,
      "kl": 0.015781402587890625,
      "learning_rate": 2.2345679012345677e-07,
      "loss": 0.0,
      "reward": 1.7285715118050575,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7285714671015739,
      "rewards/format_reward_func": 1.0,
      "step": 1810
    },
    {
      "completion_length": 237.58929920196533,
      "epoch": 0.30378473532000505,
      "grad_norm": 0.10295620756709041,
      "kl": 0.16724872589111328,
      "learning_rate": 2.237037037037037e-07,
      "loss": 0.0002,
      "reward": 1.7696429267525673,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7830357439815998,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1812
    },
    {
      "completion_length": 228.08929634094238,
      "epoch": 0.3041200385598726,
      "grad_norm": 0.15218741534858393,
      "kl": 0.2906198501586914,
      "learning_rate": 2.239506172839506e-07,
      "loss": 0.0003,
      "reward": 1.7428572177886963,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 1814
    },
    {
      "completion_length": 224.18750762939453,
      "epoch": 0.30445534179974015,
      "grad_norm": 0.19074701356213938,
      "kl": 0.06517219543457031,
      "learning_rate": 2.241975308641975e-07,
      "loss": 0.0001,
      "reward": 1.7142858058214188,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7142857518047094,
      "rewards/format_reward_func": 1.0,
      "step": 1816
    },
    {
      "completion_length": 222.05804443359375,
      "epoch": 0.3047906450396077,
      "grad_norm": 0.1476048923105393,
      "kl": 0.012349128723144531,
      "learning_rate": 2.2444444444444442e-07,
      "loss": 0.0,
      "reward": 1.742857202887535,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 1818
    },
    {
      "completion_length": 233.8750114440918,
      "epoch": 0.30512594827947526,
      "grad_norm": 0.17331623039744623,
      "kl": 0.01285552978515625,
      "learning_rate": 2.2469135802469137e-07,
      "loss": 0.0,
      "reward": 1.8000000566244125,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8000000193715096,
      "rewards/format_reward_func": 1.0,
      "step": 1820
    },
    {
      "completion_length": 231.1250114440918,
      "epoch": 0.3054612515193428,
      "grad_norm": 0.3307768489017182,
      "kl": 0.1770782470703125,
      "learning_rate": 2.2493827160493825e-07,
      "loss": 0.0002,
      "reward": 1.744642935693264,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7491071708500385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1822
    },
    {
      "completion_length": 233.92858028411865,
      "epoch": 0.30579655475921036,
      "grad_norm": 0.22175695720606406,
      "kl": 0.16938114166259766,
      "learning_rate": 2.2518518518518517e-07,
      "loss": 0.0002,
      "reward": 1.7892857789993286,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 1824
    },
    {
      "completion_length": 230.14733219146729,
      "epoch": 0.30613185799907794,
      "grad_norm": 0.27675034844790847,
      "kl": 0.020910263061523438,
      "learning_rate": 2.2543209876543208e-07,
      "loss": 0.0,
      "reward": 1.8107143640518188,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8107143081724644,
      "rewards/format_reward_func": 1.0,
      "step": 1826
    },
    {
      "completion_length": 247.6160831451416,
      "epoch": 0.30646716123894546,
      "grad_norm": 0.44542823477944804,
      "kl": 0.24846267700195312,
      "learning_rate": 2.2567901234567903e-07,
      "loss": 0.0002,
      "reward": 1.739285796880722,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7482143267989159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1828
    },
    {
      "completion_length": 226.4062623977661,
      "epoch": 0.30680246447881304,
      "grad_norm": 0.3461096636929037,
      "kl": 0.05735588073730469,
      "learning_rate": 2.2592592592592591e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857618629932,
      "rewards/format_reward_func": 1.0,
      "step": 1830
    },
    {
      "completion_length": 237.54465198516846,
      "epoch": 0.30713776771868057,
      "grad_norm": 0.4290936324075917,
      "kl": 0.08965682983398438,
      "learning_rate": 2.2617283950617283e-07,
      "loss": 0.0001,
      "reward": 1.7321429401636124,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7410714589059353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1832
    },
    {
      "completion_length": 234.8839406967163,
      "epoch": 0.30747307095854814,
      "grad_norm": 0.28822848873036533,
      "kl": 0.035064697265625,
      "learning_rate": 2.2641975308641974e-07,
      "loss": 0.0,
      "reward": 1.7821428924798965,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428906172514,
      "rewards/format_reward_func": 1.0,
      "step": 1834
    },
    {
      "completion_length": 226.7634038925171,
      "epoch": 0.30780837419841567,
      "grad_norm": 0.14622024873230458,
      "kl": 0.35060882568359375,
      "learning_rate": 2.2666666666666663e-07,
      "loss": 0.0004,
      "reward": 1.7767857611179352,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7812500484287739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1836
    },
    {
      "completion_length": 241.6919755935669,
      "epoch": 0.30814367743828325,
      "grad_norm": 0.35153050154322957,
      "kl": 0.012337684631347656,
      "learning_rate": 2.2691358024691357e-07,
      "loss": 0.0,
      "reward": 1.7267858162522316,
      "reward_std": 0.09343910869210958,
      "rewards/equation_reward_func": 0.7312500290572643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1838
    },
    {
      "completion_length": 224.90179634094238,
      "epoch": 0.3084789806781508,
      "grad_norm": 0.2655829602093783,
      "kl": 0.0985097885131836,
      "learning_rate": 2.271604938271605e-07,
      "loss": 0.0001,
      "reward": 1.801785759627819,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8062500357627869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1840
    },
    {
      "completion_length": 241.04911994934082,
      "epoch": 0.30881428391801835,
      "grad_norm": 0.2453221248498777,
      "kl": 0.4122467041015625,
      "learning_rate": 2.274074074074074e-07,
      "loss": 0.0004,
      "reward": 1.7589286044239998,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7633928991854191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1842
    },
    {
      "completion_length": 231.6741180419922,
      "epoch": 0.30914958715788593,
      "grad_norm": 0.2234689362529436,
      "kl": 0.09752082824707031,
      "learning_rate": 2.276543209876543e-07,
      "loss": 0.0001,
      "reward": 1.7142857983708382,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857518047094,
      "rewards/format_reward_func": 1.0,
      "step": 1844
    },
    {
      "completion_length": 229.68751049041748,
      "epoch": 0.30948489039775345,
      "grad_norm": 0.33753612337173233,
      "kl": 0.017671585083007812,
      "learning_rate": 2.2790123456790123e-07,
      "loss": 0.0,
      "reward": 1.7642857879400253,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 1846
    },
    {
      "completion_length": 236.9821538925171,
      "epoch": 0.30982019363762103,
      "grad_norm": 0.34463075474830307,
      "kl": 0.06949996948242188,
      "learning_rate": 2.2814814814814815e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 1848
    },
    {
      "completion_length": 222.60268783569336,
      "epoch": 0.31015549687748856,
      "grad_norm": 0.3017657562993159,
      "kl": 0.17605209350585938,
      "learning_rate": 2.2839506172839504e-07,
      "loss": 0.0002,
      "reward": 1.7410714998841286,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.745535746216774,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1850
    },
    {
      "completion_length": 233.35268878936768,
      "epoch": 0.31049080011735614,
      "grad_norm": 0.19114987608251482,
      "kl": 0.11333656311035156,
      "learning_rate": 2.2864197530864195e-07,
      "loss": 0.0001,
      "reward": 1.7446429207921028,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.74910718947649,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1852
    },
    {
      "completion_length": 239.08929443359375,
      "epoch": 0.3108261033572237,
      "grad_norm": 0.38085970590024615,
      "kl": 0.1593780517578125,
      "learning_rate": 2.288888888888889e-07,
      "loss": 0.0002,
      "reward": 1.7250000536441803,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000461935997,
      "rewards/format_reward_func": 1.0,
      "step": 1854
    },
    {
      "completion_length": 231.82143878936768,
      "epoch": 0.31116140659709124,
      "grad_norm": 0.7838052230054621,
      "kl": 0.08046340942382812,
      "learning_rate": 2.291358024691358e-07,
      "loss": 0.0001,
      "reward": 1.8071429207921028,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428798139095,
      "rewards/format_reward_func": 1.0,
      "step": 1856
    },
    {
      "completion_length": 231.53126049041748,
      "epoch": 0.3114967098369588,
      "grad_norm": 0.22527110312924714,
      "kl": 0.16312026977539062,
      "learning_rate": 2.293827160493827e-07,
      "loss": 0.0002,
      "reward": 1.8035714775323868,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714402794838,
      "rewards/format_reward_func": 1.0,
      "step": 1858
    },
    {
      "completion_length": 226.75447273254395,
      "epoch": 0.31183201307682634,
      "grad_norm": 0.20671855052520396,
      "kl": 0.08982467651367188,
      "learning_rate": 2.296296296296296e-07,
      "loss": 0.0001,
      "reward": 1.7392857670783997,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 1.0,
      "step": 1860
    },
    {
      "completion_length": 229.11161518096924,
      "epoch": 0.3121673163166939,
      "grad_norm": 0.23576458309190282,
      "kl": 0.07481575012207031,
      "learning_rate": 2.2987654320987655e-07,
      "loss": 0.0001,
      "reward": 1.74642863124609,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286107569933,
      "rewards/format_reward_func": 1.0,
      "step": 1862
    },
    {
      "completion_length": 220.23215293884277,
      "epoch": 0.31250261955656145,
      "grad_norm": 0.2948337250310543,
      "kl": 0.048417091369628906,
      "learning_rate": 2.3012345679012347e-07,
      "loss": 0.0,
      "reward": 1.7910714745521545,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7955357506871223,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1864
    },
    {
      "completion_length": 231.8035831451416,
      "epoch": 0.312837922796429,
      "grad_norm": 0.3180397707419142,
      "kl": 0.17154312133789062,
      "learning_rate": 2.3037037037037035e-07,
      "loss": 0.0002,
      "reward": 1.6535715013742447,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.6535714697092772,
      "rewards/format_reward_func": 1.0,
      "step": 1866
    },
    {
      "completion_length": 222.42858219146729,
      "epoch": 0.31317322603629655,
      "grad_norm": 0.23005009438891452,
      "kl": 0.07344245910644531,
      "learning_rate": 2.3061728395061727e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7410714626312256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1868
    },
    {
      "completion_length": 241.3794765472412,
      "epoch": 0.3135085292761641,
      "grad_norm": 0.27632991514710453,
      "kl": 0.26630401611328125,
      "learning_rate": 2.3086419753086418e-07,
      "loss": 0.0003,
      "reward": 1.807142935693264,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.807142898440361,
      "rewards/format_reward_func": 1.0,
      "step": 1870
    },
    {
      "completion_length": 231.10268783569336,
      "epoch": 0.3138438325160317,
      "grad_norm": 0.24458798323898348,
      "kl": 0.024705886840820312,
      "learning_rate": 2.311111111111111e-07,
      "loss": 0.0,
      "reward": 1.7160714864730835,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7205357383936644,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1872
    },
    {
      "completion_length": 231.321439743042,
      "epoch": 0.31417913575589923,
      "grad_norm": 0.22131545732884386,
      "kl": 0.034271240234375,
      "learning_rate": 2.31358024691358e-07,
      "loss": 0.0,
      "reward": 1.8000000342726707,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.800000037997961,
      "rewards/format_reward_func": 1.0,
      "step": 1874
    },
    {
      "completion_length": 229.5089406967163,
      "epoch": 0.3145144389957668,
      "grad_norm": 0.22465986688885642,
      "kl": 0.165924072265625,
      "learning_rate": 2.3160493827160493e-07,
      "loss": 0.0002,
      "reward": 1.7660714834928513,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1876
    },
    {
      "completion_length": 233.3884038925171,
      "epoch": 0.31484974223563433,
      "grad_norm": 0.23910415798432272,
      "kl": 0.18875694274902344,
      "learning_rate": 2.3185185185185184e-07,
      "loss": 0.0002,
      "reward": 1.7535714879631996,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714656114578,
      "rewards/format_reward_func": 1.0,
      "step": 1878
    },
    {
      "completion_length": 233.0357265472412,
      "epoch": 0.3151850454755019,
      "grad_norm": 0.40358730438180807,
      "kl": 0.04170417785644531,
      "learning_rate": 2.3209876543209876e-07,
      "loss": 0.0,
      "reward": 1.7607143595814705,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.769642885774374,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1880
    },
    {
      "completion_length": 225.48661994934082,
      "epoch": 0.31552034871536944,
      "grad_norm": 0.2931343957440977,
      "kl": 0.08726978302001953,
      "learning_rate": 2.3234567901234567e-07,
      "loss": 0.0001,
      "reward": 1.7607143446803093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143223285675,
      "rewards/format_reward_func": 1.0,
      "step": 1882
    },
    {
      "completion_length": 231.7009048461914,
      "epoch": 0.315855651955237,
      "grad_norm": 0.5066000684278723,
      "kl": 0.06025505065917969,
      "learning_rate": 2.325925925925926e-07,
      "loss": 0.0001,
      "reward": 1.7053572237491608,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7098214663565159,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1884
    },
    {
      "completion_length": 225.73661613464355,
      "epoch": 0.3161909551951046,
      "grad_norm": 0.11965847860757209,
      "kl": 0.09498214721679688,
      "learning_rate": 2.3283950617283948e-07,
      "loss": 0.0001,
      "reward": 1.8071429058909416,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8071428872644901,
      "rewards/format_reward_func": 1.0,
      "step": 1886
    },
    {
      "completion_length": 235.83483123779297,
      "epoch": 0.3165262584349721,
      "grad_norm": 0.27708246702194467,
      "kl": 0.26693153381347656,
      "learning_rate": 2.3308641975308642e-07,
      "loss": 0.0003,
      "reward": 1.7107143625617027,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.71071432903409,
      "rewards/format_reward_func": 1.0,
      "step": 1888
    },
    {
      "completion_length": 230.55358219146729,
      "epoch": 0.3168615616748397,
      "grad_norm": 0.2196535389068709,
      "kl": 0.1350536346435547,
      "learning_rate": 2.3333333333333333e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143256813288,
      "rewards/format_reward_func": 1.0,
      "step": 1890
    },
    {
      "completion_length": 233.94643783569336,
      "epoch": 0.3171968649147072,
      "grad_norm": 0.16739288082692094,
      "kl": 0.018312454223632812,
      "learning_rate": 2.3358024691358025e-07,
      "loss": 0.0,
      "reward": 1.6982143446803093,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.702678618952632,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1892
    },
    {
      "completion_length": 227.74554634094238,
      "epoch": 0.3175321681545748,
      "grad_norm": 0.23554998821278342,
      "kl": 0.04662895202636719,
      "learning_rate": 2.3382716049382713e-07,
      "loss": 0.0,
      "reward": 1.764285758137703,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 1894
    },
    {
      "completion_length": 230.40625953674316,
      "epoch": 0.3178674713944423,
      "grad_norm": 0.3893511360014083,
      "kl": 0.03877544403076172,
      "learning_rate": 2.3407407407407405e-07,
      "loss": 0.0,
      "reward": 1.7571429312229156,
      "reward_std": 0.09091372694820166,
      "rewards/equation_reward_func": 0.7660714574158192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1896
    },
    {
      "completion_length": 235.1250114440918,
      "epoch": 0.3182027746343099,
      "grad_norm": 0.2841157377433794,
      "kl": 0.19602394104003906,
      "learning_rate": 2.34320987654321e-07,
      "loss": 0.0002,
      "reward": 1.7428572252392769,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 1898
    },
    {
      "completion_length": 226.2366189956665,
      "epoch": 0.3185380778741775,
      "grad_norm": 0.17757173837637225,
      "kl": 0.0174407958984375,
      "learning_rate": 2.3456790123456788e-07,
      "loss": 0.0,
      "reward": 1.7357143387198448,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 1900
    },
    {
      "completion_length": 237.0000123977661,
      "epoch": 0.318873381114045,
      "grad_norm": 0.43075952842159304,
      "kl": 0.13301658630371094,
      "learning_rate": 2.348148148148148e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857301980257,
      "rewards/format_reward_func": 1.0,
      "step": 1902
    },
    {
      "completion_length": 230.7544755935669,
      "epoch": 0.3192086843539126,
      "grad_norm": 0.2751534698278088,
      "kl": 0.07329368591308594,
      "learning_rate": 2.350617283950617e-07,
      "loss": 0.0001,
      "reward": 1.7053572237491608,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7098214663565159,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1904
    },
    {
      "completion_length": 228.5669755935669,
      "epoch": 0.3195439875937801,
      "grad_norm": 0.25202227295897695,
      "kl": 0.03522491455078125,
      "learning_rate": 2.3530864197530865e-07,
      "loss": 0.0,
      "reward": 1.7500000670552254,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000447034836,
      "rewards/format_reward_func": 1.0,
      "step": 1906
    },
    {
      "completion_length": 238.35715675354004,
      "epoch": 0.3198792908336477,
      "grad_norm": 0.17488179771033238,
      "kl": 0.021076202392578125,
      "learning_rate": 2.3555555555555554e-07,
      "loss": 0.0,
      "reward": 1.7714286595582962,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 1908
    },
    {
      "completion_length": 226.04465579986572,
      "epoch": 0.3202145940735152,
      "grad_norm": 0.29303369817907765,
      "kl": 0.011088371276855469,
      "learning_rate": 2.3580246913580245e-07,
      "loss": 0.0,
      "reward": 1.764285795390606,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 1910
    },
    {
      "completion_length": 238.8259048461914,
      "epoch": 0.3205498973133828,
      "grad_norm": 0.28677067456421385,
      "kl": 0.06038475036621094,
      "learning_rate": 2.3604938271604937e-07,
      "loss": 0.0001,
      "reward": 1.7589286267757416,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7633928842842579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1912
    },
    {
      "completion_length": 230.89286613464355,
      "epoch": 0.32088520055325037,
      "grad_norm": 0.23104718016621956,
      "kl": 0.023987770080566406,
      "learning_rate": 2.362962962962963e-07,
      "loss": 0.0,
      "reward": 1.7535714954137802,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714469850063,
      "rewards/format_reward_func": 1.0,
      "step": 1914
    },
    {
      "completion_length": 238.77679824829102,
      "epoch": 0.3212205037931179,
      "grad_norm": 0.24706180736743527,
      "kl": 0.2645835876464844,
      "learning_rate": 2.365432098765432e-07,
      "loss": 0.0003,
      "reward": 1.7500000596046448,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 1916
    },
    {
      "completion_length": 240.2544755935669,
      "epoch": 0.3215558070329855,
      "grad_norm": 0.2133498248289064,
      "kl": 0.12011337280273438,
      "learning_rate": 2.367901234567901e-07,
      "loss": 0.0001,
      "reward": 1.758928619325161,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7633928954601288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1918
    },
    {
      "completion_length": 228.94197368621826,
      "epoch": 0.321891110272853,
      "grad_norm": 0.43973771460402583,
      "kl": 0.11147499084472656,
      "learning_rate": 2.3703703703703703e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 1920
    },
    {
      "completion_length": 230.3169755935669,
      "epoch": 0.3222264135127206,
      "grad_norm": 0.22943147719990709,
      "kl": 0.04773712158203125,
      "learning_rate": 2.3728395061728394e-07,
      "loss": 0.0,
      "reward": 1.7803571969270706,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7848214581608772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1922
    },
    {
      "completion_length": 231.18751049041748,
      "epoch": 0.3225617167525881,
      "grad_norm": 0.35686116682750085,
      "kl": 0.07299137115478516,
      "learning_rate": 2.3753086419753086e-07,
      "loss": 0.0001,
      "reward": 1.7125001028180122,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.716964315623045,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1924
    },
    {
      "completion_length": 224.07143783569336,
      "epoch": 0.3228970199924557,
      "grad_norm": 0.30472172697594374,
      "kl": 0.029535293579101562,
      "learning_rate": 2.3777777777777777e-07,
      "loss": 0.0,
      "reward": 1.7500000819563866,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 1926
    },
    {
      "completion_length": 244.20983409881592,
      "epoch": 0.32323232323232326,
      "grad_norm": 0.2661952587072812,
      "kl": 0.019370079040527344,
      "learning_rate": 2.3802469135802469e-07,
      "loss": 0.0,
      "reward": 1.7642857655882835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 1928
    },
    {
      "completion_length": 231.15179824829102,
      "epoch": 0.3235676264721908,
      "grad_norm": 0.2501130921501854,
      "kl": 0.027837753295898438,
      "learning_rate": 2.3827160493827157e-07,
      "loss": 0.0,
      "reward": 1.7928571924567223,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 1930
    },
    {
      "completion_length": 233.75001430511475,
      "epoch": 0.32390292971205836,
      "grad_norm": 0.44629294557250104,
      "kl": 0.07960128784179688,
      "learning_rate": 2.385185185185185e-07,
      "loss": 0.0001,
      "reward": 1.6946429312229156,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.6991071682423353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1932
    },
    {
      "completion_length": 235.3928680419922,
      "epoch": 0.3242382329519259,
      "grad_norm": 0.5119444231013779,
      "kl": 0.10437774658203125,
      "learning_rate": 2.3876543209876543e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.739285746589303,
      "rewards/format_reward_func": 1.0,
      "step": 1934
    },
    {
      "completion_length": 236.93304538726807,
      "epoch": 0.32457353619179347,
      "grad_norm": 0.2119830430647975,
      "kl": 0.03171539306640625,
      "learning_rate": 2.390123456790123e-07,
      "loss": 0.0,
      "reward": 1.7464286461472511,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.746428593993187,
      "rewards/format_reward_func": 1.0,
      "step": 1936
    },
    {
      "completion_length": 228.1294755935669,
      "epoch": 0.324908839431661,
      "grad_norm": 0.27788889937507205,
      "kl": 0.01113128662109375,
      "learning_rate": 2.3925925925925926e-07,
      "loss": 0.0,
      "reward": 1.6607143729925156,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.6607143338769674,
      "rewards/format_reward_func": 1.0,
      "step": 1938
    },
    {
      "completion_length": 237.4955472946167,
      "epoch": 0.32524414267152857,
      "grad_norm": 0.24299818770137305,
      "kl": 0.013406753540039062,
      "learning_rate": 2.3950617283950615e-07,
      "loss": 0.0,
      "reward": 1.7428572252392769,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 1940
    },
    {
      "completion_length": 233.36608409881592,
      "epoch": 0.3255794459113961,
      "grad_norm": 0.1810489507798674,
      "kl": 0.03422355651855469,
      "learning_rate": 2.397530864197531e-07,
      "loss": 0.0,
      "reward": 1.7392857894301414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857447266579,
      "rewards/format_reward_func": 1.0,
      "step": 1942
    },
    {
      "completion_length": 229.81697368621826,
      "epoch": 0.32591474915126367,
      "grad_norm": 0.1674598005786771,
      "kl": 0.044902801513671875,
      "learning_rate": 2.4e-07,
      "loss": 0.0,
      "reward": 1.7607143446803093,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143074274063,
      "rewards/format_reward_func": 1.0,
      "step": 1944
    },
    {
      "completion_length": 229.18304634094238,
      "epoch": 0.32625005239113125,
      "grad_norm": 0.1620064245637525,
      "kl": 0.017425537109375,
      "learning_rate": 2.402469135802469e-07,
      "loss": 0.0,
      "reward": 1.7000000923871994,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.700000025331974,
      "rewards/format_reward_func": 1.0,
      "step": 1946
    },
    {
      "completion_length": 236.3392972946167,
      "epoch": 0.3265853556309988,
      "grad_norm": 0.20727220749960834,
      "kl": 0.010746002197265625,
      "learning_rate": 2.404938271604938e-07,
      "loss": 0.0,
      "reward": 1.7107143551111221,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7196428794413805,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1948
    },
    {
      "completion_length": 234.52679634094238,
      "epoch": 0.32692065887086635,
      "grad_norm": 0.3255722185826685,
      "kl": 0.07966041564941406,
      "learning_rate": 2.407407407407407e-07,
      "loss": 0.0001,
      "reward": 1.7839286476373672,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7883928865194321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1950
    },
    {
      "completion_length": 229.72768783569336,
      "epoch": 0.3272559621107339,
      "grad_norm": 0.32015743982788275,
      "kl": 0.08532905578613281,
      "learning_rate": 2.4098765432098764e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857372760773,
      "rewards/format_reward_func": 1.0,
      "step": 1952
    },
    {
      "completion_length": 227.7232265472412,
      "epoch": 0.32759126535060146,
      "grad_norm": 0.21736216860562788,
      "kl": 0.04352378845214844,
      "learning_rate": 2.412345679012346e-07,
      "loss": 0.0,
      "reward": 1.785714328289032,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143115252256,
      "rewards/format_reward_func": 1.0,
      "step": 1954
    },
    {
      "completion_length": 229.696439743042,
      "epoch": 0.327926568590469,
      "grad_norm": 0.214901028849602,
      "kl": 0.054378509521484375,
      "learning_rate": 2.4148148148148147e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 1956
    },
    {
      "completion_length": 230.96429538726807,
      "epoch": 0.32826187183033656,
      "grad_norm": 0.35751988952802904,
      "kl": 0.04533100128173828,
      "learning_rate": 2.4172839506172836e-07,
      "loss": 0.0,
      "reward": 1.7142857909202576,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7142857611179352,
      "rewards/format_reward_func": 1.0,
      "step": 1958
    },
    {
      "completion_length": 225.32590293884277,
      "epoch": 0.32859717507020414,
      "grad_norm": 0.29350278528039353,
      "kl": 0.014001846313476562,
      "learning_rate": 2.419753086419753e-07,
      "loss": 0.0,
      "reward": 1.7321429327130318,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321428991854191,
      "rewards/format_reward_func": 1.0,
      "step": 1960
    },
    {
      "completion_length": 226.90626049041748,
      "epoch": 0.32893247831007166,
      "grad_norm": 0.2693844120927604,
      "kl": 0.07091999053955078,
      "learning_rate": 2.4222222222222224e-07,
      "loss": 0.0001,
      "reward": 1.79464291036129,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.8080357313156128,
      "rewards/format_reward_func": 0.9866071455180645,
      "step": 1962
    },
    {
      "completion_length": 232.9732255935669,
      "epoch": 0.32926778154993924,
      "grad_norm": 0.36427653784791236,
      "kl": 0.00850677490234375,
      "learning_rate": 2.4246913580246913e-07,
      "loss": 0.0,
      "reward": 1.7107143476605415,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7196428999304771,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1964
    },
    {
      "completion_length": 238.4821538925171,
      "epoch": 0.32960308478980677,
      "grad_norm": 0.18662587745852874,
      "kl": 0.014192581176757812,
      "learning_rate": 2.42716049382716e-07,
      "loss": 0.0,
      "reward": 1.7178571969270706,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7178571801632643,
      "rewards/format_reward_func": 1.0,
      "step": 1966
    },
    {
      "completion_length": 235.21429538726807,
      "epoch": 0.32993838802967435,
      "grad_norm": 0.3017454432462034,
      "kl": 0.0271148681640625,
      "learning_rate": 2.4296296296296296e-07,
      "loss": 0.0,
      "reward": 1.755357213318348,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214447498322,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1968
    },
    {
      "completion_length": 222.56697368621826,
      "epoch": 0.33027369126954187,
      "grad_norm": 0.3683075178582106,
      "kl": 0.008819580078125,
      "learning_rate": 2.432098765432099e-07,
      "loss": 0.0,
      "reward": 1.8504464700818062,
      "reward_std": 0.07007933221757412,
      "rewards/equation_reward_func": 0.8562500178813934,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 1970
    },
    {
      "completion_length": 225.54018688201904,
      "epoch": 0.33060899450940945,
      "grad_norm": 0.254113309813442,
      "kl": 0.015944480895996094,
      "learning_rate": 2.434567901234568e-07,
      "loss": 0.0,
      "reward": 1.7678572088479996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 1972
    },
    {
      "completion_length": 228.3303680419922,
      "epoch": 0.330944297749277,
      "grad_norm": 0.3838229118855729,
      "kl": 0.012636184692382812,
      "learning_rate": 2.437037037037037e-07,
      "loss": 0.0,
      "reward": 1.7071429416537285,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7160714659839869,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 1974
    },
    {
      "completion_length": 237.89733028411865,
      "epoch": 0.33127960098914455,
      "grad_norm": 0.33245679288182095,
      "kl": 0.03874683380126953,
      "learning_rate": 2.439506172839506e-07,
      "loss": 0.0,
      "reward": 1.7553572058677673,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7598214522004128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1976
    },
    {
      "completion_length": 228.4241180419922,
      "epoch": 0.33161490422901213,
      "grad_norm": 0.1817437147379376,
      "kl": 0.06694793701171875,
      "learning_rate": 2.4419753086419756e-07,
      "loss": 0.0001,
      "reward": 1.8107143566012383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8107143081724644,
      "rewards/format_reward_func": 1.0,
      "step": 1978
    },
    {
      "completion_length": 231.20090293884277,
      "epoch": 0.33195020746887965,
      "grad_norm": 0.23502872535450134,
      "kl": 0.08605766296386719,
      "learning_rate": 2.4444444444444445e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000432133675,
      "rewards/format_reward_func": 1.0,
      "step": 1980
    },
    {
      "completion_length": 228.21429538726807,
      "epoch": 0.33228551070874723,
      "grad_norm": 0.1645380719145372,
      "kl": 0.022317886352539062,
      "learning_rate": 2.4469135802469133e-07,
      "loss": 0.0,
      "reward": 1.7571429386734962,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 1982
    },
    {
      "completion_length": 230.04018878936768,
      "epoch": 0.33262081394861476,
      "grad_norm": 0.5832458221239087,
      "kl": 0.07063102722167969,
      "learning_rate": 2.449382716049383e-07,
      "loss": 0.0001,
      "reward": 1.7303572222590446,
      "reward_std": 0.0883883461356163,
      "rewards/equation_reward_func": 0.7437500357627869,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 1984
    },
    {
      "completion_length": 230.77679634094238,
      "epoch": 0.33295611718848234,
      "grad_norm": 0.32034128630556463,
      "kl": 0.05539894104003906,
      "learning_rate": 2.4518518518518516e-07,
      "loss": 0.0001,
      "reward": 1.7125000581145287,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7169643081724644,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 1986
    },
    {
      "completion_length": 221.37054538726807,
      "epoch": 0.3332914204283499,
      "grad_norm": 0.3104718829687837,
      "kl": 0.0174102783203125,
      "learning_rate": 2.454320987654321e-07,
      "loss": 0.0,
      "reward": 1.7571429163217545,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 1988
    },
    {
      "completion_length": 230.3973331451416,
      "epoch": 0.33362672366821744,
      "grad_norm": 0.28791233501393015,
      "kl": 0.02490234375,
      "learning_rate": 2.45679012345679e-07,
      "loss": 0.0,
      "reward": 1.814285770058632,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.8142857439815998,
      "rewards/format_reward_func": 1.0,
      "step": 1990
    },
    {
      "completion_length": 231.79911708831787,
      "epoch": 0.333962026908085,
      "grad_norm": 0.1954296500371838,
      "kl": 0.022606849670410156,
      "learning_rate": 2.4592592592592593e-07,
      "loss": 0.0,
      "reward": 1.757142923772335,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 1992
    },
    {
      "completion_length": 236.7053680419922,
      "epoch": 0.33429733014795254,
      "grad_norm": 0.21342070200715885,
      "kl": 0.10403156280517578,
      "learning_rate": 2.461728395061728e-07,
      "loss": 0.0001,
      "reward": 1.7214286401867867,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7214286029338837,
      "rewards/format_reward_func": 1.0,
      "step": 1994
    },
    {
      "completion_length": 225.05358409881592,
      "epoch": 0.3346326333878201,
      "grad_norm": 0.5948907828800926,
      "kl": 0.0234222412109375,
      "learning_rate": 2.4641975308641976e-07,
      "loss": 0.0,
      "reward": 1.7357143685221672,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143070548773,
      "rewards/format_reward_func": 1.0,
      "step": 1996
    },
    {
      "completion_length": 229.7366189956665,
      "epoch": 0.33496793662768765,
      "grad_norm": 0.14636131178410403,
      "kl": 0.037792205810546875,
      "learning_rate": 2.4666666666666665e-07,
      "loss": 0.0,
      "reward": 1.7142857983708382,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7142857499420643,
      "rewards/format_reward_func": 1.0,
      "step": 1998
    },
    {
      "completion_length": 228.67858219146729,
      "epoch": 0.3353032398675552,
      "grad_norm": 0.1664830030099182,
      "kl": 0.01624774932861328,
      "learning_rate": 2.4691358024691354e-07,
      "loss": 0.0,
      "reward": 1.742857202887535,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 2000
    },
    {
      "completion_length": 222.88840293884277,
      "epoch": 0.33563854310742275,
      "grad_norm": 0.26362387045617086,
      "kl": 0.012630462646484375,
      "learning_rate": 2.471604938271605e-07,
      "loss": 0.0,
      "reward": 1.796428620815277,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286133646965,
      "rewards/format_reward_func": 1.0,
      "step": 2002
    },
    {
      "completion_length": 223.72322273254395,
      "epoch": 0.33597384634729033,
      "grad_norm": 0.0014755454691630647,
      "kl": 0.009225845336914062,
      "learning_rate": 2.474074074074074e-07,
      "loss": 0.0,
      "reward": 1.7285715118050575,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7285714708268642,
      "rewards/format_reward_func": 1.0,
      "step": 2004
    },
    {
      "completion_length": 220.60715198516846,
      "epoch": 0.3363091495871579,
      "grad_norm": 0.310672175130401,
      "kl": 0.016305923461914062,
      "learning_rate": 2.476543209876543e-07,
      "loss": 0.0,
      "reward": 1.8000000417232513,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 2006
    },
    {
      "completion_length": 227.81250953674316,
      "epoch": 0.33664445282702543,
      "grad_norm": 0.2702079969416635,
      "kl": 0.10282135009765625,
      "learning_rate": 2.479012345679012e-07,
      "loss": 0.0001,
      "reward": 1.762500062584877,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2008
    },
    {
      "completion_length": 225.70536613464355,
      "epoch": 0.336979756066893,
      "grad_norm": 0.20606098388880464,
      "kl": 0.018157005310058594,
      "learning_rate": 2.4814814814814814e-07,
      "loss": 0.0,
      "reward": 1.8053572103381157,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8098214641213417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2010
    },
    {
      "completion_length": 237.58036708831787,
      "epoch": 0.33731505930676053,
      "grad_norm": 0.2482946763221189,
      "kl": 0.05824089050292969,
      "learning_rate": 2.4839506172839503e-07,
      "loss": 0.0001,
      "reward": 1.7035715132951736,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7035714760422707,
      "rewards/format_reward_func": 1.0,
      "step": 2012
    },
    {
      "completion_length": 233.10269260406494,
      "epoch": 0.3376503625466281,
      "grad_norm": 0.28002200040351294,
      "kl": 0.0264739990234375,
      "learning_rate": 2.4864197530864197e-07,
      "loss": 0.0,
      "reward": 1.7267857864499092,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7312500234693289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2014
    },
    {
      "completion_length": 236.15626049041748,
      "epoch": 0.33798566578649564,
      "grad_norm": 0.47691443247370385,
      "kl": 0.02268695831298828,
      "learning_rate": 2.4888888888888886e-07,
      "loss": 0.0,
      "reward": 1.757142923772335,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 2016
    },
    {
      "completion_length": 220.9732255935669,
      "epoch": 0.3383209690263632,
      "grad_norm": 0.4744433084933257,
      "kl": 0.03647422790527344,
      "learning_rate": 2.491358024691358e-07,
      "loss": 0.0,
      "reward": 1.7857143506407738,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 2018
    },
    {
      "completion_length": 227.18304538726807,
      "epoch": 0.3386562722662308,
      "grad_norm": 0.16060134070242493,
      "kl": 0.05447578430175781,
      "learning_rate": 2.493827160493827e-07,
      "loss": 0.0001,
      "reward": 1.7357143685221672,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 2020
    },
    {
      "completion_length": 230.07143783569336,
      "epoch": 0.3389915755060983,
      "grad_norm": 0.349626308958822,
      "kl": 0.09692955017089844,
      "learning_rate": 2.4962962962962963e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 2022
    },
    {
      "completion_length": 227.34375953674316,
      "epoch": 0.3393268787459659,
      "grad_norm": 0.229310222683427,
      "kl": 0.022785186767578125,
      "learning_rate": 2.498765432098765e-07,
      "loss": 0.0,
      "reward": 1.7803572192788124,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214562982321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2024
    },
    {
      "completion_length": 224.8839406967163,
      "epoch": 0.3396621819858334,
      "grad_norm": 0.13934129466630318,
      "kl": 0.05133819580078125,
      "learning_rate": 2.501234567901234e-07,
      "loss": 0.0001,
      "reward": 1.7464286461472511,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 2026
    },
    {
      "completion_length": 229.83036708831787,
      "epoch": 0.339997485225701,
      "grad_norm": 0.11982324549628261,
      "kl": 0.03446388244628906,
      "learning_rate": 2.5037037037037035e-07,
      "loss": 0.0,
      "reward": 1.7642857804894447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 2028
    },
    {
      "completion_length": 220.665189743042,
      "epoch": 0.3403327884655685,
      "grad_norm": 0.14265061273334256,
      "kl": 0.014574050903320312,
      "learning_rate": 2.506172839506173e-07,
      "loss": 0.0,
      "reward": 1.8107143267989159,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 2030
    },
    {
      "completion_length": 228.60715293884277,
      "epoch": 0.3406680917054361,
      "grad_norm": 0.08317394544730386,
      "kl": 0.05511188507080078,
      "learning_rate": 2.508641975308642e-07,
      "loss": 0.0001,
      "reward": 1.6892857998609543,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.689285745844245,
      "rewards/format_reward_func": 1.0,
      "step": 2032
    },
    {
      "completion_length": 224.03572368621826,
      "epoch": 0.3410033949453037,
      "grad_norm": 0.13839855068066612,
      "kl": 0.013246536254882812,
      "learning_rate": 2.511111111111111e-07,
      "loss": 0.0,
      "reward": 1.7714286148548126,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 2034
    },
    {
      "completion_length": 236.71876335144043,
      "epoch": 0.3413386981851712,
      "grad_norm": 0.46152422665186366,
      "kl": 0.03157615661621094,
      "learning_rate": 2.51358024691358e-07,
      "loss": 0.0,
      "reward": 1.7232143431901932,
      "reward_std": 0.10859139822423458,
      "rewards/equation_reward_func": 0.7366071827709675,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2036
    },
    {
      "completion_length": 219.82143783569336,
      "epoch": 0.3416740014250388,
      "grad_norm": 0.2956090340920406,
      "kl": 0.007686614990234375,
      "learning_rate": 2.5160493827160495e-07,
      "loss": 0.0,
      "reward": 1.792857214808464,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 2038
    },
    {
      "completion_length": 237.73215293884277,
      "epoch": 0.3420093046649063,
      "grad_norm": 0.3170164235407117,
      "kl": 0.04242229461669922,
      "learning_rate": 2.5185185185185184e-07,
      "loss": 0.0,
      "reward": 1.7392857819795609,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857596278191,
      "rewards/format_reward_func": 1.0,
      "step": 2040
    },
    {
      "completion_length": 229.84375858306885,
      "epoch": 0.3423446079047739,
      "grad_norm": 0.18831475868217637,
      "kl": 0.022202491760253906,
      "learning_rate": 2.520987654320987e-07,
      "loss": 0.0,
      "reward": 1.7785715162754059,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 1.0,
      "step": 2042
    },
    {
      "completion_length": 225.55358219146729,
      "epoch": 0.3426799111446414,
      "grad_norm": 0.2642266312318678,
      "kl": 0.07865715026855469,
      "learning_rate": 2.5234567901234567e-07,
      "loss": 0.0001,
      "reward": 1.7589286267757416,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7633928880095482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2044
    },
    {
      "completion_length": 235.05357933044434,
      "epoch": 0.343015214384509,
      "grad_norm": 0.2661198557562209,
      "kl": 0.028797149658203125,
      "learning_rate": 2.5259259259259255e-07,
      "loss": 0.0,
      "reward": 1.7446429207921028,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7491071708500385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2046
    },
    {
      "completion_length": 236.3259048461914,
      "epoch": 0.34335051762437657,
      "grad_norm": 0.3349485363833583,
      "kl": 0.029720306396484375,
      "learning_rate": 2.528395061728395e-07,
      "loss": 0.0,
      "reward": 1.7357143759727478,
      "reward_std": 0.09091372601687908,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 2048
    },
    {
      "completion_length": 231.45983123779297,
      "epoch": 0.3436858208642441,
      "grad_norm": 0.4207071809083083,
      "kl": 0.06764984130859375,
      "learning_rate": 2.5308641975308644e-07,
      "loss": 0.0001,
      "reward": 1.789285771548748,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 2050
    },
    {
      "completion_length": 229.0580472946167,
      "epoch": 0.3440211241041117,
      "grad_norm": 0.2937621352389509,
      "kl": 0.02901458740234375,
      "learning_rate": 2.533333333333333e-07,
      "loss": 0.0,
      "reward": 1.7321429401636124,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428880095482,
      "rewards/format_reward_func": 1.0,
      "step": 2052
    },
    {
      "completion_length": 232.07590198516846,
      "epoch": 0.3443564273439792,
      "grad_norm": 0.291425140305454,
      "kl": 0.015499114990234375,
      "learning_rate": 2.535802469135802e-07,
      "loss": 0.0,
      "reward": 1.714285783469677,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7142857387661934,
      "rewards/format_reward_func": 1.0,
      "step": 2054
    },
    {
      "completion_length": 234.2053680419922,
      "epoch": 0.3446917305838468,
      "grad_norm": 0.37230523225702566,
      "kl": 0.01446533203125,
      "learning_rate": 2.5382716049382716e-07,
      "loss": 0.0,
      "reward": 1.760714367032051,
      "reward_std": 0.09596448857337236,
      "rewards/equation_reward_func": 0.7607143223285675,
      "rewards/format_reward_func": 1.0,
      "step": 2056
    },
    {
      "completion_length": 231.3526906967163,
      "epoch": 0.3450270338237143,
      "grad_norm": 0.13254691883933248,
      "kl": 0.015127182006835938,
      "learning_rate": 2.5407407407407404e-07,
      "loss": 0.0,
      "reward": 1.8571428805589676,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.8571428842842579,
      "rewards/format_reward_func": 1.0,
      "step": 2058
    },
    {
      "completion_length": 224.11161708831787,
      "epoch": 0.3453623370635819,
      "grad_norm": 0.2855218059164058,
      "kl": 0.021585464477539062,
      "learning_rate": 2.54320987654321e-07,
      "loss": 0.0,
      "reward": 1.7392858043313026,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 1.0,
      "step": 2060
    },
    {
      "completion_length": 230.78572463989258,
      "epoch": 0.3456976403034494,
      "grad_norm": 0.2911414082155755,
      "kl": 0.01221466064453125,
      "learning_rate": 2.5456790123456787e-07,
      "loss": 0.0,
      "reward": 1.7321429550647736,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321428991854191,
      "rewards/format_reward_func": 1.0,
      "step": 2062
    },
    {
      "completion_length": 233.9642972946167,
      "epoch": 0.346032943543317,
      "grad_norm": 0.3881993855084975,
      "kl": 0.0128936767578125,
      "learning_rate": 2.548148148148148e-07,
      "loss": 0.0,
      "reward": 1.7785715237259865,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7785714380443096,
      "rewards/format_reward_func": 1.0,
      "step": 2064
    },
    {
      "completion_length": 221.6384038925171,
      "epoch": 0.34636824678318456,
      "grad_norm": 0.26392784790783114,
      "kl": 0.0257110595703125,
      "learning_rate": 2.5506172839506176e-07,
      "loss": 0.0,
      "reward": 1.792857214808464,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 2066
    },
    {
      "completion_length": 227.79911613464355,
      "epoch": 0.3467035500230521,
      "grad_norm": 0.32096854524746926,
      "kl": 0.026147842407226562,
      "learning_rate": 2.553086419753086e-07,
      "loss": 0.0,
      "reward": 1.76071435213089,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143204659224,
      "rewards/format_reward_func": 1.0,
      "step": 2068
    },
    {
      "completion_length": 235.91072750091553,
      "epoch": 0.34703885326291967,
      "grad_norm": 0.4583612266863497,
      "kl": 0.03256988525390625,
      "learning_rate": 2.5555555555555553e-07,
      "loss": 0.0,
      "reward": 1.8196429461240768,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8241071552038193,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2070
    },
    {
      "completion_length": 228.0000114440918,
      "epoch": 0.3473741565027872,
      "grad_norm": 0.18039195795853832,
      "kl": 0.03345489501953125,
      "learning_rate": 2.558024691358024e-07,
      "loss": 0.0,
      "reward": 1.7500000670552254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000242143869,
      "rewards/format_reward_func": 1.0,
      "step": 2072
    },
    {
      "completion_length": 228.9687614440918,
      "epoch": 0.34770945974265477,
      "grad_norm": 0.19847492504938696,
      "kl": 0.011264801025390625,
      "learning_rate": 2.5604938271604936e-07,
      "loss": 0.0,
      "reward": 1.7736607789993286,
      "reward_std": 0.027147850021719933,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 2074
    },
    {
      "completion_length": 233.9241189956665,
      "epoch": 0.3480447629825223,
      "grad_norm": 0.21015001646006334,
      "kl": 0.009357452392578125,
      "learning_rate": 2.562962962962963e-07,
      "loss": 0.0,
      "reward": 1.7732143476605415,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7776785977184772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2076
    },
    {
      "completion_length": 232.8482255935669,
      "epoch": 0.3483800662223899,
      "grad_norm": 0.33872680822882006,
      "kl": 0.022771835327148438,
      "learning_rate": 2.565432098765432e-07,
      "loss": 0.0,
      "reward": 1.8000000640749931,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.8089285977184772,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2078
    },
    {
      "completion_length": 229.36161708831787,
      "epoch": 0.34871536946225745,
      "grad_norm": 0.28679987177075117,
      "kl": 0.054790496826171875,
      "learning_rate": 2.5679012345679013e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 2080
    },
    {
      "completion_length": 217.6384038925171,
      "epoch": 0.349050672702125,
      "grad_norm": 0.27843672987692414,
      "kl": 0.00815582275390625,
      "learning_rate": 2.570370370370371e-07,
      "loss": 0.0,
      "reward": 1.7857143431901932,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 2082
    },
    {
      "completion_length": 237.37501049041748,
      "epoch": 0.34938597594199255,
      "grad_norm": 0.37663868418897395,
      "kl": 0.09327888488769531,
      "learning_rate": 2.572839506172839e-07,
      "loss": 0.0001,
      "reward": 1.7714286148548126,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286185801029,
      "rewards/format_reward_func": 1.0,
      "step": 2084
    },
    {
      "completion_length": 229.415189743042,
      "epoch": 0.3497212791818601,
      "grad_norm": 0.2059930556278535,
      "kl": 0.07577705383300781,
      "learning_rate": 2.5753086419753085e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 2086
    },
    {
      "completion_length": 223.26786613464355,
      "epoch": 0.35005658242172766,
      "grad_norm": 0.26716203101438885,
      "kl": 0.01068878173828125,
      "learning_rate": 2.5777777777777774e-07,
      "loss": 0.0,
      "reward": 1.7482143640518188,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7526785992085934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2088
    },
    {
      "completion_length": 223.92858505249023,
      "epoch": 0.3503918856615952,
      "grad_norm": 0.5845833867815128,
      "kl": 0.16082000732421875,
      "learning_rate": 2.580246913580247e-07,
      "loss": 0.0002,
      "reward": 1.757142923772335,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.757142897695303,
      "rewards/format_reward_func": 1.0,
      "step": 2090
    },
    {
      "completion_length": 230.1384038925171,
      "epoch": 0.35072718890146276,
      "grad_norm": 0.30983776550084197,
      "kl": 0.0106658935546875,
      "learning_rate": 2.582716049382716e-07,
      "loss": 0.0,
      "reward": 1.76071435213089,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7607143260538578,
      "rewards/format_reward_func": 1.0,
      "step": 2092
    },
    {
      "completion_length": 225.102689743042,
      "epoch": 0.35106249214133034,
      "grad_norm": 0.342229109593764,
      "kl": 0.019098281860351562,
      "learning_rate": 2.585185185185185e-07,
      "loss": 0.0,
      "reward": 1.778571479022503,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7785714790225029,
      "rewards/format_reward_func": 1.0,
      "step": 2094
    },
    {
      "completion_length": 231.91518783569336,
      "epoch": 0.35139779538119786,
      "grad_norm": 0.34481677702260877,
      "kl": 0.04109764099121094,
      "learning_rate": 2.5876543209876545e-07,
      "loss": 0.0,
      "reward": 1.7464286535978317,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.746428620070219,
      "rewards/format_reward_func": 1.0,
      "step": 2096
    },
    {
      "completion_length": 233.06697368621826,
      "epoch": 0.35173309862106544,
      "grad_norm": 0.31827473583036603,
      "kl": 0.01319122314453125,
      "learning_rate": 2.590123456790123e-07,
      "loss": 0.0,
      "reward": 1.769642911851406,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.7741071656346321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2098
    },
    {
      "completion_length": 230.33036708831787,
      "epoch": 0.35206840186093297,
      "grad_norm": 0.351989871127145,
      "kl": 0.07884025573730469,
      "learning_rate": 2.5925925925925923e-07,
      "loss": 0.0001,
      "reward": 1.7285715192556381,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7285714540630579,
      "rewards/format_reward_func": 1.0,
      "step": 2100
    },
    {
      "completion_length": 220.97768783569336,
      "epoch": 0.35240370510080055,
      "grad_norm": 0.09114508670738188,
      "kl": 0.024440765380859375,
      "learning_rate": 2.5950617283950617e-07,
      "loss": 0.0,
      "reward": 1.8107143566012383,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143044471741,
      "rewards/format_reward_func": 1.0,
      "step": 2102
    },
    {
      "completion_length": 232.7946538925171,
      "epoch": 0.35273900834066807,
      "grad_norm": 0.2583027938351751,
      "kl": 0.04311180114746094,
      "learning_rate": 2.5975308641975306e-07,
      "loss": 0.0,
      "reward": 1.7607143223285675,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7696428950875998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2104
    },
    {
      "completion_length": 225.14286613464355,
      "epoch": 0.35307431158053565,
      "grad_norm": 0.2558472070866311,
      "kl": 0.009954452514648438,
      "learning_rate": 2.6e-07,
      "loss": 0.0,
      "reward": 1.8250000402331352,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8250000383704901,
      "rewards/format_reward_func": 1.0,
      "step": 2106
    },
    {
      "completion_length": 230.54465198516846,
      "epoch": 0.35340961482040323,
      "grad_norm": 0.37407279058475107,
      "kl": 0.0380706787109375,
      "learning_rate": 2.6024691358024694e-07,
      "loss": 0.0,
      "reward": 1.7214286401867867,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7214286141097546,
      "rewards/format_reward_func": 1.0,
      "step": 2108
    },
    {
      "completion_length": 221.95536518096924,
      "epoch": 0.35374491806027075,
      "grad_norm": 0.16814010224897627,
      "kl": 0.08795928955078125,
      "learning_rate": 2.6049382716049383e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 2110
    },
    {
      "completion_length": 228.25893783569336,
      "epoch": 0.35408022130013833,
      "grad_norm": 0.45033267511988145,
      "kl": 0.026628494262695312,
      "learning_rate": 2.607407407407407e-07,
      "loss": 0.0,
      "reward": 1.730357214808464,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7348214685916901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2112
    },
    {
      "completion_length": 234.99108219146729,
      "epoch": 0.35441552454000586,
      "grad_norm": 0.13415719938476556,
      "kl": 0.062084197998046875,
      "learning_rate": 2.609876543209876e-07,
      "loss": 0.0001,
      "reward": 1.7500000894069672,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 2114
    },
    {
      "completion_length": 224.31697368621826,
      "epoch": 0.35475082777987343,
      "grad_norm": 0.4264411382430659,
      "kl": 0.043720245361328125,
      "learning_rate": 2.6123456790123455e-07,
      "loss": 0.0,
      "reward": 1.7303572073578835,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7348214592784643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2116
    },
    {
      "completion_length": 226.071439743042,
      "epoch": 0.35508613101974096,
      "grad_norm": 0.40260476168590575,
      "kl": 0.015716552734375,
      "learning_rate": 2.614814814814815e-07,
      "loss": 0.0,
      "reward": 1.7392858043313026,
      "reward_std": 0.09596448857337236,
      "rewards/equation_reward_func": 0.7392857521772385,
      "rewards/format_reward_func": 1.0,
      "step": 2118
    },
    {
      "completion_length": 233.508939743042,
      "epoch": 0.35542143425960854,
      "grad_norm": 0.24759868181676284,
      "kl": 0.024053573608398438,
      "learning_rate": 2.617283950617284e-07,
      "loss": 0.0,
      "reward": 1.7571429312229156,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 1.0,
      "step": 2120
    },
    {
      "completion_length": 224.0267972946167,
      "epoch": 0.35575673749947606,
      "grad_norm": 0.10767464037699406,
      "kl": 0.14308738708496094,
      "learning_rate": 2.619753086419753e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428775787354,
      "rewards/format_reward_func": 1.0,
      "step": 2122
    },
    {
      "completion_length": 231.52232933044434,
      "epoch": 0.35609204073934364,
      "grad_norm": 0.3298464882369941,
      "kl": 0.01337432861328125,
      "learning_rate": 2.6222222222222226e-07,
      "loss": 0.0,
      "reward": 1.7232143580913544,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7276786006987095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2124
    },
    {
      "completion_length": 223.696439743042,
      "epoch": 0.3564273439792112,
      "grad_norm": 0.24994278995491304,
      "kl": 0.025163650512695312,
      "learning_rate": 2.624691358024691e-07,
      "loss": 0.0,
      "reward": 1.7928571924567223,
      "reward_std": 0.06060914974659681,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 2126
    },
    {
      "completion_length": 224.03125953674316,
      "epoch": 0.35676264721907874,
      "grad_norm": 0.423221266506577,
      "kl": 0.08066177368164062,
      "learning_rate": 2.6271604938271604e-07,
      "loss": 0.0001,
      "reward": 1.7214286625385284,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7214285973459482,
      "rewards/format_reward_func": 1.0,
      "step": 2128
    },
    {
      "completion_length": 231.25447273254395,
      "epoch": 0.3570979504589463,
      "grad_norm": 0.22862671789859407,
      "kl": 0.025022506713867188,
      "learning_rate": 2.629629629629629e-07,
      "loss": 0.0,
      "reward": 1.7303571924567223,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.734821455553174,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2130
    },
    {
      "completion_length": 226.540189743042,
      "epoch": 0.35743325369881385,
      "grad_norm": 0.19159948818631775,
      "kl": 0.011707305908203125,
      "learning_rate": 2.6320987654320986e-07,
      "loss": 0.0,
      "reward": 1.7428572326898575,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571693599224,
      "rewards/format_reward_func": 1.0,
      "step": 2132
    },
    {
      "completion_length": 221.31697463989258,
      "epoch": 0.3577685569386814,
      "grad_norm": 0.29371425581013727,
      "kl": 0.04029273986816406,
      "learning_rate": 2.634567901234568e-07,
      "loss": 0.0,
      "reward": 1.7785714864730835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 2134
    },
    {
      "completion_length": 228.50447463989258,
      "epoch": 0.35810386017854895,
      "grad_norm": 0.12820217400008221,
      "kl": 0.10086727142333984,
      "learning_rate": 2.637037037037037e-07,
      "loss": 0.0001,
      "reward": 1.8107143491506577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8107143007218838,
      "rewards/format_reward_func": 1.0,
      "step": 2136
    },
    {
      "completion_length": 231.02233219146729,
      "epoch": 0.35843916341841653,
      "grad_norm": 0.23459826960008642,
      "kl": 0.043140411376953125,
      "learning_rate": 2.6395061728395064e-07,
      "loss": 0.0,
      "reward": 1.7964286357164383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964286096394062,
      "rewards/format_reward_func": 1.0,
      "step": 2138
    },
    {
      "completion_length": 221.94643878936768,
      "epoch": 0.3587744666582841,
      "grad_norm": 0.3008130333023876,
      "kl": 0.010679244995117188,
      "learning_rate": 2.6419753086419747e-07,
      "loss": 0.0,
      "reward": 1.7750000730156898,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 2140
    },
    {
      "completion_length": 220.62947273254395,
      "epoch": 0.35910976989815163,
      "grad_norm": 0.32973760375580985,
      "kl": 0.09985923767089844,
      "learning_rate": 2.644444444444444e-07,
      "loss": 0.0001,
      "reward": 1.7928571924567223,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571850061417,
      "rewards/format_reward_func": 1.0,
      "step": 2142
    },
    {
      "completion_length": 228.62501049041748,
      "epoch": 0.3594450731380192,
      "grad_norm": 0.48222064013341237,
      "kl": 0.10888290405273438,
      "learning_rate": 2.6469135802469135e-07,
      "loss": 0.0001,
      "reward": 1.7000000700354576,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7000000458210707,
      "rewards/format_reward_func": 1.0,
      "step": 2144
    },
    {
      "completion_length": 229.02233219146729,
      "epoch": 0.35978037637788673,
      "grad_norm": 0.23380887912828222,
      "kl": 0.08852958679199219,
      "learning_rate": 2.6493827160493824e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857670783997,
      "rewards/format_reward_func": 1.0,
      "step": 2146
    },
    {
      "completion_length": 228.0669755935669,
      "epoch": 0.3601156796177543,
      "grad_norm": 0.32278175684584354,
      "kl": 0.17383193969726562,
      "learning_rate": 2.651851851851852e-07,
      "loss": 0.0002,
      "reward": 1.735714353621006,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 2148
    },
    {
      "completion_length": 227.0759048461914,
      "epoch": 0.36045098285762184,
      "grad_norm": 0.26353297086437466,
      "kl": 0.0684661865234375,
      "learning_rate": 2.654320987654321e-07,
      "loss": 0.0001,
      "reward": 1.803571492433548,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714477300644,
      "rewards/format_reward_func": 1.0,
      "step": 2150
    },
    {
      "completion_length": 228.1428680419922,
      "epoch": 0.3607862860974894,
      "grad_norm": 0.22570297887110932,
      "kl": 0.05149078369140625,
      "learning_rate": 2.65679012345679e-07,
      "loss": 0.0001,
      "reward": 1.7000000774860382,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7000000402331352,
      "rewards/format_reward_func": 1.0,
      "step": 2152
    },
    {
      "completion_length": 220.92858028411865,
      "epoch": 0.361121589337357,
      "grad_norm": 0.3708480771834063,
      "kl": 0.08710098266601562,
      "learning_rate": 2.659259259259259e-07,
      "loss": 0.0001,
      "reward": 1.84285718947649,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8428571671247482,
      "rewards/format_reward_func": 1.0,
      "step": 2154
    },
    {
      "completion_length": 235.52679443359375,
      "epoch": 0.3614568925772245,
      "grad_norm": 0.1637799996002816,
      "kl": 0.05851173400878906,
      "learning_rate": 2.661728395061728e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.767857164144516,
      "rewards/format_reward_func": 1.0,
      "step": 2156
    },
    {
      "completion_length": 226.90626049041748,
      "epoch": 0.3617921958170921,
      "grad_norm": 0.3360529398155563,
      "kl": 0.021284103393554688,
      "learning_rate": 2.6641975308641973e-07,
      "loss": 0.0,
      "reward": 1.7821429371833801,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7821428962051868,
      "rewards/format_reward_func": 1.0,
      "step": 2158
    },
    {
      "completion_length": 233.8973331451416,
      "epoch": 0.3621274990569596,
      "grad_norm": 0.31128516923824884,
      "kl": 0.15531539916992188,
      "learning_rate": 2.6666666666666667e-07,
      "loss": 0.0002,
      "reward": 1.764285758137703,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 2160
    },
    {
      "completion_length": 223.227689743042,
      "epoch": 0.3624628022968272,
      "grad_norm": 0.23111936609267744,
      "kl": 0.038791656494140625,
      "learning_rate": 2.6691358024691356e-07,
      "loss": 0.0,
      "reward": 1.817857176065445,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8178571686148643,
      "rewards/format_reward_func": 1.0,
      "step": 2162
    },
    {
      "completion_length": 221.3035831451416,
      "epoch": 0.3627981055366947,
      "grad_norm": 0.4127253192986052,
      "kl": 0.04503440856933594,
      "learning_rate": 2.671604938271605e-07,
      "loss": 0.0,
      "reward": 1.7821429446339607,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 2164
    },
    {
      "completion_length": 221.6428680419922,
      "epoch": 0.3631334087765623,
      "grad_norm": 0.223704776342276,
      "kl": 0.061389923095703125,
      "learning_rate": 2.674074074074074e-07,
      "loss": 0.0001,
      "reward": 1.7428572252392769,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7428571712225676,
      "rewards/format_reward_func": 1.0,
      "step": 2166
    },
    {
      "completion_length": 228.24108409881592,
      "epoch": 0.3634687120164299,
      "grad_norm": 0.22441233228251525,
      "kl": 0.03566169738769531,
      "learning_rate": 2.6765432098765433e-07,
      "loss": 0.0,
      "reward": 1.7410714998841286,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7455357387661934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2168
    },
    {
      "completion_length": 225.6830472946167,
      "epoch": 0.3638040152562974,
      "grad_norm": 0.13680117104986803,
      "kl": 0.015272140502929688,
      "learning_rate": 2.679012345679012e-07,
      "loss": 0.0,
      "reward": 1.7892857789993286,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 2170
    },
    {
      "completion_length": 222.26340293884277,
      "epoch": 0.364139318496165,
      "grad_norm": 0.26671750198225047,
      "kl": 0.02013397216796875,
      "learning_rate": 2.681481481481481e-07,
      "loss": 0.0,
      "reward": 1.7214286550879478,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.721428606659174,
      "rewards/format_reward_func": 1.0,
      "step": 2172
    },
    {
      "completion_length": 221.37947463989258,
      "epoch": 0.3644746217360325,
      "grad_norm": 0.2374098871508374,
      "kl": 0.06538772583007812,
      "learning_rate": 2.6839506172839505e-07,
      "loss": 0.0001,
      "reward": 1.7785715162754059,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 2174
    },
    {
      "completion_length": 237.63393878936768,
      "epoch": 0.3648099249759001,
      "grad_norm": 0.2467816451908553,
      "kl": 0.03937530517578125,
      "learning_rate": 2.68641975308642e-07,
      "loss": 0.0,
      "reward": 1.7446429431438446,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7491071708500385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2176
    },
    {
      "completion_length": 230.14733123779297,
      "epoch": 0.3651452282157676,
      "grad_norm": 0.3179348050690457,
      "kl": 0.01219940185546875,
      "learning_rate": 2.688888888888889e-07,
      "loss": 0.0,
      "reward": 1.8178572207689285,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.8178571499884129,
      "rewards/format_reward_func": 1.0,
      "step": 2178
    },
    {
      "completion_length": 228.75447368621826,
      "epoch": 0.3654805314556352,
      "grad_norm": 0.2525358576089782,
      "kl": 0.02394866943359375,
      "learning_rate": 2.691358024691358e-07,
      "loss": 0.0,
      "reward": 1.7964286282658577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286096394062,
      "rewards/format_reward_func": 1.0,
      "step": 2180
    },
    {
      "completion_length": 222.07590293884277,
      "epoch": 0.3658158346955028,
      "grad_norm": 0.2556722238310013,
      "kl": 0.022491455078125,
      "learning_rate": 2.693827160493827e-07,
      "loss": 0.0,
      "reward": 1.6678572222590446,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.6678571812808514,
      "rewards/format_reward_func": 1.0,
      "step": 2182
    },
    {
      "completion_length": 236.40179634094238,
      "epoch": 0.3661511379353703,
      "grad_norm": 0.29412853954255963,
      "kl": 0.07938003540039062,
      "learning_rate": 2.696296296296296e-07,
      "loss": 0.0001,
      "reward": 1.7571429461240768,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7571429051458836,
      "rewards/format_reward_func": 1.0,
      "step": 2184
    },
    {
      "completion_length": 234.1428689956665,
      "epoch": 0.3664864411752379,
      "grad_norm": 0.2060815902787529,
      "kl": 0.10279083251953125,
      "learning_rate": 2.6987654320987654e-07,
      "loss": 0.0001,
      "reward": 1.698214367032051,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7026786133646965,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2186
    },
    {
      "completion_length": 227.27233219146729,
      "epoch": 0.3668217444151054,
      "grad_norm": 0.3214990798853585,
      "kl": 0.06783485412597656,
      "learning_rate": 2.7012345679012343e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7750000394880772,
      "rewards/format_reward_func": 1.0,
      "step": 2188
    },
    {
      "completion_length": 232.0803680419922,
      "epoch": 0.367157047654973,
      "grad_norm": 0.29116456137569957,
      "kl": 0.06546974182128906,
      "learning_rate": 2.7037037037037037e-07,
      "loss": 0.0001,
      "reward": 1.8035714849829674,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714440047741,
      "rewards/format_reward_func": 1.0,
      "step": 2190
    },
    {
      "completion_length": 236.8259048461914,
      "epoch": 0.3674923508948405,
      "grad_norm": 0.275873402245631,
      "kl": 0.3006572723388672,
      "learning_rate": 2.7061728395061726e-07,
      "loss": 0.0003,
      "reward": 1.7125000655651093,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7169643267989159,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2192
    },
    {
      "completion_length": 235.76786613464355,
      "epoch": 0.3678276541347081,
      "grad_norm": 0.37498091593844435,
      "kl": 0.12526512145996094,
      "learning_rate": 2.708641975308642e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714507102966,
      "rewards/format_reward_func": 1.0,
      "step": 2194
    },
    {
      "completion_length": 238.69197463989258,
      "epoch": 0.3681629573745756,
      "grad_norm": 0.19498387354928298,
      "kl": 0.19458770751953125,
      "learning_rate": 2.7111111111111114e-07,
      "loss": 0.0002,
      "reward": 1.7267857939004898,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7312500365078449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2196
    },
    {
      "completion_length": 232.6339406967163,
      "epoch": 0.3684982606144432,
      "grad_norm": 0.16930298305774352,
      "kl": 0.04517364501953125,
      "learning_rate": 2.71358024691358e-07,
      "loss": 0.0,
      "reward": 1.7821429148316383,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7910714522004128,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2198
    },
    {
      "completion_length": 225.65179538726807,
      "epoch": 0.36883356385431076,
      "grad_norm": 0.2732590185925259,
      "kl": 0.09507369995117188,
      "learning_rate": 2.716049382716049e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571697324514,
      "rewards/format_reward_func": 1.0,
      "step": 2200
    },
    {
      "completion_length": 220.92858028411865,
      "epoch": 0.3691688670941783,
      "grad_norm": 0.3480826700798807,
      "kl": 0.2006244659423828,
      "learning_rate": 2.7185185185185186e-07,
      "loss": 0.0002,
      "reward": 1.7571429386734962,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 2202
    },
    {
      "completion_length": 228.95983028411865,
      "epoch": 0.36950417033404587,
      "grad_norm": 0.23230064612406418,
      "kl": 0.09466743469238281,
      "learning_rate": 2.7209876543209875e-07,
      "loss": 0.0001,
      "reward": 1.7410715073347092,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357499420643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2204
    },
    {
      "completion_length": 225.86161613464355,
      "epoch": 0.3698394735739134,
      "grad_norm": 0.37423496124042877,
      "kl": 0.07233428955078125,
      "learning_rate": 2.723456790123457e-07,
      "loss": 0.0001,
      "reward": 1.7000000849366188,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7000000383704901,
      "rewards/format_reward_func": 1.0,
      "step": 2206
    },
    {
      "completion_length": 235.37501049041748,
      "epoch": 0.37017477681378097,
      "grad_norm": 0.20144566733771846,
      "kl": 0.08749961853027344,
      "learning_rate": 2.725925925925926e-07,
      "loss": 0.0001,
      "reward": 1.7232143506407738,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7276786044239998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2208
    },
    {
      "completion_length": 231.55804538726807,
      "epoch": 0.3705100800536485,
      "grad_norm": 0.9344322170031661,
      "kl": 0.2306499481201172,
      "learning_rate": 2.728395061728395e-07,
      "loss": 0.0002,
      "reward": 1.7214286476373672,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7214286029338837,
      "rewards/format_reward_func": 1.0,
      "step": 2210
    },
    {
      "completion_length": 232.2946538925171,
      "epoch": 0.3708453832935161,
      "grad_norm": 0.1253578302593943,
      "kl": 0.07514762878417969,
      "learning_rate": 2.730864197530864e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.753571467474103,
      "rewards/format_reward_func": 1.0,
      "step": 2212
    },
    {
      "completion_length": 232.62054824829102,
      "epoch": 0.37118068653338365,
      "grad_norm": 0.20158043710907292,
      "kl": 0.1405200958251953,
      "learning_rate": 2.733333333333333e-07,
      "loss": 0.0001,
      "reward": 1.7500000521540642,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7589286118745804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2214
    },
    {
      "completion_length": 231.17411518096924,
      "epoch": 0.3715159897732512,
      "grad_norm": 0.5074480366301922,
      "kl": 0.08559036254882812,
      "learning_rate": 2.7358024691358023e-07,
      "loss": 0.0001,
      "reward": 1.7500000894069672,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 2216
    },
    {
      "completion_length": 228.22768783569336,
      "epoch": 0.37185129301311876,
      "grad_norm": 0.16069692799159083,
      "kl": 0.05332183837890625,
      "learning_rate": 2.738271604938271e-07,
      "loss": 0.0001,
      "reward": 1.7714286595582962,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 2218
    },
    {
      "completion_length": 233.99108219146729,
      "epoch": 0.3721865962529863,
      "grad_norm": 0.45528804532488343,
      "kl": 0.2368755340576172,
      "learning_rate": 2.7407407407407406e-07,
      "loss": 0.0002,
      "reward": 1.757142923772335,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 2220
    },
    {
      "completion_length": 226.79465198516846,
      "epoch": 0.37252189949285386,
      "grad_norm": 0.4392680631342267,
      "kl": 0.0537261962890625,
      "learning_rate": 2.74320987654321e-07,
      "loss": 0.0001,
      "reward": 1.7714286148548126,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 2222
    },
    {
      "completion_length": 240.25000858306885,
      "epoch": 0.3728572027327214,
      "grad_norm": 0.42132753666272715,
      "kl": 0.18262672424316406,
      "learning_rate": 2.745679012345679e-07,
      "loss": 0.0002,
      "reward": 1.7696429044008255,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7741071656346321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2224
    },
    {
      "completion_length": 227.9419755935669,
      "epoch": 0.37319250597258896,
      "grad_norm": 0.29602895586363986,
      "kl": 0.07367324829101562,
      "learning_rate": 2.748148148148148e-07,
      "loss": 0.0001,
      "reward": 1.7571429163217545,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 2226
    },
    {
      "completion_length": 222.40179443359375,
      "epoch": 0.37352780921245654,
      "grad_norm": 0.18568951191388172,
      "kl": 0.08404731750488281,
      "learning_rate": 2.750617283950617e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714469850063,
      "rewards/format_reward_func": 1.0,
      "step": 2228
    },
    {
      "completion_length": 230.79018878936768,
      "epoch": 0.37386311245232406,
      "grad_norm": 0.2065159293766093,
      "kl": 0.041645050048828125,
      "learning_rate": 2.753086419753086e-07,
      "loss": 0.0,
      "reward": 1.796428643167019,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.796428594738245,
      "rewards/format_reward_func": 1.0,
      "step": 2230
    },
    {
      "completion_length": 224.0625114440918,
      "epoch": 0.37419841569219164,
      "grad_norm": 0.33197065162865186,
      "kl": 0.0833282470703125,
      "learning_rate": 2.7555555555555555e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.782142873853445,
      "rewards/format_reward_func": 1.0,
      "step": 2232
    },
    {
      "completion_length": 235.4776906967163,
      "epoch": 0.37453371893205917,
      "grad_norm": 0.25584059088231476,
      "kl": 0.040119171142578125,
      "learning_rate": 2.7580246913580244e-07,
      "loss": 0.0,
      "reward": 1.7107143625617027,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7107143253087997,
      "rewards/format_reward_func": 1.0,
      "step": 2234
    },
    {
      "completion_length": 231.67411708831787,
      "epoch": 0.37486902217192675,
      "grad_norm": 0.35658627996828857,
      "kl": 0.6641654968261719,
      "learning_rate": 2.760493827160494e-07,
      "loss": 0.0007,
      "reward": 1.7589286491274834,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928917348385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2236
    },
    {
      "completion_length": 231.4196548461914,
      "epoch": 0.37520432541179427,
      "grad_norm": 0.36641847714887776,
      "kl": 0.057590484619140625,
      "learning_rate": 2.762962962962963e-07,
      "loss": 0.0001,
      "reward": 1.7750000730156898,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 2238
    },
    {
      "completion_length": 228.5223331451416,
      "epoch": 0.37553962865166185,
      "grad_norm": 0.24546364124161799,
      "kl": 0.059139251708984375,
      "learning_rate": 2.7654320987654316e-07,
      "loss": 0.0001,
      "reward": 1.7589286267757416,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7633928898721933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2240
    },
    {
      "completion_length": 233.8750123977661,
      "epoch": 0.37587493189152943,
      "grad_norm": 0.2251870227230481,
      "kl": 0.18067550659179688,
      "learning_rate": 2.767901234567901e-07,
      "loss": 0.0002,
      "reward": 1.7196429297327995,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7241071723401546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2242
    },
    {
      "completion_length": 225.83036613464355,
      "epoch": 0.37621023513139695,
      "grad_norm": 0.18177423931815112,
      "kl": 0.13021469116210938,
      "learning_rate": 2.77037037037037e-07,
      "loss": 0.0001,
      "reward": 1.7321429178118706,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7321428898721933,
      "rewards/format_reward_func": 1.0,
      "step": 2244
    },
    {
      "completion_length": 224.49108028411865,
      "epoch": 0.37654553837126453,
      "grad_norm": 0.156748013206083,
      "kl": 0.1608562469482422,
      "learning_rate": 2.7728395061728393e-07,
      "loss": 0.0002,
      "reward": 1.7785714864730835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 2246
    },
    {
      "completion_length": 236.4464406967163,
      "epoch": 0.37688084161113206,
      "grad_norm": 0.25260715945203926,
      "kl": 0.2948493957519531,
      "learning_rate": 2.7753086419753087e-07,
      "loss": 0.0003,
      "reward": 1.6928572282195091,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.6928571742027998,
      "rewards/format_reward_func": 1.0,
      "step": 2248
    },
    {
      "completion_length": 235.70983409881592,
      "epoch": 0.37721614485099964,
      "grad_norm": 0.17868279769604858,
      "kl": 0.021900177001953125,
      "learning_rate": 2.7777777777777776e-07,
      "loss": 0.0,
      "reward": 1.7464286535978317,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464285902678967,
      "rewards/format_reward_func": 1.0,
      "step": 2250
    },
    {
      "completion_length": 228.9509038925171,
      "epoch": 0.37755144809086716,
      "grad_norm": 0.24309000324636804,
      "kl": 0.2515907287597656,
      "learning_rate": 2.780246913580247e-07,
      "loss": 0.0003,
      "reward": 1.7696429267525673,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.774107176810503,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2252
    },
    {
      "completion_length": 226.20983219146729,
      "epoch": 0.37788675133073474,
      "grad_norm": 0.2029313002323469,
      "kl": 0.1733074188232422,
      "learning_rate": 2.782716049382716e-07,
      "loss": 0.0002,
      "reward": 1.760714367032051,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.760714303702116,
      "rewards/format_reward_func": 1.0,
      "step": 2254
    },
    {
      "completion_length": 228.9732265472412,
      "epoch": 0.37822205457060226,
      "grad_norm": 0.3040059058343939,
      "kl": 0.12060165405273438,
      "learning_rate": 2.785185185185185e-07,
      "loss": 0.0001,
      "reward": 1.7392857819795609,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 1.0,
      "step": 2256
    },
    {
      "completion_length": 233.83036518096924,
      "epoch": 0.37855735781046984,
      "grad_norm": 0.3822294853412589,
      "kl": 0.17465782165527344,
      "learning_rate": 2.787654320987654e-07,
      "loss": 0.0002,
      "reward": 1.7446429133415222,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071783006191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2258
    },
    {
      "completion_length": 228.77233028411865,
      "epoch": 0.3788926610503374,
      "grad_norm": 0.3139393157242552,
      "kl": 0.5619621276855469,
      "learning_rate": 2.790123456790123e-07,
      "loss": 0.0006,
      "reward": 1.7410714849829674,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7455357499420643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2260
    },
    {
      "completion_length": 228.11608123779297,
      "epoch": 0.37922796429020494,
      "grad_norm": 0.1977537144990948,
      "kl": 0.09502792358398438,
      "learning_rate": 2.7925925925925925e-07,
      "loss": 0.0001,
      "reward": 1.7000000923871994,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.700000025331974,
      "rewards/format_reward_func": 1.0,
      "step": 2262
    },
    {
      "completion_length": 231.3035831451416,
      "epoch": 0.3795632675300725,
      "grad_norm": 0.262065947559945,
      "kl": 0.09780120849609375,
      "learning_rate": 2.795061728395062e-07,
      "loss": 0.0001,
      "reward": 1.7035714983940125,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7035714611411095,
      "rewards/format_reward_func": 1.0,
      "step": 2264
    },
    {
      "completion_length": 230.77679634094238,
      "epoch": 0.37989857076994005,
      "grad_norm": 0.37533314523360195,
      "kl": 0.018627166748046875,
      "learning_rate": 2.797530864197531e-07,
      "loss": 0.0,
      "reward": 1.7571429088711739,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 2266
    },
    {
      "completion_length": 229.883939743042,
      "epoch": 0.3802338740098076,
      "grad_norm": 0.2793404214719423,
      "kl": 0.12833023071289062,
      "learning_rate": 2.8e-07,
      "loss": 0.0001,
      "reward": 1.7553571835160255,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7598214708268642,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2268
    },
    {
      "completion_length": 223.883939743042,
      "epoch": 0.38056917724967515,
      "grad_norm": 0.12149878381355204,
      "kl": 0.021205902099609375,
      "learning_rate": 2.8024691358024685e-07,
      "loss": 0.0,
      "reward": 1.7714286148548126,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286148548126,
      "rewards/format_reward_func": 1.0,
      "step": 2270
    },
    {
      "completion_length": 225.12947463989258,
      "epoch": 0.38090448048954273,
      "grad_norm": 0.20368255516424766,
      "kl": 0.05774688720703125,
      "learning_rate": 2.804938271604938e-07,
      "loss": 0.0001,
      "reward": 1.6821429431438446,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.6821428947150707,
      "rewards/format_reward_func": 1.0,
      "step": 2272
    },
    {
      "completion_length": 229.54465293884277,
      "epoch": 0.3812397837294103,
      "grad_norm": 0.2472125611890382,
      "kl": 0.2638816833496094,
      "learning_rate": 2.8074074074074074e-07,
      "loss": 0.0003,
      "reward": 1.7392857819795609,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7482143081724644,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2274
    },
    {
      "completion_length": 224.26340293884277,
      "epoch": 0.38157508696927783,
      "grad_norm": 0.33149997659240854,
      "kl": 0.07962608337402344,
      "learning_rate": 2.809876543209876e-07,
      "loss": 0.0001,
      "reward": 1.7785715162754059,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7785714454948902,
      "rewards/format_reward_func": 1.0,
      "step": 2276
    },
    {
      "completion_length": 227.98661708831787,
      "epoch": 0.3819103902091454,
      "grad_norm": 0.3172056878755379,
      "kl": 0.09324264526367188,
      "learning_rate": 2.8123456790123457e-07,
      "loss": 0.0001,
      "reward": 1.7464286461472511,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 2278
    },
    {
      "completion_length": 225.8660831451416,
      "epoch": 0.38224569344901294,
      "grad_norm": 0.5301463048688589,
      "kl": 0.4278106689453125,
      "learning_rate": 2.814814814814815e-07,
      "loss": 0.0004,
      "reward": 1.74642863124609,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286163449287,
      "rewards/format_reward_func": 1.0,
      "step": 2280
    },
    {
      "completion_length": 229.35715293884277,
      "epoch": 0.3825809966888805,
      "grad_norm": 0.38356873327089946,
      "kl": 0.5712699890136719,
      "learning_rate": 2.817283950617284e-07,
      "loss": 0.0006,
      "reward": 1.7267857864499092,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.731250049546361,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2282
    },
    {
      "completion_length": 235.3884038925171,
      "epoch": 0.38291629992874804,
      "grad_norm": 0.934076897552111,
      "kl": 0.042568206787109375,
      "learning_rate": 2.819753086419753e-07,
      "loss": 0.0,
      "reward": 1.7892857566475868,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 2284
    },
    {
      "completion_length": 227.19643878936768,
      "epoch": 0.3832516031686156,
      "grad_norm": 0.16375574573485474,
      "kl": 0.04029083251953125,
      "learning_rate": 2.8222222222222217e-07,
      "loss": 0.0,
      "reward": 1.789285771548748,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 2286
    },
    {
      "completion_length": 227.34375953674316,
      "epoch": 0.3835869064084832,
      "grad_norm": 0.3301045316257568,
      "kl": 0.05418968200683594,
      "learning_rate": 2.824691358024691e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 2288
    },
    {
      "completion_length": 224.05804443359375,
      "epoch": 0.3839222096483507,
      "grad_norm": 0.2926858428633974,
      "kl": 0.021707534790039062,
      "learning_rate": 2.8271604938271606e-07,
      "loss": 0.0,
      "reward": 1.7642857804894447,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7732143104076385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2290
    },
    {
      "completion_length": 230.9553689956665,
      "epoch": 0.3842575128882183,
      "grad_norm": 0.32705642260561363,
      "kl": 0.13091659545898438,
      "learning_rate": 2.8296296296296294e-07,
      "loss": 0.0001,
      "reward": 1.6982143595814705,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7026786096394062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2292
    },
    {
      "completion_length": 229.50447273254395,
      "epoch": 0.3845928161280858,
      "grad_norm": 0.21564649196585978,
      "kl": 0.039546966552734375,
      "learning_rate": 2.832098765432099e-07,
      "loss": 0.0,
      "reward": 1.7464286759495735,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.746428593993187,
      "rewards/format_reward_func": 1.0,
      "step": 2294
    },
    {
      "completion_length": 219.53125953674316,
      "epoch": 0.3849281193679534,
      "grad_norm": 0.26692613221581457,
      "kl": 0.033473968505859375,
      "learning_rate": 2.834567901234568e-07,
      "loss": 0.0,
      "reward": 1.7785714864730835,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 1.0,
      "step": 2296
    },
    {
      "completion_length": 233.65179538726807,
      "epoch": 0.3852634226078209,
      "grad_norm": 0.2678121392019473,
      "kl": 0.06494903564453125,
      "learning_rate": 2.8370370370370366e-07,
      "loss": 0.0001,
      "reward": 1.7285715118050575,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714782774448,
      "rewards/format_reward_func": 1.0,
      "step": 2298
    },
    {
      "completion_length": 228.22322368621826,
      "epoch": 0.3855987258476885,
      "grad_norm": 0.39709523911952316,
      "kl": 0.040225982666015625,
      "learning_rate": 2.839506172839506e-07,
      "loss": 0.0,
      "reward": 1.7250000834465027,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7339286096394062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2300
    },
    {
      "completion_length": 217.65179634094238,
      "epoch": 0.3859340290875561,
      "grad_norm": 0.39208732176897304,
      "kl": 0.064727783203125,
      "learning_rate": 2.841975308641975e-07,
      "loss": 0.0001,
      "reward": 1.8071429133415222,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428798139095,
      "rewards/format_reward_func": 1.0,
      "step": 2302
    },
    {
      "completion_length": 229.55804920196533,
      "epoch": 0.3862693323274236,
      "grad_norm": 0.31110812623321255,
      "kl": 0.036376953125,
      "learning_rate": 2.8444444444444443e-07,
      "loss": 0.0,
      "reward": 1.8035714998841286,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8035714663565159,
      "rewards/format_reward_func": 1.0,
      "step": 2304
    },
    {
      "completion_length": 227.11161708831787,
      "epoch": 0.3866046355672912,
      "grad_norm": 0.2816325042184154,
      "kl": 0.028156280517578125,
      "learning_rate": 2.846913580246914e-07,
      "loss": 0.0,
      "reward": 1.7464286461472511,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 2306
    },
    {
      "completion_length": 216.4241189956665,
      "epoch": 0.3869399388071587,
      "grad_norm": 0.14324961565346786,
      "kl": 0.02130126953125,
      "learning_rate": 2.8493827160493826e-07,
      "loss": 0.0,
      "reward": 1.821428619325161,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.821428582072258,
      "rewards/format_reward_func": 1.0,
      "step": 2308
    },
    {
      "completion_length": 221.38393878936768,
      "epoch": 0.3872752420470263,
      "grad_norm": 0.26443223145301653,
      "kl": 0.026519775390625,
      "learning_rate": 2.851851851851852e-07,
      "loss": 0.0,
      "reward": 1.742857187986374,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571730852127,
      "rewards/format_reward_func": 1.0,
      "step": 2310
    },
    {
      "completion_length": 234.01340293884277,
      "epoch": 0.3876105452868938,
      "grad_norm": 0.16721530630211082,
      "kl": 0.02655792236328125,
      "learning_rate": 2.8543209876543204e-07,
      "loss": 0.0,
      "reward": 1.7142857909202576,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.714285746216774,
      "rewards/format_reward_func": 1.0,
      "step": 2312
    },
    {
      "completion_length": 236.1384048461914,
      "epoch": 0.3879458485267614,
      "grad_norm": 0.18018195259232184,
      "kl": 0.025665283203125,
      "learning_rate": 2.85679012345679e-07,
      "loss": 0.0,
      "reward": 1.7142857760190964,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7142857611179352,
      "rewards/format_reward_func": 1.0,
      "step": 2314
    },
    {
      "completion_length": 231.79465293884277,
      "epoch": 0.3882811517666289,
      "grad_norm": 0.2801403066500493,
      "kl": 0.034091949462890625,
      "learning_rate": 2.859259259259259e-07,
      "loss": 0.0,
      "reward": 1.7464286535978317,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.746428620070219,
      "rewards/format_reward_func": 1.0,
      "step": 2316
    },
    {
      "completion_length": 222.54465103149414,
      "epoch": 0.3886164550064965,
      "grad_norm": 0.19240949356448653,
      "kl": 0.026641845703125,
      "learning_rate": 2.861728395061728e-07,
      "loss": 0.0,
      "reward": 1.7821429073810577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 2318
    },
    {
      "completion_length": 227.81251335144043,
      "epoch": 0.3889517582463641,
      "grad_norm": 0.14613030642705596,
      "kl": 0.030071258544921875,
      "learning_rate": 2.8641975308641975e-07,
      "loss": 0.0,
      "reward": 1.7035714983940125,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7035714592784643,
      "rewards/format_reward_func": 1.0,
      "step": 2320
    },
    {
      "completion_length": 227.46876049041748,
      "epoch": 0.3892870614862316,
      "grad_norm": 0.39846413926302476,
      "kl": 0.030303955078125,
      "learning_rate": 2.866666666666667e-07,
      "loss": 0.0,
      "reward": 1.7767857685685158,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7812500223517418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2322
    },
    {
      "completion_length": 221.89733123779297,
      "epoch": 0.3896223647260992,
      "grad_norm": 0.41518256668271264,
      "kl": 0.040134429931640625,
      "learning_rate": 2.869135802469136e-07,
      "loss": 0.0,
      "reward": 1.7821429073810577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428775787354,
      "rewards/format_reward_func": 1.0,
      "step": 2324
    },
    {
      "completion_length": 234.54465579986572,
      "epoch": 0.3899576679659667,
      "grad_norm": 0.4979212066521887,
      "kl": 0.04520416259765625,
      "learning_rate": 2.8716049382716047e-07,
      "loss": 0.0,
      "reward": 1.7500000596046448,
      "reward_std": 0.09091372601687908,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 2326
    },
    {
      "completion_length": 225.66518783569336,
      "epoch": 0.3902929712058343,
      "grad_norm": 0.68471046288006,
      "kl": 0.06603240966796875,
      "learning_rate": 2.8740740740740736e-07,
      "loss": 0.0001,
      "reward": 1.7232143580913544,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7276786118745804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2328
    },
    {
      "completion_length": 230.5134038925171,
      "epoch": 0.3906282744457018,
      "grad_norm": 0.1734750680599172,
      "kl": 0.022487640380859375,
      "learning_rate": 2.876543209876543e-07,
      "loss": 0.0,
      "reward": 1.7642857879400253,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857357859612,
      "rewards/format_reward_func": 1.0,
      "step": 2330
    },
    {
      "completion_length": 226.7009038925171,
      "epoch": 0.3909635776855694,
      "grad_norm": 0.17619003683755471,
      "kl": 0.052631378173828125,
      "learning_rate": 2.8790123456790124e-07,
      "loss": 0.0001,
      "reward": 1.6821429207921028,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.682142898440361,
      "rewards/format_reward_func": 1.0,
      "step": 2332
    },
    {
      "completion_length": 223.3928680419922,
      "epoch": 0.39129888092543696,
      "grad_norm": 0.2130792522496975,
      "kl": 0.0273590087890625,
      "learning_rate": 2.8814814814814813e-07,
      "loss": 0.0,
      "reward": 1.7625000849366188,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643089175224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2334
    },
    {
      "completion_length": 230.91518878936768,
      "epoch": 0.3916341841653045,
      "grad_norm": 0.16563670418519555,
      "kl": 0.0242462158203125,
      "learning_rate": 2.8839506172839507e-07,
      "loss": 0.0,
      "reward": 1.7607143223285675,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 2336
    },
    {
      "completion_length": 223.83036708831787,
      "epoch": 0.39196948740517207,
      "grad_norm": 0.2537452887729266,
      "kl": 0.0279083251953125,
      "learning_rate": 2.8864197530864196e-07,
      "loss": 0.0,
      "reward": 1.7857143506407738,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 2338
    },
    {
      "completion_length": 233.96876430511475,
      "epoch": 0.3923047906450396,
      "grad_norm": 0.25310078074433373,
      "kl": 0.04262542724609375,
      "learning_rate": 2.8888888888888885e-07,
      "loss": 0.0,
      "reward": 1.8000000566244125,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 2340
    },
    {
      "completion_length": 231.4732255935669,
      "epoch": 0.39264009388490717,
      "grad_norm": 0.6108374179336262,
      "kl": 0.07244873046875,
      "learning_rate": 2.891358024691358e-07,
      "loss": 0.0001,
      "reward": 1.7750000804662704,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000096857548,
      "rewards/format_reward_func": 1.0,
      "step": 2342
    },
    {
      "completion_length": 230.8437614440918,
      "epoch": 0.3929753971247747,
      "grad_norm": 0.20767365117804806,
      "kl": 0.031772613525390625,
      "learning_rate": 2.893827160493827e-07,
      "loss": 0.0,
      "reward": 1.6892857998609543,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.6892857402563095,
      "rewards/format_reward_func": 1.0,
      "step": 2344
    },
    {
      "completion_length": 229.22768783569336,
      "epoch": 0.3933107003646423,
      "grad_norm": 0.18029574615767352,
      "kl": 0.055999755859375,
      "learning_rate": 2.896296296296296e-07,
      "loss": 0.0001,
      "reward": 1.7357143759727478,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.735714316368103,
      "rewards/format_reward_func": 1.0,
      "step": 2346
    },
    {
      "completion_length": 231.33483123779297,
      "epoch": 0.39364600360450985,
      "grad_norm": 0.174887596588021,
      "kl": 0.020839691162109375,
      "learning_rate": 2.8987654320987656e-07,
      "loss": 0.0,
      "reward": 1.7142857983708382,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7142857443541288,
      "rewards/format_reward_func": 1.0,
      "step": 2348
    },
    {
      "completion_length": 239.0803680419922,
      "epoch": 0.3939813068443774,
      "grad_norm": 0.24591155442162888,
      "kl": 0.028659820556640625,
      "learning_rate": 2.9012345679012345e-07,
      "loss": 0.0,
      "reward": 1.7821429073810577,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 2350
    },
    {
      "completion_length": 232.0580472946167,
      "epoch": 0.39431661008424496,
      "grad_norm": 0.16364288828454512,
      "kl": 0.031280517578125,
      "learning_rate": 2.903703703703704e-07,
      "loss": 0.0,
      "reward": 1.7428572177886963,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571823984385,
      "rewards/format_reward_func": 1.0,
      "step": 2352
    },
    {
      "completion_length": 238.96429634094238,
      "epoch": 0.3946519133241125,
      "grad_norm": 0.25228292667784574,
      "kl": 0.03216552734375,
      "learning_rate": 2.906172839506173e-07,
      "loss": 0.0,
      "reward": 1.7625000700354576,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7669643089175224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2354
    },
    {
      "completion_length": 237.01786994934082,
      "epoch": 0.39498721656398006,
      "grad_norm": 0.2681810554388981,
      "kl": 0.0478515625,
      "learning_rate": 2.9086419753086416e-07,
      "loss": 0.0,
      "reward": 1.7214286476373672,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.721428606659174,
      "rewards/format_reward_func": 1.0,
      "step": 2356
    },
    {
      "completion_length": 231.8660831451416,
      "epoch": 0.3953225198038476,
      "grad_norm": 0.2985047069341516,
      "kl": 0.048858642578125,
      "learning_rate": 2.911111111111111e-07,
      "loss": 0.0,
      "reward": 1.7732143327593803,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7776785958558321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2358
    },
    {
      "completion_length": 240.35268878936768,
      "epoch": 0.39565782304371516,
      "grad_norm": 0.2855045435472761,
      "kl": 0.023204803466796875,
      "learning_rate": 2.91358024691358e-07,
      "loss": 0.0,
      "reward": 1.7428571954369545,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7517857439815998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2360
    },
    {
      "completion_length": 237.57143878936768,
      "epoch": 0.39599312628358274,
      "grad_norm": 0.290504962458472,
      "kl": 0.030864715576171875,
      "learning_rate": 2.9160493827160494e-07,
      "loss": 0.0,
      "reward": 1.7607143744826317,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 2362
    },
    {
      "completion_length": 231.0669755935669,
      "epoch": 0.39632842952345027,
      "grad_norm": 0.3066050981425015,
      "kl": 0.025148391723632812,
      "learning_rate": 2.918518518518518e-07,
      "loss": 0.0,
      "reward": 1.791071504354477,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.795535746961832,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2364
    },
    {
      "completion_length": 238.04465579986572,
      "epoch": 0.39666373276331784,
      "grad_norm": 0.1959750175843398,
      "kl": 0.034114837646484375,
      "learning_rate": 2.9209876543209877e-07,
      "loss": 0.0,
      "reward": 1.6928572431206703,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.692857164889574,
      "rewards/format_reward_func": 1.0,
      "step": 2366
    },
    {
      "completion_length": 224.20090293884277,
      "epoch": 0.39699903600318537,
      "grad_norm": 0.18280818205958912,
      "kl": 0.0224151611328125,
      "learning_rate": 2.923456790123457e-07,
      "loss": 0.0,
      "reward": 1.753571517765522,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 2368
    },
    {
      "completion_length": 228.18750953674316,
      "epoch": 0.39733433924305295,
      "grad_norm": 0.40446503207075296,
      "kl": 0.059787750244140625,
      "learning_rate": 2.9259259259259254e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 2370
    },
    {
      "completion_length": 238.78572750091553,
      "epoch": 0.39766964248292047,
      "grad_norm": 0.3511749525167252,
      "kl": 0.02414703369140625,
      "learning_rate": 2.928395061728395e-07,
      "loss": 0.0,
      "reward": 1.7464286386966705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 2372
    },
    {
      "completion_length": 250.41518783569336,
      "epoch": 0.39800494572278805,
      "grad_norm": 0.4279324335140687,
      "kl": 0.04547119140625,
      "learning_rate": 2.930864197530864e-07,
      "loss": 0.0,
      "reward": 1.712500087916851,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7169643044471741,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2374
    },
    {
      "completion_length": 233.25893878936768,
      "epoch": 0.3983402489626556,
      "grad_norm": 0.3870702027447503,
      "kl": 0.03582000732421875,
      "learning_rate": 2.933333333333333e-07,
      "loss": 0.0,
      "reward": 1.6857143640518188,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.6857143249362707,
      "rewards/format_reward_func": 1.0,
      "step": 2376
    },
    {
      "completion_length": 243.75000953674316,
      "epoch": 0.39867555220252315,
      "grad_norm": 0.2756062262293934,
      "kl": 0.03237152099609375,
      "learning_rate": 2.9358024691358025e-07,
      "loss": 0.0,
      "reward": 1.7607143744826317,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143074274063,
      "rewards/format_reward_func": 1.0,
      "step": 2378
    },
    {
      "completion_length": 232.4821538925171,
      "epoch": 0.39901085544239073,
      "grad_norm": 0.21033752055447466,
      "kl": 0.022663116455078125,
      "learning_rate": 2.9382716049382714e-07,
      "loss": 0.0,
      "reward": 1.7017858177423477,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7062500212341547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2380
    },
    {
      "completion_length": 233.0535831451416,
      "epoch": 0.39934615868225826,
      "grad_norm": 0.30371212478863435,
      "kl": 0.02435302734375,
      "learning_rate": 2.940740740740741e-07,
      "loss": 0.0,
      "reward": 1.7821429446339607,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.782142873853445,
      "rewards/format_reward_func": 1.0,
      "step": 2382
    },
    {
      "completion_length": 233.5491180419922,
      "epoch": 0.39968146192212584,
      "grad_norm": 0.29400370741808635,
      "kl": 0.0242462158203125,
      "learning_rate": 2.9432098765432097e-07,
      "loss": 0.0,
      "reward": 1.7357143387198448,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7357143182307482,
      "rewards/format_reward_func": 1.0,
      "step": 2384
    },
    {
      "completion_length": 235.47768878936768,
      "epoch": 0.40001676516199336,
      "grad_norm": 0.24495863289215739,
      "kl": 0.0254669189453125,
      "learning_rate": 2.9456790123456786e-07,
      "loss": 0.0,
      "reward": 1.7607143372297287,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7696428783237934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2386
    },
    {
      "completion_length": 234.45090293884277,
      "epoch": 0.40035206840186094,
      "grad_norm": 0.18587902604819076,
      "kl": 0.0247650146484375,
      "learning_rate": 2.948148148148148e-07,
      "loss": 0.0,
      "reward": 1.807142898440361,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.807142898440361,
      "rewards/format_reward_func": 1.0,
      "step": 2388
    },
    {
      "completion_length": 233.30358409881592,
      "epoch": 0.40068737164172846,
      "grad_norm": 0.41121728830485704,
      "kl": 0.036991119384765625,
      "learning_rate": 2.950617283950617e-07,
      "loss": 0.0,
      "reward": 1.7928572073578835,
      "reward_std": 0.07071067485958338,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 2390
    },
    {
      "completion_length": 234.9732255935669,
      "epoch": 0.40102267488159604,
      "grad_norm": 0.2577178462756106,
      "kl": 0.027858734130859375,
      "learning_rate": 2.9530864197530863e-07,
      "loss": 0.0,
      "reward": 1.839285746216774,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.839285746216774,
      "rewards/format_reward_func": 1.0,
      "step": 2392
    },
    {
      "completion_length": 233.92411994934082,
      "epoch": 0.4013579781214636,
      "grad_norm": 0.08764588825744794,
      "kl": 0.02735137939453125,
      "learning_rate": 2.9555555555555557e-07,
      "loss": 0.0,
      "reward": 1.8035714700818062,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 2394
    },
    {
      "completion_length": 238.86161708831787,
      "epoch": 0.40169328136133114,
      "grad_norm": 0.1686954172881697,
      "kl": 0.026401519775390625,
      "learning_rate": 2.9580246913580246e-07,
      "loss": 0.0,
      "reward": 1.7428571954369545,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7517857570201159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2396
    },
    {
      "completion_length": 240.46875953674316,
      "epoch": 0.4020285846011987,
      "grad_norm": 0.28063510826087984,
      "kl": 0.034069061279296875,
      "learning_rate": 2.9604938271604935e-07,
      "loss": 0.0,
      "reward": 1.782142922282219,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 2398
    },
    {
      "completion_length": 243.3928680419922,
      "epoch": 0.40236388784106625,
      "grad_norm": 0.4889857363092545,
      "kl": 0.03224945068359375,
      "learning_rate": 2.962962962962963e-07,
      "loss": 0.0,
      "reward": 1.7535714954137802,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.762500025331974,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2400
    },
    {
      "completion_length": 236.2500123977661,
      "epoch": 0.4026991910809338,
      "grad_norm": 0.10869538161812134,
      "kl": 0.03144073486328125,
      "learning_rate": 2.965432098765432e-07,
      "loss": 0.0,
      "reward": 1.7178572118282318,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7178571708500385,
      "rewards/format_reward_func": 1.0,
      "step": 2402
    },
    {
      "completion_length": 244.38840579986572,
      "epoch": 0.40303449432080135,
      "grad_norm": 0.40208088618904964,
      "kl": 0.033901214599609375,
      "learning_rate": 2.967901234567901e-07,
      "loss": 0.0,
      "reward": 1.7714286372065544,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7803571596741676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2404
    },
    {
      "completion_length": 232.1964406967163,
      "epoch": 0.40336979756066893,
      "grad_norm": 0.290996152868013,
      "kl": 0.026607513427734375,
      "learning_rate": 2.97037037037037e-07,
      "loss": 0.0,
      "reward": 1.7875000461935997,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7919643074274063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2406
    },
    {
      "completion_length": 243.1785831451416,
      "epoch": 0.4037051008005365,
      "grad_norm": 0.17181050036006004,
      "kl": 0.06661605834960938,
      "learning_rate": 2.9728395061728395e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 2408
    },
    {
      "completion_length": 239.33483219146729,
      "epoch": 0.40404040404040403,
      "grad_norm": 0.20456214359825808,
      "kl": 0.06856155395507812,
      "learning_rate": 2.975308641975309e-07,
      "loss": 0.0001,
      "reward": 1.7625000402331352,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643051922321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2410
    },
    {
      "completion_length": 236.35268878936768,
      "epoch": 0.4043757072802716,
      "grad_norm": 0.10846825466310321,
      "kl": 0.02588653564453125,
      "learning_rate": 2.9777777777777773e-07,
      "loss": 0.0,
      "reward": 1.8053571805357933,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.8098214603960514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2412
    },
    {
      "completion_length": 227.97768783569336,
      "epoch": 0.40471101052013914,
      "grad_norm": 0.3076350136465618,
      "kl": 0.043216705322265625,
      "learning_rate": 2.9802469135802467e-07,
      "loss": 0.0,
      "reward": 1.7839286103844643,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2414
    },
    {
      "completion_length": 241.9509048461914,
      "epoch": 0.4050463137600067,
      "grad_norm": 0.22497192914816633,
      "kl": 0.0537261962890625,
      "learning_rate": 2.9827160493827156e-07,
      "loss": 0.0001,
      "reward": 1.6750000715255737,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.6839286033064127,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2416
    },
    {
      "completion_length": 237.3259048461914,
      "epoch": 0.40538161699987424,
      "grad_norm": 0.2892909531482391,
      "kl": 0.0372314453125,
      "learning_rate": 2.985185185185185e-07,
      "loss": 0.0,
      "reward": 1.7535714954137802,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 2418
    },
    {
      "completion_length": 248.46429920196533,
      "epoch": 0.4057169202397418,
      "grad_norm": 0.20048739142980532,
      "kl": 0.025386810302734375,
      "learning_rate": 2.9876543209876544e-07,
      "loss": 0.0,
      "reward": 1.717857226729393,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7178571783006191,
      "rewards/format_reward_func": 1.0,
      "step": 2420
    },
    {
      "completion_length": 238.56250953674316,
      "epoch": 0.4060522234796094,
      "grad_norm": 0.3425412260910249,
      "kl": 0.06465911865234375,
      "learning_rate": 2.9901234567901233e-07,
      "loss": 0.0001,
      "reward": 1.783928632736206,
      "reward_std": 0.09343910776078701,
      "rewards/equation_reward_func": 0.7883928958326578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2422
    },
    {
      "completion_length": 243.9151906967163,
      "epoch": 0.4063875267194769,
      "grad_norm": 0.2759994660120673,
      "kl": 0.034027099609375,
      "learning_rate": 2.9925925925925927e-07,
      "loss": 0.0,
      "reward": 1.703571505844593,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7035714723169804,
      "rewards/format_reward_func": 1.0,
      "step": 2424
    },
    {
      "completion_length": 241.92858123779297,
      "epoch": 0.4067228299593445,
      "grad_norm": 0.21101376319229112,
      "kl": 0.023448944091796875,
      "learning_rate": 2.9950617283950616e-07,
      "loss": 0.0,
      "reward": 1.70000009983778,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7000000402331352,
      "rewards/format_reward_func": 1.0,
      "step": 2426
    },
    {
      "completion_length": 239.5357265472412,
      "epoch": 0.407058133199212,
      "grad_norm": 0.1853824828635671,
      "kl": 0.022472381591796875,
      "learning_rate": 2.9975308641975305e-07,
      "loss": 0.0,
      "reward": 1.719642922282219,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7241071835160255,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2428
    },
    {
      "completion_length": 230.90625953674316,
      "epoch": 0.4073934364390796,
      "grad_norm": 0.10183301453604178,
      "kl": 0.033657073974609375,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 1.7553572058677673,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7598214540630579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2430
    },
    {
      "completion_length": 242.81251049041748,
      "epoch": 0.4077287396789471,
      "grad_norm": 0.046270323750636944,
      "kl": 0.024127960205078125,
      "learning_rate": 3.002469135802469e-07,
      "loss": 0.0,
      "reward": 1.7696429193019867,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7741071917116642,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2432
    },
    {
      "completion_length": 240.46875953674316,
      "epoch": 0.4080640429188147,
      "grad_norm": 0.3617812467572274,
      "kl": 0.0596771240234375,
      "learning_rate": 3.004938271604938e-07,
      "loss": 0.0001,
      "reward": 1.7482143640518188,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.752678606659174,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2434
    },
    {
      "completion_length": 238.13840293884277,
      "epoch": 0.4083993461586823,
      "grad_norm": 0.474064872648244,
      "kl": 0.042621612548828125,
      "learning_rate": 3.0074074074074076e-07,
      "loss": 0.0,
      "reward": 1.717857226729393,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7267857417464256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2436
    },
    {
      "completion_length": 229.0267972946167,
      "epoch": 0.4087346493985498,
      "grad_norm": 0.25945331903624114,
      "kl": 0.061847686767578125,
      "learning_rate": 3.0098765432098765e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 2438
    },
    {
      "completion_length": 246.94197845458984,
      "epoch": 0.4090699526384174,
      "grad_norm": 0.29150718449013036,
      "kl": 0.19774627685546875,
      "learning_rate": 3.012345679012346e-07,
      "loss": 0.0002,
      "reward": 1.7267857939004898,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.7401785925030708,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2440
    },
    {
      "completion_length": 243.5937623977661,
      "epoch": 0.4094052558782849,
      "grad_norm": 0.47884281229914716,
      "kl": 0.08807373046875,
      "learning_rate": 3.014814814814814e-07,
      "loss": 0.0001,
      "reward": 1.7017857804894447,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7151786088943481,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2442
    },
    {
      "completion_length": 233.33036994934082,
      "epoch": 0.4097405591181525,
      "grad_norm": 0.39819654959775574,
      "kl": 0.07984161376953125,
      "learning_rate": 3.0172839506172836e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 2444
    },
    {
      "completion_length": 247.2232265472412,
      "epoch": 0.41007586235802,
      "grad_norm": 0.2930877870059224,
      "kl": 0.07854843139648438,
      "learning_rate": 3.019753086419753e-07,
      "loss": 0.0001,
      "reward": 1.6589286550879478,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.6633928846567869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2446
    },
    {
      "completion_length": 242.20983219146729,
      "epoch": 0.4104111655978876,
      "grad_norm": 0.327769941363608,
      "kl": 0.04544830322265625,
      "learning_rate": 3.022222222222222e-07,
      "loss": 0.0,
      "reward": 1.6892857924103737,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.6982143186032772,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2448
    },
    {
      "completion_length": 234.95090293884277,
      "epoch": 0.4107464688377551,
      "grad_norm": 0.2819124888649559,
      "kl": 0.103240966796875,
      "learning_rate": 3.0246913580246913e-07,
      "loss": 0.0001,
      "reward": 1.710714340209961,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7196428794413805,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2450
    },
    {
      "completion_length": 238.6250114440918,
      "epoch": 0.4110817720776227,
      "grad_norm": 0.13389653861381598,
      "kl": 0.024539947509765625,
      "learning_rate": 3.027160493827161e-07,
      "loss": 0.0,
      "reward": 1.7714286297559738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285925030708,
      "rewards/format_reward_func": 1.0,
      "step": 2452
    },
    {
      "completion_length": 249.633939743042,
      "epoch": 0.4114170753174903,
      "grad_norm": 0.2228403364562828,
      "kl": 0.033233642578125,
      "learning_rate": 3.0296296296296296e-07,
      "loss": 0.0,
      "reward": 1.7946429029107094,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7991071715950966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2454
    },
    {
      "completion_length": 239.16519260406494,
      "epoch": 0.4117523785573578,
      "grad_norm": 0.29195230568320907,
      "kl": 0.02500152587890625,
      "learning_rate": 3.0320987654320985e-07,
      "loss": 0.0,
      "reward": 1.8071429133415222,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428909897804,
      "rewards/format_reward_func": 1.0,
      "step": 2456
    },
    {
      "completion_length": 243.6875114440918,
      "epoch": 0.4120876817972254,
      "grad_norm": 0.17428188867849037,
      "kl": 0.048297882080078125,
      "learning_rate": 3.0345679012345674e-07,
      "loss": 0.0,
      "reward": 1.7571429535746574,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7660714574158192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2458
    },
    {
      "completion_length": 236.196439743042,
      "epoch": 0.4124229850370929,
      "grad_norm": 0.36680589539865105,
      "kl": 0.061859130859375,
      "learning_rate": 3.037037037037037e-07,
      "loss": 0.0001,
      "reward": 1.785714365541935,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7946428768336773,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2460
    },
    {
      "completion_length": 241.49554824829102,
      "epoch": 0.4127582882769605,
      "grad_norm": 0.482576488789002,
      "kl": 0.05983734130859375,
      "learning_rate": 3.039506172839506e-07,
      "loss": 0.0001,
      "reward": 1.762500062584877,
      "reward_std": 0.10354063659906387,
      "rewards/equation_reward_func": 0.7758928947150707,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2462
    },
    {
      "completion_length": 251.9509048461914,
      "epoch": 0.413093591516828,
      "grad_norm": 0.5102705709365054,
      "kl": 0.23926925659179688,
      "learning_rate": 3.041975308641975e-07,
      "loss": 0.0002,
      "reward": 1.7821429148316383,
      "reward_std": 0.11616754066199064,
      "rewards/equation_reward_func": 0.8000000156462193,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2464
    },
    {
      "completion_length": 243.87054538726807,
      "epoch": 0.4134288947566956,
      "grad_norm": 0.6208793238312089,
      "kl": 0.6168899536132812,
      "learning_rate": 3.0444444444444445e-07,
      "loss": 0.0006,
      "reward": 1.78035718947649,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7848214507102966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2466
    },
    {
      "completion_length": 230.75447463989258,
      "epoch": 0.41376419799656317,
      "grad_norm": 0.20628736823470975,
      "kl": 0.09053802490234375,
      "learning_rate": 3.046913580246914e-07,
      "loss": 0.0001,
      "reward": 1.8357143327593803,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8357142955064774,
      "rewards/format_reward_func": 1.0,
      "step": 2468
    },
    {
      "completion_length": 247.08036708831787,
      "epoch": 0.4140995012364307,
      "grad_norm": 0.34160066484399093,
      "kl": 0.14976119995117188,
      "learning_rate": 3.0493827160493823e-07,
      "loss": 0.0001,
      "reward": 1.726785771548748,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.731250025331974,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2470
    },
    {
      "completion_length": 240.43304824829102,
      "epoch": 0.41443480447629827,
      "grad_norm": 0.22342851163494104,
      "kl": 0.24057769775390625,
      "learning_rate": 3.0518518518518517e-07,
      "loss": 0.0002,
      "reward": 1.7160714864730835,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7205357383936644,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2472
    },
    {
      "completion_length": 241.9062623977661,
      "epoch": 0.4147701077161658,
      "grad_norm": 0.1707649901922291,
      "kl": 0.0610198974609375,
      "learning_rate": 3.0543209876543206e-07,
      "loss": 0.0001,
      "reward": 1.762500062584877,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643219560385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2474
    },
    {
      "completion_length": 244.6607255935669,
      "epoch": 0.41510541095603337,
      "grad_norm": 0.28532945804476273,
      "kl": 0.042633056640625,
      "learning_rate": 3.05679012345679e-07,
      "loss": 0.0,
      "reward": 1.8250000402331352,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.825000025331974,
      "rewards/format_reward_func": 1.0,
      "step": 2476
    },
    {
      "completion_length": 246.70983409881592,
      "epoch": 0.4154407141959009,
      "grad_norm": 0.20875711813517592,
      "kl": 0.06849288940429688,
      "learning_rate": 3.0592592592592594e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286033064127,
      "rewards/format_reward_func": 1.0,
      "step": 2478
    },
    {
      "completion_length": 242.1428680419922,
      "epoch": 0.4157760174357685,
      "grad_norm": 0.16125349654244475,
      "kl": 0.026233673095703125,
      "learning_rate": 3.0617283950617283e-07,
      "loss": 0.0,
      "reward": 1.7321429252624512,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7321428768336773,
      "rewards/format_reward_func": 1.0,
      "step": 2480
    },
    {
      "completion_length": 242.0625123977661,
      "epoch": 0.41611132067563605,
      "grad_norm": 0.24992527870915915,
      "kl": 0.032123565673828125,
      "learning_rate": 3.0641975308641977e-07,
      "loss": 0.0,
      "reward": 1.7714286297559738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 2482
    },
    {
      "completion_length": 251.56251335144043,
      "epoch": 0.4164466239155036,
      "grad_norm": 0.277668143360467,
      "kl": 0.05127716064453125,
      "learning_rate": 3.066666666666666e-07,
      "loss": 0.0001,
      "reward": 1.7000001072883606,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7000000290572643,
      "rewards/format_reward_func": 1.0,
      "step": 2484
    },
    {
      "completion_length": 244.11608123779297,
      "epoch": 0.41678192715537116,
      "grad_norm": 0.1966415408927424,
      "kl": 0.033660888671875,
      "learning_rate": 3.0691358024691355e-07,
      "loss": 0.0,
      "reward": 1.678571529686451,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.6785714775323868,
      "rewards/format_reward_func": 1.0,
      "step": 2486
    },
    {
      "completion_length": 243.00000858306885,
      "epoch": 0.4171172303952387,
      "grad_norm": 0.4143451684580404,
      "kl": 0.0799713134765625,
      "learning_rate": 3.071604938271605e-07,
      "loss": 0.0001,
      "reward": 1.7665179148316383,
      "reward_std": 0.06755394861102104,
      "rewards/equation_reward_func": 0.7678571790456772,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 2488
    },
    {
      "completion_length": 237.1473331451416,
      "epoch": 0.41745253363510626,
      "grad_norm": 0.26423638528297066,
      "kl": 0.03952789306640625,
      "learning_rate": 3.074074074074074e-07,
      "loss": 0.0,
      "reward": 1.814285784959793,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8142857402563095,
      "rewards/format_reward_func": 1.0,
      "step": 2490
    },
    {
      "completion_length": 244.64286708831787,
      "epoch": 0.4177878368749738,
      "grad_norm": 0.2312057942169059,
      "kl": 0.04471588134765625,
      "learning_rate": 3.076543209876543e-07,
      "loss": 0.0,
      "reward": 1.7593750730156898,
      "reward_std": 0.027147849323228,
      "rewards/equation_reward_func": 0.7607143074274063,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 2492
    },
    {
      "completion_length": 246.00447273254395,
      "epoch": 0.41812314011484136,
      "grad_norm": 0.4950536770067875,
      "kl": 0.04170989990234375,
      "learning_rate": 3.0790123456790126e-07,
      "loss": 0.0,
      "reward": 1.7196429297327995,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7241071779280901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2494
    },
    {
      "completion_length": 244.5491189956665,
      "epoch": 0.41845844335470894,
      "grad_norm": 0.32448324165354375,
      "kl": 0.02623748779296875,
      "learning_rate": 3.0814814814814815e-07,
      "loss": 0.0,
      "reward": 1.7821429446339607,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.7910714540630579,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2496
    },
    {
      "completion_length": 247.46430015563965,
      "epoch": 0.41879374659457647,
      "grad_norm": 0.23471168214360996,
      "kl": 0.0288848876953125,
      "learning_rate": 3.0839506172839504e-07,
      "loss": 0.0,
      "reward": 1.7267857864499092,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7401786036789417,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2498
    },
    {
      "completion_length": 244.1384038925171,
      "epoch": 0.41912904983444405,
      "grad_norm": 0.2761535326936902,
      "kl": 0.025539398193359375,
      "learning_rate": 3.086419753086419e-07,
      "loss": 0.0,
      "reward": 1.783928632736206,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7883928865194321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2500
    },
    {
      "completion_length": 248.1651906967163,
      "epoch": 0.41946435307431157,
      "grad_norm": 0.2309364519086521,
      "kl": 0.037647247314453125,
      "learning_rate": 3.0888888888888887e-07,
      "loss": 0.0,
      "reward": 1.7642858028411865,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 2502
    },
    {
      "completion_length": 250.14733219146729,
      "epoch": 0.41979965631417915,
      "grad_norm": 0.29956528286167283,
      "kl": 0.07854461669921875,
      "learning_rate": 3.091358024691358e-07,
      "loss": 0.0001,
      "reward": 1.7607143372297287,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.769642885774374,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2504
    },
    {
      "completion_length": 246.00893783569336,
      "epoch": 0.42013495955404667,
      "grad_norm": 0.3199713423763226,
      "kl": 0.08681488037109375,
      "learning_rate": 3.093827160493827e-07,
      "loss": 0.0001,
      "reward": 1.7428572177886963,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7517857439815998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2506
    },
    {
      "completion_length": 243.0134048461914,
      "epoch": 0.42047026279391425,
      "grad_norm": 0.11689773260946061,
      "kl": 0.06568527221679688,
      "learning_rate": 3.0962962962962964e-07,
      "loss": 0.0001,
      "reward": 1.7357143461704254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143126428127,
      "rewards/format_reward_func": 1.0,
      "step": 2508
    },
    {
      "completion_length": 240.65179920196533,
      "epoch": 0.4208055660337818,
      "grad_norm": 0.3331850418354822,
      "kl": 0.0395355224609375,
      "learning_rate": 3.098765432098765e-07,
      "loss": 0.0,
      "reward": 1.7553572058677673,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7598214522004128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2510
    },
    {
      "completion_length": 239.3125114440918,
      "epoch": 0.42114086927364935,
      "grad_norm": 0.24332880806945797,
      "kl": 0.026187896728515625,
      "learning_rate": 3.101234567901234e-07,
      "loss": 0.0,
      "reward": 1.771428644657135,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7803571745753288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2512
    },
    {
      "completion_length": 256.1294775009155,
      "epoch": 0.42147617251351693,
      "grad_norm": 0.2725284304146817,
      "kl": 0.04680633544921875,
      "learning_rate": 3.1037037037037036e-07,
      "loss": 0.0,
      "reward": 1.721428632736206,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7214285973459482,
      "rewards/format_reward_func": 1.0,
      "step": 2514
    },
    {
      "completion_length": 242.76340103149414,
      "epoch": 0.42181147575338446,
      "grad_norm": 0.4089842216356524,
      "kl": 0.0549163818359375,
      "learning_rate": 3.1061728395061724e-07,
      "loss": 0.0001,
      "reward": 1.700000062584877,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7000000327825546,
      "rewards/format_reward_func": 1.0,
      "step": 2516
    },
    {
      "completion_length": 239.29465103149414,
      "epoch": 0.42214677899325204,
      "grad_norm": 0.22043840909969697,
      "kl": 0.05002593994140625,
      "learning_rate": 3.108641975308642e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000149011612,
      "rewards/format_reward_func": 1.0,
      "step": 2518
    },
    {
      "completion_length": 243.62947368621826,
      "epoch": 0.42248208223311956,
      "grad_norm": 0.1826835616573898,
      "kl": 0.0255279541015625,
      "learning_rate": 3.111111111111111e-07,
      "loss": 0.0,
      "reward": 1.7714286670088768,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 2520
    },
    {
      "completion_length": 243.39287090301514,
      "epoch": 0.42281738547298714,
      "grad_norm": 0.216918530888218,
      "kl": 0.035877227783203125,
      "learning_rate": 3.11358024691358e-07,
      "loss": 0.0,
      "reward": 1.7750000730156898,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000394880772,
      "rewards/format_reward_func": 1.0,
      "step": 2522
    },
    {
      "completion_length": 242.32590579986572,
      "epoch": 0.42315268871285466,
      "grad_norm": 0.16724150439793084,
      "kl": 0.03157806396484375,
      "learning_rate": 3.1160493827160496e-07,
      "loss": 0.0,
      "reward": 1.7928571999073029,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 2524
    },
    {
      "completion_length": 244.15179634094238,
      "epoch": 0.42348799195272224,
      "grad_norm": 0.2824979326339329,
      "kl": 0.04889678955078125,
      "learning_rate": 3.118518518518518e-07,
      "loss": 0.0,
      "reward": 1.7571429461240768,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7660714536905289,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2526
    },
    {
      "completion_length": 243.45536708831787,
      "epoch": 0.4238232951925898,
      "grad_norm": 0.37376178454239156,
      "kl": 0.03272247314453125,
      "learning_rate": 3.1209876543209873e-07,
      "loss": 0.0,
      "reward": 1.7660714983940125,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7705357521772385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2528
    },
    {
      "completion_length": 235.4509048461914,
      "epoch": 0.42415859843245735,
      "grad_norm": 0.40747437616592286,
      "kl": 0.04550933837890625,
      "learning_rate": 3.123456790123457e-07,
      "loss": 0.0,
      "reward": 1.757142923772335,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 2530
    },
    {
      "completion_length": 248.3750123977661,
      "epoch": 0.4244939016723249,
      "grad_norm": 0.2179129790820734,
      "kl": 0.059764862060546875,
      "learning_rate": 3.1259259259259256e-07,
      "loss": 0.0001,
      "reward": 1.7553571984171867,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7598214577883482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2532
    },
    {
      "completion_length": 255.93304538726807,
      "epoch": 0.42482920491219245,
      "grad_norm": 0.253351262982618,
      "kl": 0.0557861328125,
      "learning_rate": 3.128395061728395e-07,
      "loss": 0.0001,
      "reward": 1.7642857655882835,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.77321432903409,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2534
    },
    {
      "completion_length": 242.6294755935669,
      "epoch": 0.42516450815206003,
      "grad_norm": 0.21174747359151216,
      "kl": 0.0252227783203125,
      "learning_rate": 3.130864197530864e-07,
      "loss": 0.0,
      "reward": 1.8178572058677673,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.817857176065445,
      "rewards/format_reward_func": 1.0,
      "step": 2536
    },
    {
      "completion_length": 236.34376049041748,
      "epoch": 0.42549981139192755,
      "grad_norm": 0.07787793090487856,
      "kl": 0.049343109130859375,
      "learning_rate": 3.1333333333333333e-07,
      "loss": 0.0,
      "reward": 1.735714353621006,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7357143182307482,
      "rewards/format_reward_func": 1.0,
      "step": 2538
    },
    {
      "completion_length": 244.9241189956665,
      "epoch": 0.42583511463179513,
      "grad_norm": 0.2622517959405339,
      "kl": 0.04746246337890625,
      "learning_rate": 3.135802469135803e-07,
      "loss": 0.0,
      "reward": 1.7517858073115349,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7562500238418579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2540
    },
    {
      "completion_length": 239.76786613464355,
      "epoch": 0.4261704178716627,
      "grad_norm": 0.34941696728550165,
      "kl": 0.02500152587890625,
      "learning_rate": 3.138271604938271e-07,
      "loss": 0.0,
      "reward": 1.7303572073578835,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7348214611411095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2542
    },
    {
      "completion_length": 237.49554443359375,
      "epoch": 0.42650572111153023,
      "grad_norm": 0.20175321039359187,
      "kl": 0.022891998291015625,
      "learning_rate": 3.1407407407407405e-07,
      "loss": 0.0,
      "reward": 1.7357143461704254,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.735714303329587,
      "rewards/format_reward_func": 1.0,
      "step": 2544
    },
    {
      "completion_length": 251.40626335144043,
      "epoch": 0.4268410243513978,
      "grad_norm": 0.13603414263062696,
      "kl": 0.08626937866210938,
      "learning_rate": 3.14320987654321e-07,
      "loss": 0.0001,
      "reward": 1.7178572490811348,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7357143126428127,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2546
    },
    {
      "completion_length": 249.23215293884277,
      "epoch": 0.42717632759126534,
      "grad_norm": 0.25842443169289875,
      "kl": 0.039806365966796875,
      "learning_rate": 3.145679012345679e-07,
      "loss": 0.0,
      "reward": 1.723214365541935,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7276785802096128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2548
    },
    {
      "completion_length": 248.36608219146729,
      "epoch": 0.4275116308311329,
      "grad_norm": 0.2714677153001056,
      "kl": 0.12255859375,
      "learning_rate": 3.148148148148148e-07,
      "loss": 0.0001,
      "reward": 1.7178572192788124,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7178571745753288,
      "rewards/format_reward_func": 1.0,
      "step": 2550
    },
    {
      "completion_length": 236.23215293884277,
      "epoch": 0.42784693407100044,
      "grad_norm": 0.35098529470563555,
      "kl": 0.03569793701171875,
      "learning_rate": 3.150617283950617e-07,
      "loss": 0.0,
      "reward": 1.7535714879631996,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7535714507102966,
      "rewards/format_reward_func": 1.0,
      "step": 2552
    },
    {
      "completion_length": 251.02679443359375,
      "epoch": 0.428182237310868,
      "grad_norm": 0.12539207610264913,
      "kl": 0.08616256713867188,
      "learning_rate": 3.1530864197530865e-07,
      "loss": 0.0001,
      "reward": 1.7946429029107094,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.799107164144516,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2554
    },
    {
      "completion_length": 238.7901906967163,
      "epoch": 0.4285175405507356,
      "grad_norm": 0.3178237413778879,
      "kl": 0.023960113525390625,
      "learning_rate": 3.1555555555555554e-07,
      "loss": 0.0,
      "reward": 1.7446429207921028,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2556
    },
    {
      "completion_length": 245.50001049041748,
      "epoch": 0.4288528437906031,
      "grad_norm": 0.45006753486366646,
      "kl": 0.10959625244140625,
      "learning_rate": 3.1580246913580243e-07,
      "loss": 0.0001,
      "reward": 1.7267857939004898,
      "reward_std": 0.0732360566034913,
      "rewards/equation_reward_func": 0.7312500216066837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2558
    },
    {
      "completion_length": 247.4196548461914,
      "epoch": 0.4291881470304707,
      "grad_norm": 0.8192667075625969,
      "kl": 0.22243881225585938,
      "learning_rate": 3.1604938271604937e-07,
      "loss": 0.0002,
      "reward": 1.7750000581145287,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7839286047965288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2560
    },
    {
      "completion_length": 251.14286708831787,
      "epoch": 0.4295234502703382,
      "grad_norm": 0.22454144415636515,
      "kl": 0.024169921875,
      "learning_rate": 3.1629629629629626e-07,
      "loss": 0.0,
      "reward": 1.726785808801651,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7312500290572643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2562
    },
    {
      "completion_length": 246.77679634094238,
      "epoch": 0.4298587535102058,
      "grad_norm": 0.3261496817001146,
      "kl": 0.19002532958984375,
      "learning_rate": 3.165432098765432e-07,
      "loss": 0.0002,
      "reward": 1.7892857939004898,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 2564
    },
    {
      "completion_length": 243.477689743042,
      "epoch": 0.43019405675007333,
      "grad_norm": 0.33328892831898055,
      "kl": 0.037197113037109375,
      "learning_rate": 3.1679012345679014e-07,
      "loss": 0.0,
      "reward": 1.737500049173832,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7508929036557674,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2566
    },
    {
      "completion_length": 236.32590198516846,
      "epoch": 0.4305293599899409,
      "grad_norm": 0.145820177741581,
      "kl": 0.02503204345703125,
      "learning_rate": 3.1703703703703703e-07,
      "loss": 0.0,
      "reward": 1.7500000521540642,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7589285969734192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2568
    },
    {
      "completion_length": 246.85715293884277,
      "epoch": 0.43086466322980843,
      "grad_norm": 0.29553026796035753,
      "kl": 0.06223297119140625,
      "learning_rate": 3.172839506172839e-07,
      "loss": 0.0001,
      "reward": 1.7625000774860382,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.7758928798139095,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2570
    },
    {
      "completion_length": 247.07590198516846,
      "epoch": 0.431199966469676,
      "grad_norm": 0.15917956082740858,
      "kl": 0.30030059814453125,
      "learning_rate": 3.1753086419753086e-07,
      "loss": 0.0003,
      "reward": 1.7714286372065544,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 2572
    },
    {
      "completion_length": 259.2946538925171,
      "epoch": 0.4315352697095436,
      "grad_norm": 0.45168312035874253,
      "kl": 0.3671760559082031,
      "learning_rate": 3.1777777777777775e-07,
      "loss": 0.0004,
      "reward": 1.7446429282426834,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7491071783006191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2574
    },
    {
      "completion_length": 240.11608409881592,
      "epoch": 0.4318705729494111,
      "grad_norm": 0.3496117852276753,
      "kl": 0.02508544921875,
      "learning_rate": 3.180246913580247e-07,
      "loss": 0.0,
      "reward": 1.7375000715255737,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7419643253087997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2576
    },
    {
      "completion_length": 243.9821538925171,
      "epoch": 0.4322058761892787,
      "grad_norm": 0.3225535714095935,
      "kl": 0.02301025390625,
      "learning_rate": 3.182716049382716e-07,
      "loss": 0.0,
      "reward": 1.7250001057982445,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7428571842610836,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2578
    },
    {
      "completion_length": 250.12054634094238,
      "epoch": 0.4325411794291462,
      "grad_norm": 0.26559242472932765,
      "kl": 0.034000396728515625,
      "learning_rate": 3.185185185185185e-07,
      "loss": 0.0,
      "reward": 1.7857143506407738,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7857143320143223,
      "rewards/format_reward_func": 1.0,
      "step": 2580
    },
    {
      "completion_length": 250.25447463989258,
      "epoch": 0.4328764826690138,
      "grad_norm": 0.2759127477860899,
      "kl": 0.06327056884765625,
      "learning_rate": 3.1876543209876546e-07,
      "loss": 0.0001,
      "reward": 1.7017857730388641,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7151785958558321,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2582
    },
    {
      "completion_length": 244.22769165039062,
      "epoch": 0.4332117859088813,
      "grad_norm": 0.207482672686499,
      "kl": 0.1029205322265625,
      "learning_rate": 3.190123456790123e-07,
      "loss": 0.0001,
      "reward": 1.7089286521077156,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7133928947150707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2584
    },
    {
      "completion_length": 245.727689743042,
      "epoch": 0.4335470891487489,
      "grad_norm": 0.5793295166969608,
      "kl": 0.1691436767578125,
      "learning_rate": 3.1925925925925924e-07,
      "loss": 0.0002,
      "reward": 1.7196429371833801,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7241071723401546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2586
    },
    {
      "completion_length": 251.60715293884277,
      "epoch": 0.4338823923886165,
      "grad_norm": 0.5828281237924078,
      "kl": 0.9253044128417969,
      "learning_rate": 3.195061728395061e-07,
      "loss": 0.0009,
      "reward": 1.7607143223285675,
      "reward_std": 0.07576144114136696,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2588
    },
    {
      "completion_length": 255.88394260406494,
      "epoch": 0.434217695628484,
      "grad_norm": 0.25950153452146346,
      "kl": 0.05170440673828125,
      "learning_rate": 3.1975308641975307e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 2590
    },
    {
      "completion_length": 251.83929443359375,
      "epoch": 0.4345529988683516,
      "grad_norm": 0.31718516852729894,
      "kl": 0.11458206176757812,
      "learning_rate": 3.2e-07,
      "loss": 0.0001,
      "reward": 1.7803571820259094,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7848214656114578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2592
    },
    {
      "completion_length": 251.4419765472412,
      "epoch": 0.4348883021082191,
      "grad_norm": 0.1884583384444149,
      "kl": 0.12668609619140625,
      "learning_rate": 3.202469135802469e-07,
      "loss": 0.0001,
      "reward": 1.7339286357164383,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.73839289881289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2594
    },
    {
      "completion_length": 252.9375123977661,
      "epoch": 0.4352236053480867,
      "grad_norm": 0.22589381749613982,
      "kl": 0.035221099853515625,
      "learning_rate": 3.2049382716049384e-07,
      "loss": 0.0,
      "reward": 1.8482143506407738,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.852678582072258,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2596
    },
    {
      "completion_length": 246.90179538726807,
      "epoch": 0.4355589085879542,
      "grad_norm": 0.15749758438094288,
      "kl": 0.04557037353515625,
      "learning_rate": 3.207407407407407e-07,
      "loss": 0.0,
      "reward": 1.7821429073810577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 2598
    },
    {
      "completion_length": 265.1696529388428,
      "epoch": 0.4358942118278218,
      "grad_norm": 0.22695902010162736,
      "kl": 0.19362640380859375,
      "learning_rate": 3.209876543209876e-07,
      "loss": 0.0002,
      "reward": 1.7160714864730835,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7294643111526966,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2600
    },
    {
      "completion_length": 246.68751049041748,
      "epoch": 0.43622951506768937,
      "grad_norm": 0.3494971527672536,
      "kl": 0.2670745849609375,
      "learning_rate": 3.2123456790123455e-07,
      "loss": 0.0003,
      "reward": 1.7857143729925156,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7857142984867096,
      "rewards/format_reward_func": 1.0,
      "step": 2602
    },
    {
      "completion_length": 252.92858505249023,
      "epoch": 0.4365648183075569,
      "grad_norm": 0.2570008866666796,
      "kl": 0.08303451538085938,
      "learning_rate": 3.2148148148148144e-07,
      "loss": 0.0001,
      "reward": 1.7446429207921028,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2604
    },
    {
      "completion_length": 254.10269165039062,
      "epoch": 0.43690012154742447,
      "grad_norm": 0.3532176636632374,
      "kl": 1.0090484619140625,
      "learning_rate": 3.217283950617284e-07,
      "loss": 0.001,
      "reward": 1.7375000640749931,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643215835094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2606
    },
    {
      "completion_length": 252.3526906967163,
      "epoch": 0.437235424787292,
      "grad_norm": 0.1619935140323748,
      "kl": 0.17085647583007812,
      "learning_rate": 3.219753086419753e-07,
      "loss": 0.0002,
      "reward": 1.7375000789761543,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.750892873853445,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2608
    },
    {
      "completion_length": 253.14286708831787,
      "epoch": 0.4375707280271596,
      "grad_norm": 0.19724744628716312,
      "kl": 0.15912628173828125,
      "learning_rate": 3.222222222222222e-07,
      "loss": 0.0002,
      "reward": 1.7232143878936768,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7276786081492901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2610
    },
    {
      "completion_length": 257.7009057998657,
      "epoch": 0.4379060312670271,
      "grad_norm": 0.500823014531597,
      "kl": 0.095916748046875,
      "learning_rate": 3.224691358024691e-07,
      "loss": 0.0001,
      "reward": 1.741071492433548,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7455357424914837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2612
    },
    {
      "completion_length": 242.83929824829102,
      "epoch": 0.4382413345068947,
      "grad_norm": 0.30820036330593226,
      "kl": 0.0233306884765625,
      "learning_rate": 3.22716049382716e-07,
      "loss": 0.0,
      "reward": 1.76071435213089,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143055647612,
      "rewards/format_reward_func": 1.0,
      "step": 2614
    },
    {
      "completion_length": 253.11608219146729,
      "epoch": 0.43857663774676225,
      "grad_norm": 0.2536925262111839,
      "kl": 0.04438018798828125,
      "learning_rate": 3.2296296296296293e-07,
      "loss": 0.0,
      "reward": 1.708928644657135,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7133928816765547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2616
    },
    {
      "completion_length": 258.24554920196533,
      "epoch": 0.4389119409866298,
      "grad_norm": 0.28150699335408913,
      "kl": 0.10482025146484375,
      "learning_rate": 3.2320987654320987e-07,
      "loss": 0.0001,
      "reward": 1.7339286357164383,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.7473214454948902,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2618
    },
    {
      "completion_length": 254.0580472946167,
      "epoch": 0.43924724422649736,
      "grad_norm": 0.12402202969919433,
      "kl": 0.6347122192382812,
      "learning_rate": 3.2345679012345676e-07,
      "loss": 0.0006,
      "reward": 1.7178572192788124,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7267857454717159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2620
    },
    {
      "completion_length": 257.6473321914673,
      "epoch": 0.4395825474663649,
      "grad_norm": 0.3870609356353709,
      "kl": 0.027385711669921875,
      "learning_rate": 3.237037037037037e-07,
      "loss": 0.0,
      "reward": 1.7107143551111221,
      "reward_std": 0.08586296439170837,
      "rewards/equation_reward_func": 0.7196428887546062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2622
    },
    {
      "completion_length": 250.04912090301514,
      "epoch": 0.43991785070623246,
      "grad_norm": 0.4512728132951566,
      "kl": 0.14263916015625,
      "learning_rate": 3.2395061728395064e-07,
      "loss": 0.0001,
      "reward": 1.7303572297096252,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7348214723169804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2624
    },
    {
      "completion_length": 266.42858695983887,
      "epoch": 0.4402531539461,
      "grad_norm": 0.2475325032392611,
      "kl": 0.025585174560546875,
      "learning_rate": 3.2419753086419753e-07,
      "loss": 0.0,
      "reward": 1.7392857670783997,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.748214315623045,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2626
    },
    {
      "completion_length": 240.79019260406494,
      "epoch": 0.44058845718596756,
      "grad_norm": 0.24164859505408506,
      "kl": 0.3480949401855469,
      "learning_rate": 3.244444444444444e-07,
      "loss": 0.0003,
      "reward": 1.7964286357164383,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8053571805357933,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2628
    },
    {
      "completion_length": 254.13393878936768,
      "epoch": 0.4409237604258351,
      "grad_norm": 0.31436550452414275,
      "kl": 0.12886428833007812,
      "learning_rate": 3.246913580246913e-07,
      "loss": 0.0001,
      "reward": 1.74642863124609,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.7553571872413158,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2630
    },
    {
      "completion_length": 251.0580472946167,
      "epoch": 0.44125906366570267,
      "grad_norm": 0.29791950286400054,
      "kl": 0.037143707275390625,
      "learning_rate": 3.2493827160493825e-07,
      "loss": 0.0,
      "reward": 1.783928632736206,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7883928865194321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2632
    },
    {
      "completion_length": 270.63840675354004,
      "epoch": 0.44159436690557025,
      "grad_norm": 0.24728104547594096,
      "kl": 0.06447219848632812,
      "learning_rate": 3.251851851851852e-07,
      "loss": 0.0001,
      "reward": 1.7410714998841286,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7544643133878708,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2634
    },
    {
      "completion_length": 248.92858409881592,
      "epoch": 0.44192967014543777,
      "grad_norm": 0.31478920467178306,
      "kl": 0.051105499267578125,
      "learning_rate": 3.254320987654321e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7696428894996643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2636
    },
    {
      "completion_length": 249.5759038925171,
      "epoch": 0.44226497338530535,
      "grad_norm": 0.24993023096821998,
      "kl": 0.13302993774414062,
      "learning_rate": 3.25679012345679e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.045456865802407265,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2638
    },
    {
      "completion_length": 260.30804920196533,
      "epoch": 0.4426002766251729,
      "grad_norm": 1.3821262018529439,
      "kl": 0.258209228515625,
      "learning_rate": 3.2592592592592596e-07,
      "loss": 0.0003,
      "reward": 1.7089286297559738,
      "reward_std": 0.08838834799826145,
      "rewards/equation_reward_func": 0.7312500216066837,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 2640
    },
    {
      "completion_length": 262.3973340988159,
      "epoch": 0.44293557986504045,
      "grad_norm": 0.23587870226894336,
      "kl": 0.12923049926757812,
      "learning_rate": 3.261728395061728e-07,
      "loss": 0.0001,
      "reward": 1.7589286044239998,
      "reward_std": 0.10859139915555716,
      "rewards/equation_reward_func": 0.7812500335276127,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 2642
    },
    {
      "completion_length": 265.0803699493408,
      "epoch": 0.443270883104908,
      "grad_norm": 0.260573659490083,
      "kl": 0.06291961669921875,
      "learning_rate": 3.2641975308641974e-07,
      "loss": 0.0001,
      "reward": 1.7142857983708382,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7232143115252256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2644
    },
    {
      "completion_length": 266.5848331451416,
      "epoch": 0.44360618634477555,
      "grad_norm": 0.29066419116889264,
      "kl": 0.4171791076660156,
      "learning_rate": 3.2666666666666663e-07,
      "loss": 0.0004,
      "reward": 1.7571428939700127,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2646
    },
    {
      "completion_length": 255.9464406967163,
      "epoch": 0.44394148958464313,
      "grad_norm": 0.2812603360108659,
      "kl": 0.05731964111328125,
      "learning_rate": 3.2691358024691357e-07,
      "loss": 0.0001,
      "reward": 1.7303571701049805,
      "reward_std": 0.1085914010182023,
      "rewards/equation_reward_func": 0.7616071719676256,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 2648
    },
    {
      "completion_length": 262.03572940826416,
      "epoch": 0.44427679282451066,
      "grad_norm": 0.30695546284,
      "kl": 0.057254791259765625,
      "learning_rate": 3.271604938271605e-07,
      "loss": 0.0001,
      "reward": 1.7053572461009026,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7187500298023224,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2650
    },
    {
      "completion_length": 264.9910831451416,
      "epoch": 0.44461209606437824,
      "grad_norm": 0.3474524782514886,
      "kl": 0.06933212280273438,
      "learning_rate": 3.274074074074074e-07,
      "loss": 0.0001,
      "reward": 1.7571429461240768,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7660714648663998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2652
    },
    {
      "completion_length": 249.60268878936768,
      "epoch": 0.44494739930424576,
      "grad_norm": 0.2084635078884685,
      "kl": 0.029857635498046875,
      "learning_rate": 3.2765432098765434e-07,
      "loss": 0.0,
      "reward": 1.7553572058677673,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214596509933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2654
    },
    {
      "completion_length": 247.1919755935669,
      "epoch": 0.44528270254411334,
      "grad_norm": 0.1889165270559453,
      "kl": 0.0886993408203125,
      "learning_rate": 3.279012345679012e-07,
      "loss": 0.0001,
      "reward": 1.78035718947649,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214618861675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2656
    },
    {
      "completion_length": 245.50894165039062,
      "epoch": 0.44561800578398086,
      "grad_norm": 0.1117686353305998,
      "kl": 0.03253936767578125,
      "learning_rate": 3.281481481481481e-07,
      "loss": 0.0,
      "reward": 1.7714286521077156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 2658
    },
    {
      "completion_length": 250.4375123977661,
      "epoch": 0.44595330902384844,
      "grad_norm": 0.2595808740548275,
      "kl": 0.1915740966796875,
      "learning_rate": 3.2839506172839506e-07,
      "loss": 0.0002,
      "reward": 1.750000074505806,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 2660
    },
    {
      "completion_length": 251.02233600616455,
      "epoch": 0.446288612263716,
      "grad_norm": 0.2233516502035073,
      "kl": 0.02516937255859375,
      "learning_rate": 3.2864197530864195e-07,
      "loss": 0.0,
      "reward": 1.7214286401867867,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7214285992085934,
      "rewards/format_reward_func": 1.0,
      "step": 2662
    },
    {
      "completion_length": 257.571439743042,
      "epoch": 0.44662391550358355,
      "grad_norm": 0.2758449396807126,
      "kl": 0.046142578125,
      "learning_rate": 3.288888888888889e-07,
      "loss": 0.0,
      "reward": 1.7553572207689285,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7598214596509933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2664
    },
    {
      "completion_length": 270.15179920196533,
      "epoch": 0.4469592187434511,
      "grad_norm": 0.16203992665582476,
      "kl": 0.20227813720703125,
      "learning_rate": 3.2913580246913583e-07,
      "loss": 0.0002,
      "reward": 1.748214341700077,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.761607164517045,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2666
    },
    {
      "completion_length": 257.2455472946167,
      "epoch": 0.44729452198331865,
      "grad_norm": 0.31669039361071327,
      "kl": 0.17920303344726562,
      "learning_rate": 3.293827160493827e-07,
      "loss": 0.0002,
      "reward": 1.803571492433548,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 2668
    },
    {
      "completion_length": 248.63393783569336,
      "epoch": 0.44762982522318623,
      "grad_norm": 0.14210074166673994,
      "kl": 0.06937408447265625,
      "learning_rate": 3.296296296296296e-07,
      "loss": 0.0001,
      "reward": 1.7625000774860382,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.766964316368103,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2670
    },
    {
      "completion_length": 250.87947368621826,
      "epoch": 0.44796512846305375,
      "grad_norm": 0.6957656299935766,
      "kl": 0.3000144958496094,
      "learning_rate": 3.298765432098765e-07,
      "loss": 0.0003,
      "reward": 1.7718750685453415,
      "reward_std": 0.059977806406095624,
      "rewards/equation_reward_func": 0.7776785902678967,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 2672
    },
    {
      "completion_length": 249.4732265472412,
      "epoch": 0.44830043170292133,
      "grad_norm": 0.28613974030082806,
      "kl": 0.361419677734375,
      "learning_rate": 3.3012345679012343e-07,
      "loss": 0.0004,
      "reward": 1.82321435213089,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.8276785984635353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2674
    },
    {
      "completion_length": 255.8571538925171,
      "epoch": 0.4486357349427889,
      "grad_norm": 0.15867424298767457,
      "kl": 0.026119232177734375,
      "learning_rate": 3.303703703703704e-07,
      "loss": 0.0,
      "reward": 1.7232143431901932,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7366071715950966,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2676
    },
    {
      "completion_length": 253.98215579986572,
      "epoch": 0.44897103818265643,
      "grad_norm": 0.7509986400487191,
      "kl": 0.14146804809570312,
      "learning_rate": 3.3061728395061726e-07,
      "loss": 0.0001,
      "reward": 1.7553572207689285,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7598214522004128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2678
    },
    {
      "completion_length": 252.8482255935669,
      "epoch": 0.449306341422524,
      "grad_norm": 0.38029749797861767,
      "kl": 0.07612991333007812,
      "learning_rate": 3.308641975308642e-07,
      "loss": 0.0001,
      "reward": 1.7267857864499092,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7312500365078449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2680
    },
    {
      "completion_length": 251.9285831451416,
      "epoch": 0.44964164466239154,
      "grad_norm": 0.39717612267346547,
      "kl": 0.026416778564453125,
      "learning_rate": 3.311111111111111e-07,
      "loss": 0.0,
      "reward": 1.7428572252392769,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7428571619093418,
      "rewards/format_reward_func": 1.0,
      "step": 2682
    },
    {
      "completion_length": 253.34822463989258,
      "epoch": 0.4499769479022591,
      "grad_norm": 0.643773820689036,
      "kl": 1.5482711791992188,
      "learning_rate": 3.31358024691358e-07,
      "loss": 0.0016,
      "reward": 1.7107143700122833,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7196429036557674,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2684
    },
    {
      "completion_length": 251.17858600616455,
      "epoch": 0.45031225114212664,
      "grad_norm": 0.34670280364640704,
      "kl": 0.042018890380859375,
      "learning_rate": 3.316049382716049e-07,
      "loss": 0.0,
      "reward": 1.7267857864499092,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7312500346451998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2686
    },
    {
      "completion_length": 245.5312614440918,
      "epoch": 0.4506475543819942,
      "grad_norm": 0.25670407143953416,
      "kl": 0.06372833251953125,
      "learning_rate": 3.318518518518518e-07,
      "loss": 0.0001,
      "reward": 1.8107143640518188,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8107143081724644,
      "rewards/format_reward_func": 1.0,
      "step": 2688
    },
    {
      "completion_length": 244.03572463989258,
      "epoch": 0.45098285762186174,
      "grad_norm": 0.23765421348038518,
      "kl": 0.027862548828125,
      "learning_rate": 3.3209876543209875e-07,
      "loss": 0.0,
      "reward": 1.7857143357396126,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143059372902,
      "rewards/format_reward_func": 1.0,
      "step": 2690
    },
    {
      "completion_length": 247.9151906967163,
      "epoch": 0.4513181608617293,
      "grad_norm": 0.33600943743007694,
      "kl": 0.559478759765625,
      "learning_rate": 3.323456790123457e-07,
      "loss": 0.0006,
      "reward": 1.7571429461240768,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 1.0,
      "step": 2692
    },
    {
      "completion_length": 237.571439743042,
      "epoch": 0.4516534641015969,
      "grad_norm": 0.5969202483502274,
      "kl": 0.6981925964355469,
      "learning_rate": 3.325925925925926e-07,
      "loss": 0.0007,
      "reward": 1.7446429505944252,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071671247482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2694
    },
    {
      "completion_length": 244.20090103149414,
      "epoch": 0.4519887673414644,
      "grad_norm": 1.2490766104629545,
      "kl": 0.19190216064453125,
      "learning_rate": 3.328395061728395e-07,
      "loss": 0.0002,
      "reward": 1.7553572207689285,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.7687500305473804,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2696
    },
    {
      "completion_length": 256.09375953674316,
      "epoch": 0.452324070581332,
      "grad_norm": 0.3169020842125548,
      "kl": 0.19821548461914062,
      "learning_rate": 3.3308641975308636e-07,
      "loss": 0.0002,
      "reward": 1.744642935693264,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7491071671247482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2698
    },
    {
      "completion_length": 251.383939743042,
      "epoch": 0.45265937382119953,
      "grad_norm": 0.17454350064179114,
      "kl": 0.209808349609375,
      "learning_rate": 3.333333333333333e-07,
      "loss": 0.0002,
      "reward": 1.796428643167019,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964286096394062,
      "rewards/format_reward_func": 1.0,
      "step": 2700
    },
    {
      "completion_length": 250.05805110931396,
      "epoch": 0.4529946770610671,
      "grad_norm": 0.270399027082715,
      "kl": 0.09576034545898438,
      "learning_rate": 3.3358024691358024e-07,
      "loss": 0.0001,
      "reward": 1.7964286655187607,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285910129547,
      "rewards/format_reward_func": 1.0,
      "step": 2702
    },
    {
      "completion_length": 235.66072463989258,
      "epoch": 0.45332998030093463,
      "grad_norm": 0.25215380030085344,
      "kl": 0.033405303955078125,
      "learning_rate": 3.3382716049382713e-07,
      "loss": 0.0,
      "reward": 1.7535715028643608,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 2704
    },
    {
      "completion_length": 258.839298248291,
      "epoch": 0.4536652835408022,
      "grad_norm": 0.507861698652949,
      "kl": 0.32860565185546875,
      "learning_rate": 3.3407407407407407e-07,
      "loss": 0.0003,
      "reward": 1.7625000923871994,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2706
    },
    {
      "completion_length": 248.53572368621826,
      "epoch": 0.4540005867806698,
      "grad_norm": 0.30873013848794667,
      "kl": 0.034900665283203125,
      "learning_rate": 3.3432098765432096e-07,
      "loss": 0.0,
      "reward": 1.7625000551342964,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7669643200933933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2708
    },
    {
      "completion_length": 242.1696548461914,
      "epoch": 0.4543358900205373,
      "grad_norm": 0.36290902428851235,
      "kl": 0.1857452392578125,
      "learning_rate": 3.345679012345679e-07,
      "loss": 0.0002,
      "reward": 1.7875000685453415,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7919643111526966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2710
    },
    {
      "completion_length": 265.44197273254395,
      "epoch": 0.4546711932604049,
      "grad_norm": 0.18967670433015887,
      "kl": 0.16097640991210938,
      "learning_rate": 3.348148148148148e-07,
      "loss": 0.0002,
      "reward": 1.7464286461472511,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286033064127,
      "rewards/format_reward_func": 1.0,
      "step": 2712
    },
    {
      "completion_length": 266.29019260406494,
      "epoch": 0.4550064965002724,
      "grad_norm": 0.2752528674594683,
      "kl": 0.06867218017578125,
      "learning_rate": 3.350617283950617e-07,
      "loss": 0.0001,
      "reward": 1.6803572475910187,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.6848214659839869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2714
    },
    {
      "completion_length": 249.27233505249023,
      "epoch": 0.45534179974014,
      "grad_norm": 0.17644565403173249,
      "kl": 0.35277557373046875,
      "learning_rate": 3.353086419753086e-07,
      "loss": 0.0004,
      "reward": 1.7946429252624512,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7991071715950966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2716
    },
    {
      "completion_length": 257.3259038925171,
      "epoch": 0.4556771029800075,
      "grad_norm": 0.18222821575997314,
      "kl": 0.06207275390625,
      "learning_rate": 3.3555555555555556e-07,
      "loss": 0.0001,
      "reward": 1.7410715147852898,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357424914837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2718
    },
    {
      "completion_length": 258.589298248291,
      "epoch": 0.4560124062198751,
      "grad_norm": 0.40252505186651977,
      "kl": 0.049884796142578125,
      "learning_rate": 3.3580246913580245e-07,
      "loss": 0.0,
      "reward": 1.7500000819563866,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7589285857975483,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2720
    },
    {
      "completion_length": 244.1696548461914,
      "epoch": 0.4563477094597427,
      "grad_norm": 0.44376256831190036,
      "kl": 0.5896148681640625,
      "learning_rate": 3.360493827160494e-07,
      "loss": 0.0006,
      "reward": 1.7410715147852898,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7455357406288385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2722
    },
    {
      "completion_length": 246.1919765472412,
      "epoch": 0.4566830126996102,
      "grad_norm": 0.29343191916555844,
      "kl": 0.07892227172851562,
      "learning_rate": 3.362962962962963e-07,
      "loss": 0.0001,
      "reward": 1.700000062584877,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7000000327825546,
      "rewards/format_reward_func": 1.0,
      "step": 2724
    },
    {
      "completion_length": 245.77679443359375,
      "epoch": 0.4570183159394778,
      "grad_norm": 0.2667512601424003,
      "kl": 0.09537887573242188,
      "learning_rate": 3.365432098765432e-07,
      "loss": 0.0001,
      "reward": 1.7660714909434319,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7705357372760773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2726
    },
    {
      "completion_length": 249.2634038925171,
      "epoch": 0.4573536191793453,
      "grad_norm": 0.38442369886561234,
      "kl": 0.36904144287109375,
      "learning_rate": 3.367901234567901e-07,
      "loss": 0.0004,
      "reward": 1.7571429386734962,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 1.0,
      "step": 2728
    },
    {
      "completion_length": 254.82144260406494,
      "epoch": 0.4576889224192129,
      "grad_norm": 0.3137500522359342,
      "kl": 0.061252593994140625,
      "learning_rate": 3.37037037037037e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7625000402331352,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2730
    },
    {
      "completion_length": 246.1785831451416,
      "epoch": 0.4580242256590804,
      "grad_norm": 0.2809572400397591,
      "kl": 0.11748123168945312,
      "learning_rate": 3.3728395061728394e-07,
      "loss": 0.0001,
      "reward": 1.79464291036129,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7991071753203869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2732
    },
    {
      "completion_length": 255.915189743042,
      "epoch": 0.458359528898948,
      "grad_norm": 0.2971358870435252,
      "kl": 0.028736114501953125,
      "learning_rate": 3.375308641975308e-07,
      "loss": 0.0,
      "reward": 1.8017857819795609,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8062500320374966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2734
    },
    {
      "completion_length": 257.7544765472412,
      "epoch": 0.45869483213881557,
      "grad_norm": 0.3195023386657291,
      "kl": 0.030345916748046875,
      "learning_rate": 3.3777777777777777e-07,
      "loss": 0.0,
      "reward": 1.7517857551574707,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7562500201165676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2736
    },
    {
      "completion_length": 246.2366180419922,
      "epoch": 0.4590301353786831,
      "grad_norm": 0.9720315208811928,
      "kl": 0.024379730224609375,
      "learning_rate": 3.380246913580247e-07,
      "loss": 0.0,
      "reward": 1.79464291036129,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7991071715950966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2738
    },
    {
      "completion_length": 250.5803689956665,
      "epoch": 0.45936543861855067,
      "grad_norm": 0.26845835890331887,
      "kl": 0.0423736572265625,
      "learning_rate": 3.382716049382716e-07,
      "loss": 0.0,
      "reward": 1.7696429267525673,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7741071656346321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2740
    },
    {
      "completion_length": 261.7544765472412,
      "epoch": 0.4597007418584182,
      "grad_norm": 0.19155715201137047,
      "kl": 0.08993148803710938,
      "learning_rate": 3.385185185185185e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285887777805,
      "rewards/format_reward_func": 1.0,
      "step": 2742
    },
    {
      "completion_length": 255.4375123977661,
      "epoch": 0.4600360450982858,
      "grad_norm": 0.24483028716397143,
      "kl": 0.07487106323242188,
      "learning_rate": 3.387654320987654e-07,
      "loss": 0.0001,
      "reward": 1.6910715326666832,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.704464316368103,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2744
    },
    {
      "completion_length": 251.3973331451416,
      "epoch": 0.4603713483381533,
      "grad_norm": 0.208316662402822,
      "kl": 0.02557373046875,
      "learning_rate": 3.390123456790123e-07,
      "loss": 0.0,
      "reward": 1.7714286372065544,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 2746
    },
    {
      "completion_length": 259.19197845458984,
      "epoch": 0.4607066515780209,
      "grad_norm": 0.1999658236725068,
      "kl": 0.037628173828125,
      "learning_rate": 3.3925925925925926e-07,
      "loss": 0.0,
      "reward": 1.7767857536673546,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7812500298023224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2748
    },
    {
      "completion_length": 247.6741180419922,
      "epoch": 0.46104195481788846,
      "grad_norm": 0.26139475662583994,
      "kl": 0.0295257568359375,
      "learning_rate": 3.3950617283950614e-07,
      "loss": 0.0,
      "reward": 1.7464286461472511,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 2750
    },
    {
      "completion_length": 258.54018783569336,
      "epoch": 0.461377258057756,
      "grad_norm": 0.13583613800255978,
      "kl": 0.0282440185546875,
      "learning_rate": 3.397530864197531e-07,
      "loss": 0.0,
      "reward": 1.7642857730388641,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7732143141329288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2752
    },
    {
      "completion_length": 251.55358123779297,
      "epoch": 0.46171256129762356,
      "grad_norm": 0.24682817309475594,
      "kl": 0.0307464599609375,
      "learning_rate": 3.4000000000000003e-07,
      "loss": 0.0,
      "reward": 1.8017857745289803,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.8062500357627869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2754
    },
    {
      "completion_length": 257.0848321914673,
      "epoch": 0.4620478645374911,
      "grad_norm": 0.15096030885914136,
      "kl": 0.02770233154296875,
      "learning_rate": 3.4024691358024686e-07,
      "loss": 0.0,
      "reward": 1.7678571790456772,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.7767857424914837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2756
    },
    {
      "completion_length": 252.02680206298828,
      "epoch": 0.46238316777735866,
      "grad_norm": 0.23076502570985755,
      "kl": 0.026927947998046875,
      "learning_rate": 3.404938271604938e-07,
      "loss": 0.0,
      "reward": 1.7375000640749931,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643215835094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2758
    },
    {
      "completion_length": 260.3348331451416,
      "epoch": 0.4627184710172262,
      "grad_norm": 0.34134054770080197,
      "kl": 0.050151824951171875,
      "learning_rate": 3.407407407407407e-07,
      "loss": 0.0001,
      "reward": 1.758928619325161,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7633928917348385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2760
    },
    {
      "completion_length": 255.1071548461914,
      "epoch": 0.46305377425709376,
      "grad_norm": 0.43200431098358355,
      "kl": 0.082275390625,
      "learning_rate": 3.4098765432098763e-07,
      "loss": 0.0001,
      "reward": 1.7482143715023994,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.7616071663796902,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2762
    },
    {
      "completion_length": 262.69643783569336,
      "epoch": 0.4633890774969613,
      "grad_norm": 0.37310963538113023,
      "kl": 0.16982269287109375,
      "learning_rate": 3.412345679012346e-07,
      "loss": 0.0002,
      "reward": 1.8035714700818062,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 2764
    },
    {
      "completion_length": 251.98215579986572,
      "epoch": 0.46372438073682887,
      "grad_norm": 0.5049977633001252,
      "kl": 0.03232574462890625,
      "learning_rate": 3.4148148148148146e-07,
      "loss": 0.0,
      "reward": 1.7392857745289803,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7392857521772385,
      "rewards/format_reward_func": 1.0,
      "step": 2766
    },
    {
      "completion_length": 249.11608219146729,
      "epoch": 0.46405968397669645,
      "grad_norm": 0.16067851846385553,
      "kl": 0.03436279296875,
      "learning_rate": 3.417283950617284e-07,
      "loss": 0.0,
      "reward": 1.7410714775323868,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7455357480794191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2768
    },
    {
      "completion_length": 247.83929634094238,
      "epoch": 0.46439498721656397,
      "grad_norm": 0.23395496207649558,
      "kl": 0.054126739501953125,
      "learning_rate": 3.419753086419753e-07,
      "loss": 0.0001,
      "reward": 1.7482143566012383,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2770
    },
    {
      "completion_length": 250.0937623977661,
      "epoch": 0.46473029045643155,
      "grad_norm": 0.18523182105694547,
      "kl": 0.024806976318359375,
      "learning_rate": 3.422222222222222e-07,
      "loss": 0.0,
      "reward": 1.821428619325161,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8214285895228386,
      "rewards/format_reward_func": 1.0,
      "step": 2772
    },
    {
      "completion_length": 256.6875114440918,
      "epoch": 0.4650655936962991,
      "grad_norm": 0.24536718633510723,
      "kl": 0.19739151000976562,
      "learning_rate": 3.424691358024691e-07,
      "loss": 0.0002,
      "reward": 1.7464286461472511,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.746428593993187,
      "rewards/format_reward_func": 1.0,
      "step": 2774
    },
    {
      "completion_length": 256.1562614440918,
      "epoch": 0.46540089693616665,
      "grad_norm": 0.3292800179362696,
      "kl": 0.12223052978515625,
      "learning_rate": 3.42716049382716e-07,
      "loss": 0.0001,
      "reward": 1.714285783469677,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7232143152505159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2776
    },
    {
      "completion_length": 242.8571538925171,
      "epoch": 0.4657362001760342,
      "grad_norm": 0.40450119722326106,
      "kl": 0.03858184814453125,
      "learning_rate": 3.4296296296296295e-07,
      "loss": 0.0,
      "reward": 1.8250000551342964,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8250000216066837,
      "rewards/format_reward_func": 1.0,
      "step": 2778
    },
    {
      "completion_length": 250.78572463989258,
      "epoch": 0.46607150341590176,
      "grad_norm": 0.40163561926349534,
      "kl": 0.05098724365234375,
      "learning_rate": 3.432098765432099e-07,
      "loss": 0.0001,
      "reward": 1.7250000834465027,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7250000350177288,
      "rewards/format_reward_func": 1.0,
      "step": 2780
    },
    {
      "completion_length": 247.16072750091553,
      "epoch": 0.46640680665576933,
      "grad_norm": 0.2388646035496535,
      "kl": 0.050304412841796875,
      "learning_rate": 3.434567901234568e-07,
      "loss": 0.0001,
      "reward": 1.7982143610715866,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.8026785925030708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2782
    },
    {
      "completion_length": 251.508939743042,
      "epoch": 0.46674210989563686,
      "grad_norm": 0.23875992787468778,
      "kl": 0.04918670654296875,
      "learning_rate": 3.4370370370370367e-07,
      "loss": 0.0,
      "reward": 1.7571429088711739,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7660714592784643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2784
    },
    {
      "completion_length": 251.3437623977661,
      "epoch": 0.46707741313550444,
      "grad_norm": 0.188319172046665,
      "kl": 0.3220100402832031,
      "learning_rate": 3.4395061728395056e-07,
      "loss": 0.0003,
      "reward": 1.7785715013742447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714510828257,
      "rewards/format_reward_func": 1.0,
      "step": 2786
    },
    {
      "completion_length": 247.6607255935669,
      "epoch": 0.46741271637537196,
      "grad_norm": 0.3122224174936983,
      "kl": 0.0294189453125,
      "learning_rate": 3.441975308641975e-07,
      "loss": 0.0,
      "reward": 1.776785746216774,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7812500167638063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2788
    },
    {
      "completion_length": 254.26787090301514,
      "epoch": 0.46774801961523954,
      "grad_norm": 0.22040518881569834,
      "kl": 0.098358154296875,
      "learning_rate": 3.4444444444444444e-07,
      "loss": 0.0001,
      "reward": 1.7178572341799736,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7178571745753288,
      "rewards/format_reward_func": 1.0,
      "step": 2790
    },
    {
      "completion_length": 242.29911708831787,
      "epoch": 0.46808332285510706,
      "grad_norm": 0.3139530838726713,
      "kl": 0.039592742919921875,
      "learning_rate": 3.4469135802469133e-07,
      "loss": 0.0,
      "reward": 1.7714286297559738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 2792
    },
    {
      "completion_length": 254.37947463989258,
      "epoch": 0.46841862609497464,
      "grad_norm": 0.2681590517820234,
      "kl": 0.03741455078125,
      "learning_rate": 3.4493827160493827e-07,
      "loss": 0.0,
      "reward": 1.7857143506407738,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7946428842842579,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2794
    },
    {
      "completion_length": 256.9241180419922,
      "epoch": 0.4687539293348422,
      "grad_norm": 0.17318988896415513,
      "kl": 0.1388397216796875,
      "learning_rate": 3.451851851851852e-07,
      "loss": 0.0001,
      "reward": 1.7446429133415222,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.749107176437974,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2796
    },
    {
      "completion_length": 264.62500953674316,
      "epoch": 0.46908923257470975,
      "grad_norm": 0.5645059371676063,
      "kl": 0.10883331298828125,
      "learning_rate": 3.4543209876543205e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7946428842842579,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2798
    },
    {
      "completion_length": 264.3080463409424,
      "epoch": 0.4694245358145773,
      "grad_norm": 0.3930393143745379,
      "kl": 0.0624847412109375,
      "learning_rate": 3.45679012345679e-07,
      "loss": 0.0001,
      "reward": 1.698214367032051,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7026786059141159,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2800
    },
    {
      "completion_length": 257.2098340988159,
      "epoch": 0.46975983905444485,
      "grad_norm": 0.2631303652587504,
      "kl": 0.04084014892578125,
      "learning_rate": 3.459259259259259e-07,
      "loss": 0.0,
      "reward": 1.7107143625617027,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7196428962051868,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2802
    },
    {
      "completion_length": 257.65179443359375,
      "epoch": 0.47009514229431243,
      "grad_norm": 0.22529626467128386,
      "kl": 0.0889739990234375,
      "learning_rate": 3.461728395061728e-07,
      "loss": 0.0001,
      "reward": 1.7732143625617027,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7776785921305418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2804
    },
    {
      "completion_length": 254.2678689956665,
      "epoch": 0.47043044553417995,
      "grad_norm": 0.6203961669806676,
      "kl": 0.20491790771484375,
      "learning_rate": 3.4641975308641976e-07,
      "loss": 0.0002,
      "reward": 1.7517857924103737,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500387430191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2806
    },
    {
      "completion_length": 256.9107275009155,
      "epoch": 0.47076574877404753,
      "grad_norm": 0.24761485503433714,
      "kl": 0.09076690673828125,
      "learning_rate": 3.4666666666666665e-07,
      "loss": 0.0001,
      "reward": 1.7839286252856255,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2808
    },
    {
      "completion_length": 258.24108028411865,
      "epoch": 0.4711010520139151,
      "grad_norm": 0.2496260686532331,
      "kl": 0.03143310546875,
      "learning_rate": 3.469135802469136e-07,
      "loss": 0.0,
      "reward": 1.7482143864035606,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7526785917580128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2810
    },
    {
      "completion_length": 254.9509038925171,
      "epoch": 0.47143635525378264,
      "grad_norm": 0.6028251611292678,
      "kl": 0.3297767639160156,
      "learning_rate": 3.4716049382716053e-07,
      "loss": 0.0003,
      "reward": 1.7464286237955093,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.755357164889574,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2812
    },
    {
      "completion_length": 269.39286708831787,
      "epoch": 0.4717716584936502,
      "grad_norm": 0.3415395676744013,
      "kl": 0.1169281005859375,
      "learning_rate": 3.4740740740740737e-07,
      "loss": 0.0001,
      "reward": 1.7392857745289803,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7571428753435612,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2814
    },
    {
      "completion_length": 263.53125953674316,
      "epoch": 0.47210696173351774,
      "grad_norm": 0.23960314502568703,
      "kl": 0.050708770751953125,
      "learning_rate": 3.476543209876543e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.0909137288108468,
      "rewards/equation_reward_func": 0.7750000171363354,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2816
    },
    {
      "completion_length": 268.6651887893677,
      "epoch": 0.4724422649733853,
      "grad_norm": 0.22502083720491892,
      "kl": 0.3942832946777344,
      "learning_rate": 3.479012345679012e-07,
      "loss": 0.0004,
      "reward": 1.7125000581145287,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7258929070085287,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2818
    },
    {
      "completion_length": 266.383939743042,
      "epoch": 0.47277756821325284,
      "grad_norm": 0.3852008894067241,
      "kl": 0.5472488403320312,
      "learning_rate": 3.4814814814814814e-07,
      "loss": 0.0005,
      "reward": 1.7785715088248253,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 2820
    },
    {
      "completion_length": 277.2678737640381,
      "epoch": 0.4731128714531204,
      "grad_norm": 0.4898862429233096,
      "kl": 0.44490814208984375,
      "learning_rate": 3.483950617283951e-07,
      "loss": 0.0004,
      "reward": 1.7464286535978317,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2822
    },
    {
      "completion_length": 271.20983505249023,
      "epoch": 0.47344817469298794,
      "grad_norm": 0.3050807925241347,
      "kl": 0.32114410400390625,
      "learning_rate": 3.4864197530864197e-07,
      "loss": 0.0003,
      "reward": 1.7107143327593803,
      "reward_std": 0.05555839091539383,
      "rewards/equation_reward_func": 0.7285714615136385,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2824
    },
    {
      "completion_length": 262.2857255935669,
      "epoch": 0.4737834779328555,
      "grad_norm": 0.32417959842074145,
      "kl": 0.27597808837890625,
      "learning_rate": 3.488888888888889e-07,
      "loss": 0.0003,
      "reward": 1.7196429446339607,
      "reward_std": 0.06313453428447247,
      "rewards/equation_reward_func": 0.7330357395112514,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2826
    },
    {
      "completion_length": 256.62947940826416,
      "epoch": 0.4741187811727231,
      "grad_norm": 0.43158639531492315,
      "kl": 0.3496856689453125,
      "learning_rate": 3.4913580246913574e-07,
      "loss": 0.0003,
      "reward": 1.7857143506407738,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.7946428786963224,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2828
    },
    {
      "completion_length": 257.06251430511475,
      "epoch": 0.4744540844125906,
      "grad_norm": 0.16878279995351914,
      "kl": 0.02948760986328125,
      "learning_rate": 3.493827160493827e-07,
      "loss": 0.0,
      "reward": 1.7964286133646965,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 2830
    },
    {
      "completion_length": 268.3035821914673,
      "epoch": 0.4747893876524582,
      "grad_norm": 1.3822497181999542,
      "kl": 1.5914840698242188,
      "learning_rate": 3.496296296296296e-07,
      "loss": 0.0016,
      "reward": 1.723214365541935,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7366071864962578,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2832
    },
    {
      "completion_length": 272.46876430511475,
      "epoch": 0.47512469089232573,
      "grad_norm": 0.31405924815564323,
      "kl": 0.60906982421875,
      "learning_rate": 3.498765432098765e-07,
      "loss": 0.0006,
      "reward": 1.721428632736206,
      "reward_std": 0.08586296439170837,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2834
    },
    {
      "completion_length": 265.4419765472412,
      "epoch": 0.4754599941321933,
      "grad_norm": 0.3334541089600988,
      "kl": 0.1193695068359375,
      "learning_rate": 3.5012345679012345e-07,
      "loss": 0.0001,
      "reward": 1.7196429297327995,
      "reward_std": 0.09343910869210958,
      "rewards/equation_reward_func": 0.7330357357859612,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2836
    },
    {
      "completion_length": 261.68751430511475,
      "epoch": 0.47579529737206083,
      "grad_norm": 0.2359020883684203,
      "kl": 0.02530670166015625,
      "learning_rate": 3.503703703703704e-07,
      "loss": 0.0,
      "reward": 1.7392857819795609,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7392857521772385,
      "rewards/format_reward_func": 1.0,
      "step": 2838
    },
    {
      "completion_length": 270.17858123779297,
      "epoch": 0.4761306006119284,
      "grad_norm": 0.19232319568658027,
      "kl": 0.028003692626953125,
      "learning_rate": 3.506172839506173e-07,
      "loss": 0.0,
      "reward": 1.6857143491506577,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7035714518278837,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2840
    },
    {
      "completion_length": 248.52679824829102,
      "epoch": 0.476465903851796,
      "grad_norm": 0.14569116729208106,
      "kl": 0.08068084716796875,
      "learning_rate": 3.5086419753086417e-07,
      "loss": 0.0001,
      "reward": 1.8071429133415222,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8071428835391998,
      "rewards/format_reward_func": 1.0,
      "step": 2842
    },
    {
      "completion_length": 254.58929634094238,
      "epoch": 0.4768012070916635,
      "grad_norm": 1.0018321660589764,
      "kl": 0.047954559326171875,
      "learning_rate": 3.5111111111111106e-07,
      "loss": 0.0,
      "reward": 1.8000000640749931,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.8089286014437675,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2844
    },
    {
      "completion_length": 242.83483219146729,
      "epoch": 0.4771365103315311,
      "grad_norm": 0.1876792247373185,
      "kl": 0.19379425048828125,
      "learning_rate": 3.51358024691358e-07,
      "loss": 0.0002,
      "reward": 1.7428572177886963,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571693599224,
      "rewards/format_reward_func": 1.0,
      "step": 2846
    },
    {
      "completion_length": 257.9866199493408,
      "epoch": 0.4774718135713986,
      "grad_norm": 0.21949993388796094,
      "kl": 0.054508209228515625,
      "learning_rate": 3.5160493827160494e-07,
      "loss": 0.0001,
      "reward": 1.7696429416537285,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7741071656346321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2848
    },
    {
      "completion_length": 248.25447463989258,
      "epoch": 0.4778071168112662,
      "grad_norm": 0.47622915884966727,
      "kl": 0.5040206909179688,
      "learning_rate": 3.5185185185185183e-07,
      "loss": 0.0005,
      "reward": 1.7625000849366188,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643089175224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2850
    },
    {
      "completion_length": 249.97768783569336,
      "epoch": 0.4781424200511337,
      "grad_norm": 0.40265000030374465,
      "kl": 0.05511474609375,
      "learning_rate": 3.5209876543209877e-07,
      "loss": 0.0001,
      "reward": 1.7803572043776512,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7848214618861675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2852
    },
    {
      "completion_length": 264.2812614440918,
      "epoch": 0.4784777232910013,
      "grad_norm": 0.38387369362331086,
      "kl": 0.31607818603515625,
      "learning_rate": 3.5234567901234566e-07,
      "loss": 0.0003,
      "reward": 1.7267857939004898,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.7401785887777805,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2854
    },
    {
      "completion_length": 257.3392972946167,
      "epoch": 0.4788130265308689,
      "grad_norm": 0.5951750074514578,
      "kl": 0.13494873046875,
      "learning_rate": 3.5259259259259255e-07,
      "loss": 0.0001,
      "reward": 1.741071492433548,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7455357499420643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2856
    },
    {
      "completion_length": 254.5134048461914,
      "epoch": 0.4791483297707364,
      "grad_norm": 0.17450689785228546,
      "kl": 0.03231048583984375,
      "learning_rate": 3.528395061728395e-07,
      "loss": 0.0,
      "reward": 1.741071492433548,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357536673546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2858
    },
    {
      "completion_length": 257.2321529388428,
      "epoch": 0.479483633010604,
      "grad_norm": 0.1944576284948148,
      "kl": 0.288116455078125,
      "learning_rate": 3.530864197530864e-07,
      "loss": 0.0003,
      "reward": 1.7428571954369545,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7517857421189547,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2860
    },
    {
      "completion_length": 260.0357275009155,
      "epoch": 0.4798189362504715,
      "grad_norm": 0.19695027022142356,
      "kl": 0.067718505859375,
      "learning_rate": 3.533333333333333e-07,
      "loss": 0.0001,
      "reward": 1.805357202887535,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8098214454948902,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2862
    },
    {
      "completion_length": 255.81251430511475,
      "epoch": 0.4801542394903391,
      "grad_norm": 0.23358145138386263,
      "kl": 0.33197021484375,
      "learning_rate": 3.5358024691358026e-07,
      "loss": 0.0003,
      "reward": 1.733928643167019,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7383928894996643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2864
    },
    {
      "completion_length": 248.89733505249023,
      "epoch": 0.4804895427302066,
      "grad_norm": 0.349013914148386,
      "kl": 0.0921783447265625,
      "learning_rate": 3.5382716049382715e-07,
      "loss": 0.0001,
      "reward": 1.725000075995922,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7339286021888256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2866
    },
    {
      "completion_length": 246.35715675354004,
      "epoch": 0.4808248459700742,
      "grad_norm": 0.5646698298864009,
      "kl": 0.31024169921875,
      "learning_rate": 3.540740740740741e-07,
      "loss": 0.0003,
      "reward": 1.6839286461472511,
      "reward_std": 0.10354063473641872,
      "rewards/equation_reward_func": 0.6973214633762836,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2868
    },
    {
      "completion_length": 258.14287090301514,
      "epoch": 0.48116014920994177,
      "grad_norm": 0.52150068072927,
      "kl": 0.208526611328125,
      "learning_rate": 3.5432098765432093e-07,
      "loss": 0.0002,
      "reward": 1.73214291036129,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7410714700818062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2870
    },
    {
      "completion_length": 249.07590198516846,
      "epoch": 0.4814954524498093,
      "grad_norm": 0.31669922163507147,
      "kl": 0.0349273681640625,
      "learning_rate": 3.5456790123456787e-07,
      "loss": 0.0,
      "reward": 1.7464286237955093,
      "reward_std": 0.07576144114136696,
      "rewards/equation_reward_func": 0.755357189103961,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2872
    },
    {
      "completion_length": 249.3392972946167,
      "epoch": 0.48183075568967687,
      "grad_norm": 0.25581827702063864,
      "kl": 0.05368804931640625,
      "learning_rate": 3.548148148148148e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428954601288,
      "rewards/format_reward_func": 1.0,
      "step": 2874
    },
    {
      "completion_length": 259.40179538726807,
      "epoch": 0.4821660589295444,
      "grad_norm": 0.2812439109626435,
      "kl": 1.0510787963867188,
      "learning_rate": 3.550617283950617e-07,
      "loss": 0.0011,
      "reward": 1.6892857775092125,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.6982143223285675,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2876
    },
    {
      "completion_length": 250.75447273254395,
      "epoch": 0.482501362169412,
      "grad_norm": 0.19207748486652276,
      "kl": 0.322998046875,
      "learning_rate": 3.5530864197530864e-07,
      "loss": 0.0003,
      "reward": 1.7035715207457542,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7035714648663998,
      "rewards/format_reward_func": 1.0,
      "step": 2878
    },
    {
      "completion_length": 260.4017963409424,
      "epoch": 0.4828366654092795,
      "grad_norm": 0.22444574368780876,
      "kl": 0.17897415161132812,
      "learning_rate": 3.5555555555555553e-07,
      "loss": 0.0002,
      "reward": 1.7642857730388641,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7732143364846706,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2880
    },
    {
      "completion_length": 250.5178680419922,
      "epoch": 0.4831719686491471,
      "grad_norm": 0.5271841775692403,
      "kl": 0.5541343688964844,
      "learning_rate": 3.5580246913580247e-07,
      "loss": 0.0006,
      "reward": 1.7732143476605415,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7776785977184772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2882
    },
    {
      "completion_length": 255.05804634094238,
      "epoch": 0.4835072718890146,
      "grad_norm": 0.465173051193132,
      "kl": 0.31304931640625,
      "learning_rate": 3.5604938271604936e-07,
      "loss": 0.0003,
      "reward": 1.694642923772335,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.6991071719676256,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2884
    },
    {
      "completion_length": 261.0982255935669,
      "epoch": 0.4838425751288822,
      "grad_norm": 0.3209358019665937,
      "kl": 0.6431045532226562,
      "learning_rate": 3.5629629629629625e-07,
      "loss": 0.0006,
      "reward": 1.742857202887535,
      "reward_std": 0.1010152529925108,
      "rewards/equation_reward_func": 0.7607143074274063,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2886
    },
    {
      "completion_length": 253.11608409881592,
      "epoch": 0.48417787836874976,
      "grad_norm": 0.2680813222428663,
      "kl": 0.21021652221679688,
      "learning_rate": 3.565432098765432e-07,
      "loss": 0.0002,
      "reward": 1.7303572222590446,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7348214685916901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2888
    },
    {
      "completion_length": 253.08929443359375,
      "epoch": 0.4845131816086173,
      "grad_norm": 0.5583623489192696,
      "kl": 0.2989044189453125,
      "learning_rate": 3.5679012345679013e-07,
      "loss": 0.0003,
      "reward": 1.753571480512619,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7625000402331352,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2890
    },
    {
      "completion_length": 248.2455472946167,
      "epoch": 0.48484848484848486,
      "grad_norm": 0.2820007992710127,
      "kl": 0.31207275390625,
      "learning_rate": 3.57037037037037e-07,
      "loss": 0.0003,
      "reward": 1.7875000536441803,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7910714484751225,
      "rewards/format_reward_func": 0.9964285790920258,
      "step": 2892
    },
    {
      "completion_length": 253.58483409881592,
      "epoch": 0.4851837880883524,
      "grad_norm": 0.7035434638322244,
      "kl": 0.6702957153320312,
      "learning_rate": 3.5728395061728396e-07,
      "loss": 0.0007,
      "reward": 1.7000000923871994,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7089285925030708,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2894
    },
    {
      "completion_length": 256.2410831451416,
      "epoch": 0.48551909132821996,
      "grad_norm": 0.3080211232059757,
      "kl": 0.267303466796875,
      "learning_rate": 3.5753086419753085e-07,
      "loss": 0.0003,
      "reward": 1.7410714998841286,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7544643096625805,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2896
    },
    {
      "completion_length": 257.6384029388428,
      "epoch": 0.4858543945680875,
      "grad_norm": 0.28553151949283306,
      "kl": 0.15687179565429688,
      "learning_rate": 3.5777777777777773e-07,
      "loss": 0.0002,
      "reward": 1.7339286133646965,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.7473214752972126,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2898
    },
    {
      "completion_length": 255.65625667572021,
      "epoch": 0.48618969780795507,
      "grad_norm": 0.21765843503980004,
      "kl": 0.5928115844726562,
      "learning_rate": 3.580246913580247e-07,
      "loss": 0.0006,
      "reward": 1.6803572103381157,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.6937500424683094,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2900
    },
    {
      "completion_length": 254.33929920196533,
      "epoch": 0.48652500104782265,
      "grad_norm": 0.44552416758581037,
      "kl": 0.3790931701660156,
      "learning_rate": 3.5827160493827156e-07,
      "loss": 0.0004,
      "reward": 1.6964286491274834,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7053571753203869,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2902
    },
    {
      "completion_length": 253.09376335144043,
      "epoch": 0.48686030428769017,
      "grad_norm": 0.9647608524452777,
      "kl": 1.1598663330078125,
      "learning_rate": 3.585185185185185e-07,
      "loss": 0.0012,
      "reward": 1.714285783469677,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7142857518047094,
      "rewards/format_reward_func": 1.0,
      "step": 2904
    },
    {
      "completion_length": 255.12054920196533,
      "epoch": 0.48719560752755775,
      "grad_norm": 0.29048605992776716,
      "kl": 0.21368408203125,
      "learning_rate": 3.587654320987654e-07,
      "loss": 0.0002,
      "reward": 1.7589286342263222,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7723214663565159,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2906
    },
    {
      "completion_length": 255.02679920196533,
      "epoch": 0.4875309107674253,
      "grad_norm": 0.2027351239625565,
      "kl": 0.7143478393554688,
      "learning_rate": 3.5901234567901234e-07,
      "loss": 0.0007,
      "reward": 1.7660714983940125,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.7794643118977547,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2908
    },
    {
      "completion_length": 252.3973331451416,
      "epoch": 0.48786621400729285,
      "grad_norm": 0.19049513041306962,
      "kl": 0.08922576904296875,
      "learning_rate": 3.592592592592593e-07,
      "loss": 0.0001,
      "reward": 1.791071504354477,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.795535746961832,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2910
    },
    {
      "completion_length": 256.40625953674316,
      "epoch": 0.4882015172471604,
      "grad_norm": 0.35376853781284395,
      "kl": 0.48523712158203125,
      "learning_rate": 3.5950617283950616e-07,
      "loss": 0.0005,
      "reward": 1.769642911851406,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7741071805357933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2912
    },
    {
      "completion_length": 263.09822273254395,
      "epoch": 0.48853682048702796,
      "grad_norm": 0.36369623998455325,
      "kl": 0.08524322509765625,
      "learning_rate": 3.5975308641975305e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 2914
    },
    {
      "completion_length": 245.4553689956665,
      "epoch": 0.48887212372689554,
      "grad_norm": 0.1363731743599683,
      "kl": 0.03830718994140625,
      "learning_rate": 3.6e-07,
      "loss": 0.0,
      "reward": 1.7232143729925156,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7276785969734192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2916
    },
    {
      "completion_length": 259.6875123977661,
      "epoch": 0.48920742696676306,
      "grad_norm": 0.22073589348512726,
      "kl": 0.175567626953125,
      "learning_rate": 3.602469135802469e-07,
      "loss": 0.0002,
      "reward": 1.7821428999304771,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.791071455925703,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2918
    },
    {
      "completion_length": 256.28572845458984,
      "epoch": 0.48954273020663064,
      "grad_norm": 0.2817423794616662,
      "kl": 0.23409271240234375,
      "learning_rate": 3.604938271604938e-07,
      "loss": 0.0002,
      "reward": 1.782142922282219,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428775787354,
      "rewards/format_reward_func": 1.0,
      "step": 2920
    },
    {
      "completion_length": 262.0446500778198,
      "epoch": 0.48987803344649816,
      "grad_norm": 0.5882471776016766,
      "kl": 0.3173255920410156,
      "learning_rate": 3.607407407407407e-07,
      "loss": 0.0003,
      "reward": 1.6803572252392769,
      "reward_std": 0.06818529777228832,
      "rewards/equation_reward_func": 0.6937500387430191,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2922
    },
    {
      "completion_length": 247.25447463989258,
      "epoch": 0.49021333668636574,
      "grad_norm": 0.2754932860486546,
      "kl": 0.6283416748046875,
      "learning_rate": 3.6098765432098765e-07,
      "loss": 0.0006,
      "reward": 1.758928656578064,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7633928880095482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2924
    },
    {
      "completion_length": 242.98661994934082,
      "epoch": 0.49054863992623327,
      "grad_norm": 0.31778700835528595,
      "kl": 0.11323165893554688,
      "learning_rate": 3.612345679012346e-07,
      "loss": 0.0001,
      "reward": 1.7339286729693413,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7383928932249546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2926
    },
    {
      "completion_length": 255.58929920196533,
      "epoch": 0.49088394316610084,
      "grad_norm": 0.2705317676366948,
      "kl": 1.1042404174804688,
      "learning_rate": 3.6148148148148143e-07,
      "loss": 0.0011,
      "reward": 1.7803571820259094,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7848214693367481,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2928
    },
    {
      "completion_length": 248.96430015563965,
      "epoch": 0.4912192464059684,
      "grad_norm": 0.1774856462598626,
      "kl": 0.22403717041015625,
      "learning_rate": 3.6172839506172837e-07,
      "loss": 0.0002,
      "reward": 1.810714341700077,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8107143193483353,
      "rewards/format_reward_func": 1.0,
      "step": 2930
    },
    {
      "completion_length": 250.04911613464355,
      "epoch": 0.49155454964583595,
      "grad_norm": 0.2009737631916188,
      "kl": 0.3294029235839844,
      "learning_rate": 3.6197530864197526e-07,
      "loss": 0.0003,
      "reward": 1.7714286297559738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 2932
    },
    {
      "completion_length": 252.9062623977661,
      "epoch": 0.4918898528857035,
      "grad_norm": 0.24142261752824526,
      "kl": 0.37865447998046875,
      "learning_rate": 3.622222222222222e-07,
      "loss": 0.0004,
      "reward": 1.7660714760422707,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7705357484519482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2934
    },
    {
      "completion_length": 245.23215198516846,
      "epoch": 0.49222515612557105,
      "grad_norm": 0.3224585490033016,
      "kl": 1.2538604736328125,
      "learning_rate": 3.6246913580246914e-07,
      "loss": 0.0013,
      "reward": 1.8285714834928513,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8285714499652386,
      "rewards/format_reward_func": 1.0,
      "step": 2936
    },
    {
      "completion_length": 242.7098331451416,
      "epoch": 0.49256045936543863,
      "grad_norm": 0.07482152881354606,
      "kl": 0.6891937255859375,
      "learning_rate": 3.6271604938271603e-07,
      "loss": 0.0007,
      "reward": 1.7821429446339607,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 2938
    },
    {
      "completion_length": 250.76340293884277,
      "epoch": 0.49289576260530615,
      "grad_norm": 0.3688486923883066,
      "kl": 0.46811676025390625,
      "learning_rate": 3.6296296296296297e-07,
      "loss": 0.0005,
      "reward": 1.7053572237491608,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.709821468219161,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2940
    },
    {
      "completion_length": 247.9598331451416,
      "epoch": 0.49323106584517373,
      "grad_norm": 0.38334694612665543,
      "kl": 0.5134353637695312,
      "learning_rate": 3.6320987654320986e-07,
      "loss": 0.0005,
      "reward": 1.814285770058632,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8142857346683741,
      "rewards/format_reward_func": 1.0,
      "step": 2942
    },
    {
      "completion_length": 253.24108505249023,
      "epoch": 0.49356636908504126,
      "grad_norm": 0.2466986529569209,
      "kl": 0.073577880859375,
      "learning_rate": 3.6345679012345675e-07,
      "loss": 0.0001,
      "reward": 1.682142935693264,
      "reward_std": 0.09596449043601751,
      "rewards/equation_reward_func": 0.7000000309199095,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2944
    },
    {
      "completion_length": 253.5044755935669,
      "epoch": 0.49390167232490884,
      "grad_norm": 0.22385096048864217,
      "kl": 0.6382865905761719,
      "learning_rate": 3.637037037037037e-07,
      "loss": 0.0006,
      "reward": 1.7464286461472511,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7553571835160255,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2946
    },
    {
      "completion_length": 244.7544755935669,
      "epoch": 0.4942369755647764,
      "grad_norm": 0.3057270423350129,
      "kl": 0.124908447265625,
      "learning_rate": 3.639506172839506e-07,
      "loss": 0.0001,
      "reward": 1.7678572237491608,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 2948
    },
    {
      "completion_length": 251.68304634094238,
      "epoch": 0.49457227880464394,
      "grad_norm": 0.8529840796001801,
      "kl": 1.887176513671875,
      "learning_rate": 3.641975308641975e-07,
      "loss": 0.0019,
      "reward": 1.7017858028411865,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7062500342726707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2950
    },
    {
      "completion_length": 250.33929824829102,
      "epoch": 0.4949075820445115,
      "grad_norm": 0.32756161445028503,
      "kl": 0.14838409423828125,
      "learning_rate": 3.6444444444444446e-07,
      "loss": 0.0001,
      "reward": 1.7089286372065544,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7133928779512644,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2952
    },
    {
      "completion_length": 259.4419775009155,
      "epoch": 0.49524288528437904,
      "grad_norm": 0.7213068467496084,
      "kl": 0.8767471313476562,
      "learning_rate": 3.6469135802469135e-07,
      "loss": 0.0009,
      "reward": 1.7714286148548126,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7803571783006191,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2954
    },
    {
      "completion_length": 253.06250953674316,
      "epoch": 0.4955781885242466,
      "grad_norm": 0.3223780611086043,
      "kl": 0.16445159912109375,
      "learning_rate": 3.6493827160493824e-07,
      "loss": 0.0002,
      "reward": 1.7607143223285675,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.769642885774374,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2956
    },
    {
      "completion_length": 250.9732265472412,
      "epoch": 0.49591349176411414,
      "grad_norm": 0.31015288975976857,
      "kl": 0.32781982421875,
      "learning_rate": 3.651851851851851e-07,
      "loss": 0.0003,
      "reward": 1.7000000774860382,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7089285999536514,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2958
    },
    {
      "completion_length": 244.3750123977661,
      "epoch": 0.4962487950039817,
      "grad_norm": 0.24747093123257072,
      "kl": 0.4291877746582031,
      "learning_rate": 3.6543209876543207e-07,
      "loss": 0.0004,
      "reward": 1.7482143491506577,
      "reward_std": 0.06313453428447247,
      "rewards/equation_reward_func": 0.7616071701049805,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2960
    },
    {
      "completion_length": 257.76786708831787,
      "epoch": 0.4965840982438493,
      "grad_norm": 0.3119423738396043,
      "kl": 0.40831756591796875,
      "learning_rate": 3.65679012345679e-07,
      "loss": 0.0004,
      "reward": 1.732142947614193,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2962
    },
    {
      "completion_length": 259.17858505249023,
      "epoch": 0.4969194014837168,
      "grad_norm": 0.1830517098733492,
      "kl": 0.05135345458984375,
      "learning_rate": 3.659259259259259e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.8035714589059353,
      "rewards/format_reward_func": 0.9821428619325161,
      "step": 2964
    },
    {
      "completion_length": 243.9419755935669,
      "epoch": 0.4972547047235844,
      "grad_norm": 0.21681102140064856,
      "kl": 0.10626220703125,
      "learning_rate": 3.6617283950617284e-07,
      "loss": 0.0001,
      "reward": 1.7964286133646965,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.8053571581840515,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2966
    },
    {
      "completion_length": 255.5312614440918,
      "epoch": 0.49759000796345193,
      "grad_norm": 0.30420774625152075,
      "kl": 0.27170562744140625,
      "learning_rate": 3.664197530864198e-07,
      "loss": 0.0003,
      "reward": 1.7839286178350449,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.7973214574158192,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2968
    },
    {
      "completion_length": 260.1562623977661,
      "epoch": 0.4979253112033195,
      "grad_norm": 0.5679008529737339,
      "kl": 0.6922416687011719,
      "learning_rate": 3.666666666666666e-07,
      "loss": 0.0007,
      "reward": 1.703571505844593,
      "reward_std": 0.09091372601687908,
      "rewards/equation_reward_func": 0.7303571589291096,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 2970
    },
    {
      "completion_length": 256.95983123779297,
      "epoch": 0.49826061444318703,
      "grad_norm": 0.21251929924314933,
      "kl": 0.3111114501953125,
      "learning_rate": 3.6691358024691356e-07,
      "loss": 0.0003,
      "reward": 1.7339286282658577,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7473214603960514,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2972
    },
    {
      "completion_length": 250.0044755935669,
      "epoch": 0.4985959176830546,
      "grad_norm": 0.2855125484052691,
      "kl": 0.6980438232421875,
      "learning_rate": 3.6716049382716044e-07,
      "loss": 0.0007,
      "reward": 1.7303571924567223,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7437500283122063,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2974
    },
    {
      "completion_length": 266.54019260406494,
      "epoch": 0.4989312209229222,
      "grad_norm": 0.4551964805783462,
      "kl": 0.3116912841796875,
      "learning_rate": 3.674074074074074e-07,
      "loss": 0.0003,
      "reward": 1.8089286163449287,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.8223214447498322,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2976
    },
    {
      "completion_length": 257.308048248291,
      "epoch": 0.4992665241627897,
      "grad_norm": 0.38176881311264943,
      "kl": 0.36710357666015625,
      "learning_rate": 3.6765432098765433e-07,
      "loss": 0.0004,
      "reward": 1.7035714983940125,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7214285992085934,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2978
    },
    {
      "completion_length": 260.9598340988159,
      "epoch": 0.4996018274026573,
      "grad_norm": 0.14641630559689583,
      "kl": 0.6771926879882812,
      "learning_rate": 3.679012345679012e-07,
      "loss": 0.0007,
      "reward": 1.74642863124609,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.755357176065445,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2980
    },
    {
      "completion_length": 258.0625114440918,
      "epoch": 0.4999371306425248,
      "grad_norm": 0.2595195319935882,
      "kl": 0.0545196533203125,
      "learning_rate": 3.6814814814814816e-07,
      "loss": 0.0001,
      "reward": 1.7357143685221672,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7446428909897804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2982
    },
    {
      "completion_length": 263.83929920196533,
      "epoch": 0.5002724338823924,
      "grad_norm": 0.24109920616356117,
      "kl": 0.05385589599609375,
      "learning_rate": 3.68395061728395e-07,
      "loss": 0.0001,
      "reward": 1.7982143461704254,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.8116071783006191,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2984
    },
    {
      "completion_length": 252.6160831451416,
      "epoch": 0.5006077371222599,
      "grad_norm": 0.31178103313116434,
      "kl": 0.11287689208984375,
      "learning_rate": 3.6864197530864193e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.8107143193483353,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2986
    },
    {
      "completion_length": 258.7276906967163,
      "epoch": 0.5009430403621274,
      "grad_norm": 0.257026450835211,
      "kl": 0.25646209716796875,
      "learning_rate": 3.688888888888889e-07,
      "loss": 0.0003,
      "reward": 1.7714286297559738,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7803571596741676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 2988
    },
    {
      "completion_length": 251.8125123977661,
      "epoch": 0.5012783436019951,
      "grad_norm": 0.26271403970330076,
      "kl": 0.6702995300292969,
      "learning_rate": 3.6913580246913576e-07,
      "loss": 0.0007,
      "reward": 1.8125000521540642,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.8169642947614193,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2990
    },
    {
      "completion_length": 256.6384038925171,
      "epoch": 0.5016136468418626,
      "grad_norm": 0.27592010057674143,
      "kl": 0.1610565185546875,
      "learning_rate": 3.693827160493827e-07,
      "loss": 0.0002,
      "reward": 1.6785715073347092,
      "reward_std": 0.08081220369786024,
      "rewards/equation_reward_func": 0.6964286081492901,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 2992
    },
    {
      "completion_length": 257.120548248291,
      "epoch": 0.5019489500817301,
      "grad_norm": 0.5610497082441622,
      "kl": 0.25261688232421875,
      "learning_rate": 3.6962962962962965e-07,
      "loss": 0.0003,
      "reward": 1.680357240140438,
      "reward_std": 0.10859139636158943,
      "rewards/equation_reward_func": 0.7026785984635353,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 2994
    },
    {
      "completion_length": 268.7410840988159,
      "epoch": 0.5022842533215978,
      "grad_norm": 0.5487386942065536,
      "kl": 0.42702484130859375,
      "learning_rate": 3.6987654320987653e-07,
      "loss": 0.0004,
      "reward": 1.7446429282426834,
      "reward_std": 0.08838834799826145,
      "rewards/equation_reward_func": 0.7580357454717159,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 2996
    },
    {
      "completion_length": 250.97322463989258,
      "epoch": 0.5026195565614653,
      "grad_norm": 0.3198474395173545,
      "kl": 0.5860786437988281,
      "learning_rate": 3.701234567901235e-07,
      "loss": 0.0006,
      "reward": 1.7339286506175995,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7383928876370192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 2998
    },
    {
      "completion_length": 255.59822463989258,
      "epoch": 0.5029548598013328,
      "grad_norm": 0.26785004048529404,
      "kl": 1.1002006530761719,
      "learning_rate": 3.703703703703703e-07,
      "loss": 0.0011,
      "reward": 1.748214341700077,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7616071663796902,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3000
    },
    {
      "completion_length": 261.35269260406494,
      "epoch": 0.5032901630412003,
      "grad_norm": 0.3310002273693556,
      "kl": 0.20697784423828125,
      "learning_rate": 3.7061728395061725e-07,
      "loss": 0.0002,
      "reward": 1.6982143595814705,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7116071823984385,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3002
    },
    {
      "completion_length": 253.9732255935669,
      "epoch": 0.503625466281068,
      "grad_norm": 0.13514736768128646,
      "kl": 0.20084381103515625,
      "learning_rate": 3.708641975308642e-07,
      "loss": 0.0002,
      "reward": 1.7875000461935997,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.800892872735858,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3004
    },
    {
      "completion_length": 259.071439743042,
      "epoch": 0.5039607695209355,
      "grad_norm": 0.3078015700685397,
      "kl": 0.37459564208984375,
      "learning_rate": 3.711111111111111e-07,
      "loss": 0.0004,
      "reward": 1.7839286252856255,
      "reward_std": 0.06313453428447247,
      "rewards/equation_reward_func": 0.7973214499652386,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3006
    },
    {
      "completion_length": 263.93751335144043,
      "epoch": 0.504296072760803,
      "grad_norm": 0.36645424767648727,
      "kl": 0.47135162353515625,
      "learning_rate": 3.71358024691358e-07,
      "loss": 0.0005,
      "reward": 1.7571429163217545,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3008
    },
    {
      "completion_length": 255.758939743042,
      "epoch": 0.5046313760006707,
      "grad_norm": 0.7864030190498497,
      "kl": 1.5985565185546875,
      "learning_rate": 3.7160493827160496e-07,
      "loss": 0.0016,
      "reward": 1.7482143342494965,
      "reward_std": 0.07828682288527489,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9776785783469677,
      "step": 3010
    },
    {
      "completion_length": 248.2634038925171,
      "epoch": 0.5049666792405382,
      "grad_norm": 0.2667122406171907,
      "kl": 0.03884124755859375,
      "learning_rate": 3.7185185185185185e-07,
      "loss": 0.0,
      "reward": 1.7642857804894447,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7732143178582191,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3012
    },
    {
      "completion_length": 267.8482303619385,
      "epoch": 0.5053019824804057,
      "grad_norm": 0.3173331245909394,
      "kl": 1.0220565795898438,
      "learning_rate": 3.7209876543209874e-07,
      "loss": 0.001,
      "reward": 1.7321429252624512,
      "reward_std": 0.09596449043601751,
      "rewards/equation_reward_func": 0.7589285932481289,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 3014
    },
    {
      "completion_length": 243.81697368621826,
      "epoch": 0.5056372857202732,
      "grad_norm": 0.34077550562632575,
      "kl": 0.040744781494140625,
      "learning_rate": 3.7234567901234563e-07,
      "loss": 0.0,
      "reward": 1.7589286267757416,
      "reward_std": 0.09848987217992544,
      "rewards/equation_reward_func": 0.7812500298023224,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 3016
    },
    {
      "completion_length": 261.33929347991943,
      "epoch": 0.5059725889601409,
      "grad_norm": 0.15313021699348667,
      "kl": 0.43424224853515625,
      "learning_rate": 3.7259259259259257e-07,
      "loss": 0.0004,
      "reward": 1.716071479022503,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7294643260538578,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3018
    },
    {
      "completion_length": 258.4509029388428,
      "epoch": 0.5063078922000084,
      "grad_norm": 0.5652791164362287,
      "kl": 0.9497528076171875,
      "learning_rate": 3.728395061728395e-07,
      "loss": 0.0009,
      "reward": 1.6750000938773155,
      "reward_std": 0.11616754159331322,
      "rewards/equation_reward_func": 0.6928571835160255,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3020
    },
    {
      "completion_length": 267.790189743042,
      "epoch": 0.5066431954398759,
      "grad_norm": 0.2861186061735038,
      "kl": 0.532745361328125,
      "learning_rate": 3.730864197530864e-07,
      "loss": 0.0005,
      "reward": 1.7785714715719223,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7875000201165676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3022
    },
    {
      "completion_length": 266.2500162124634,
      "epoch": 0.5069784986797435,
      "grad_norm": 0.4827885536321817,
      "kl": 0.32605743408203125,
      "learning_rate": 3.7333333333333334e-07,
      "loss": 0.0003,
      "reward": 1.6928572058677673,
      "reward_std": 0.1010152529925108,
      "rewards/equation_reward_func": 0.7107143234461546,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3024
    },
    {
      "completion_length": 256.75447368621826,
      "epoch": 0.5073138019196111,
      "grad_norm": 0.29762340493815626,
      "kl": 0.08817291259765625,
      "learning_rate": 3.7358024691358023e-07,
      "loss": 0.0001,
      "reward": 1.6732143610715866,
      "reward_std": 0.08838834706693888,
      "rewards/equation_reward_func": 0.6955357305705547,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 3026
    },
    {
      "completion_length": 250.3794765472412,
      "epoch": 0.5076491051594786,
      "grad_norm": 0.29653367322563745,
      "kl": 0.04032135009765625,
      "learning_rate": 3.738271604938271e-07,
      "loss": 0.0,
      "reward": 1.717857226729393,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7267857417464256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3028
    },
    {
      "completion_length": 256.73661708831787,
      "epoch": 0.5079844083993461,
      "grad_norm": 0.19002438532993526,
      "kl": 0.0325164794921875,
      "learning_rate": 3.7407407407407406e-07,
      "loss": 0.0,
      "reward": 1.7125000953674316,
      "reward_std": 0.0328299580141902,
      "rewards/equation_reward_func": 0.716964315623045,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3030
    },
    {
      "completion_length": 264.87947940826416,
      "epoch": 0.5083197116392137,
      "grad_norm": 0.47697001771334524,
      "kl": 0.1249542236328125,
      "learning_rate": 3.7432098765432095e-07,
      "loss": 0.0001,
      "reward": 1.739285796880722,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.748214315623045,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3032
    },
    {
      "completion_length": 252.37501335144043,
      "epoch": 0.5086550148790813,
      "grad_norm": 0.19164757321243217,
      "kl": 0.16481781005859375,
      "learning_rate": 3.745679012345679e-07,
      "loss": 0.0002,
      "reward": 1.6839286610484123,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.6973214689642191,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3034
    },
    {
      "completion_length": 242.11608409881592,
      "epoch": 0.5089903181189488,
      "grad_norm": 0.2965385453786233,
      "kl": 0.102325439453125,
      "learning_rate": 3.7481481481481483e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.769642885774374,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3036
    },
    {
      "completion_length": 264.0134038925171,
      "epoch": 0.5093256213588164,
      "grad_norm": 0.26828654844694805,
      "kl": 0.50030517578125,
      "learning_rate": 3.750617283950617e-07,
      "loss": 0.0005,
      "reward": 1.7232143431901932,
      "reward_std": 0.0833375845104456,
      "rewards/equation_reward_func": 0.7633928880095482,
      "rewards/format_reward_func": 0.9598214477300644,
      "step": 3038
    },
    {
      "completion_length": 254.5625114440918,
      "epoch": 0.509660924598684,
      "grad_norm": 0.193990185560616,
      "kl": 0.0839996337890625,
      "learning_rate": 3.7530864197530866e-07,
      "loss": 0.0001,
      "reward": 1.739285759627819,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7482143323868513,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3040
    },
    {
      "completion_length": 246.9419755935669,
      "epoch": 0.5099962278385515,
      "grad_norm": 0.19313521551747367,
      "kl": 0.03554534912109375,
      "learning_rate": 3.755555555555555e-07,
      "loss": 0.0,
      "reward": 1.7839286252856255,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7883928753435612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3042
    },
    {
      "completion_length": 255.01340675354004,
      "epoch": 0.510331531078419,
      "grad_norm": 0.7765180616026391,
      "kl": 0.51654052734375,
      "learning_rate": 3.7580246913580244e-07,
      "loss": 0.0005,
      "reward": 1.7482143342494965,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.7705357372760773,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 3044
    },
    {
      "completion_length": 256.6339416503906,
      "epoch": 0.5106668343182866,
      "grad_norm": 1.1364308614205811,
      "kl": 0.6205825805664062,
      "learning_rate": 3.760493827160494e-07,
      "loss": 0.0006,
      "reward": 1.7375000715255737,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7508928962051868,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3046
    },
    {
      "completion_length": 246.1830472946167,
      "epoch": 0.5110021375581542,
      "grad_norm": 0.2493496018800458,
      "kl": 0.158294677734375,
      "learning_rate": 3.7629629629629627e-07,
      "loss": 0.0002,
      "reward": 1.7750000432133675,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7839285992085934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3048
    },
    {
      "completion_length": 261.41072940826416,
      "epoch": 0.5113374407980217,
      "grad_norm": 0.19985846821705897,
      "kl": 0.2151336669921875,
      "learning_rate": 3.765432098765432e-07,
      "loss": 0.0002,
      "reward": 1.7446428835391998,
      "reward_std": 0.08838834706693888,
      "rewards/equation_reward_func": 0.766964316368103,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 3050
    },
    {
      "completion_length": 257.71429920196533,
      "epoch": 0.5116727440378893,
      "grad_norm": 0.22633980489201774,
      "kl": 0.155548095703125,
      "learning_rate": 3.767901234567901e-07,
      "loss": 0.0002,
      "reward": 1.694642961025238,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7071428969502449,
      "rewards/format_reward_func": 0.9875000044703484,
      "step": 3052
    },
    {
      "completion_length": 259.6160840988159,
      "epoch": 0.5120080472777568,
      "grad_norm": 0.13005143693972465,
      "kl": 0.12020111083984375,
      "learning_rate": 3.7703703703703704e-07,
      "loss": 0.0001,
      "reward": 1.7982143387198448,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8026785999536514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3054
    },
    {
      "completion_length": 252.43751049041748,
      "epoch": 0.5123433505176244,
      "grad_norm": 0.29216567176643266,
      "kl": 0.304046630859375,
      "learning_rate": 3.772839506172839e-07,
      "loss": 0.0003,
      "reward": 1.7129464820027351,
      "reward_std": 0.08270623860880733,
      "rewards/equation_reward_func": 0.7232143189758062,
      "rewards/format_reward_func": 0.9897321499884129,
      "step": 3056
    },
    {
      "completion_length": 251.09376525878906,
      "epoch": 0.5126786537574919,
      "grad_norm": 0.6033130029832788,
      "kl": 0.08933258056640625,
      "learning_rate": 3.775308641975308e-07,
      "loss": 0.0001,
      "reward": 1.7767857685685158,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.7901786081492901,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3058
    },
    {
      "completion_length": 252.43751335144043,
      "epoch": 0.5130139569973595,
      "grad_norm": 0.20902483387252335,
      "kl": 0.0396728515625,
      "learning_rate": 3.7777777777777775e-07,
      "loss": 0.0,
      "reward": 1.7375000715255737,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643178582191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3060
    },
    {
      "completion_length": 249.22769165039062,
      "epoch": 0.513349260237227,
      "grad_norm": 0.260245315584681,
      "kl": 0.18543243408203125,
      "learning_rate": 3.780246913580247e-07,
      "loss": 0.0002,
      "reward": 1.7446429207921028,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7491071820259094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3062
    },
    {
      "completion_length": 254.43751049041748,
      "epoch": 0.5136845634770946,
      "grad_norm": 0.26409884638490183,
      "kl": 0.037139892578125,
      "learning_rate": 3.782716049382716e-07,
      "loss": 0.0,
      "reward": 1.696428656578064,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.6964286081492901,
      "rewards/format_reward_func": 1.0,
      "step": 3064
    },
    {
      "completion_length": 260.54465103149414,
      "epoch": 0.5140198667169622,
      "grad_norm": 0.1428687655533802,
      "kl": 0.0744171142578125,
      "learning_rate": 3.785185185185185e-07,
      "loss": 0.0001,
      "reward": 1.7482143640518188,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3066
    },
    {
      "completion_length": 246.53572368621826,
      "epoch": 0.5143551699568297,
      "grad_norm": 0.16185149934506265,
      "kl": 0.0672454833984375,
      "learning_rate": 3.787654320987654e-07,
      "loss": 0.0001,
      "reward": 1.7589286491274834,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7633928917348385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3068
    },
    {
      "completion_length": 258.9821557998657,
      "epoch": 0.5146904731966973,
      "grad_norm": 0.2306501331554964,
      "kl": 0.0682220458984375,
      "learning_rate": 3.790123456790123e-07,
      "loss": 0.0001,
      "reward": 1.7035715207457542,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7125000543892384,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3070
    },
    {
      "completion_length": 264.87947845458984,
      "epoch": 0.5150257764365648,
      "grad_norm": 0.22456347881762204,
      "kl": 0.06907272338867188,
      "learning_rate": 3.7925925925925924e-07,
      "loss": 0.0001,
      "reward": 1.6892858147621155,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6892857514321804,
      "rewards/format_reward_func": 1.0,
      "step": 3072
    },
    {
      "completion_length": 258.65179538726807,
      "epoch": 0.5153610796764324,
      "grad_norm": 0.35293272384908336,
      "kl": 0.2332611083984375,
      "learning_rate": 3.7950617283950613e-07,
      "loss": 0.0002,
      "reward": 1.739285796880722,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7482143044471741,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3074
    },
    {
      "completion_length": 251.87054443359375,
      "epoch": 0.5156963829162999,
      "grad_norm": 0.1800318548596281,
      "kl": 0.1234283447265625,
      "learning_rate": 3.7975308641975307e-07,
      "loss": 0.0001,
      "reward": 1.7839286252856255,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7883928902447224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3076
    },
    {
      "completion_length": 246.4553689956665,
      "epoch": 0.5160316861561675,
      "grad_norm": 0.1740786371284468,
      "kl": 0.02825164794921875,
      "learning_rate": 3.7999999999999996e-07,
      "loss": 0.0,
      "reward": 1.7267857939004898,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7312500402331352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3078
    },
    {
      "completion_length": 258.22769260406494,
      "epoch": 0.5163669893960351,
      "grad_norm": 0.3160476219985606,
      "kl": 0.057590484619140625,
      "learning_rate": 3.802469135802469e-07,
      "loss": 0.0001,
      "reward": 1.7589286342263222,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7633928805589676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3080
    },
    {
      "completion_length": 252.28572750091553,
      "epoch": 0.5167022926359026,
      "grad_norm": 0.4762047675136293,
      "kl": 0.20147705078125,
      "learning_rate": 3.8049382716049384e-07,
      "loss": 0.0002,
      "reward": 1.751785770058632,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.7651785984635353,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3082
    },
    {
      "completion_length": 257.0401906967163,
      "epoch": 0.5170375958757701,
      "grad_norm": 0.4295341728220505,
      "kl": 0.029018402099609375,
      "learning_rate": 3.8074074074074073e-07,
      "loss": 0.0,
      "reward": 1.7089286372065544,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7133928965777159,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3084
    },
    {
      "completion_length": 258.2321529388428,
      "epoch": 0.5173728991156377,
      "grad_norm": 0.2085002371188389,
      "kl": 0.2578277587890625,
      "learning_rate": 3.809876543209876e-07,
      "loss": 0.0003,
      "reward": 1.7178572043776512,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7178571913391352,
      "rewards/format_reward_func": 1.0,
      "step": 3086
    },
    {
      "completion_length": 263.6384038925171,
      "epoch": 0.5177082023555053,
      "grad_norm": 0.564076640872581,
      "kl": 0.08950042724609375,
      "learning_rate": 3.8123456790123456e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 3088
    },
    {
      "completion_length": 252.54465675354004,
      "epoch": 0.5180435055953728,
      "grad_norm": 0.18742434123139506,
      "kl": 0.1312103271484375,
      "learning_rate": 3.8148148148148145e-07,
      "loss": 0.0001,
      "reward": 1.751785784959793,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7562500163912773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3090
    },
    {
      "completion_length": 250.01786708831787,
      "epoch": 0.5183788088352403,
      "grad_norm": 0.20853683279872454,
      "kl": 0.11835479736328125,
      "learning_rate": 3.817283950617284e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285835623741,
      "rewards/format_reward_func": 1.0,
      "step": 3092
    },
    {
      "completion_length": 261.95983695983887,
      "epoch": 0.518714112075108,
      "grad_norm": 0.20069159944437373,
      "kl": 0.052577972412109375,
      "learning_rate": 3.819753086419753e-07,
      "loss": 0.0001,
      "reward": 1.7750000432133675,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.783928606659174,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3094
    },
    {
      "completion_length": 248.60268878936768,
      "epoch": 0.5190494153149755,
      "grad_norm": 0.18308556647430432,
      "kl": 0.05411529541015625,
      "learning_rate": 3.822222222222222e-07,
      "loss": 0.0001,
      "reward": 1.7196429520845413,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7241071797907352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3096
    },
    {
      "completion_length": 252.9107255935669,
      "epoch": 0.519384718554843,
      "grad_norm": 0.23069855632685662,
      "kl": 0.0457916259765625,
      "learning_rate": 3.8246913580246916e-07,
      "loss": 0.0,
      "reward": 1.744642935693264,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7491071671247482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3098
    },
    {
      "completion_length": 250.46429920196533,
      "epoch": 0.5197200217947106,
      "grad_norm": 0.259615918588719,
      "kl": 0.05632781982421875,
      "learning_rate": 3.82716049382716e-07,
      "loss": 0.0001,
      "reward": 1.7589286342263222,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7633929066359997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3100
    },
    {
      "completion_length": 255.2142972946167,
      "epoch": 0.5200553250345782,
      "grad_norm": 0.2773907131668989,
      "kl": 0.36530303955078125,
      "learning_rate": 3.8296296296296294e-07,
      "loss": 0.0004,
      "reward": 1.758928656578064,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928693830967,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3102
    },
    {
      "completion_length": 248.92412090301514,
      "epoch": 0.5203906282744457,
      "grad_norm": 0.37992031338955395,
      "kl": 0.17604446411132812,
      "learning_rate": 3.8320987654320983e-07,
      "loss": 0.0002,
      "reward": 1.7678572162985802,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7767857499420643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3104
    },
    {
      "completion_length": 258.3125114440918,
      "epoch": 0.5207259315143132,
      "grad_norm": 0.394132585956962,
      "kl": 0.433685302734375,
      "learning_rate": 3.8345679012345677e-07,
      "loss": 0.0004,
      "reward": 1.6821429505944252,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.6910714581608772,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3106
    },
    {
      "completion_length": 264.12501430511475,
      "epoch": 0.5210612347541808,
      "grad_norm": 0.18801212145262286,
      "kl": 0.64044189453125,
      "learning_rate": 3.837037037037037e-07,
      "loss": 0.0006,
      "reward": 1.7093750983476639,
      "reward_std": 0.037249374436214566,
      "rewards/equation_reward_func": 0.7107143178582191,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 3108
    },
    {
      "completion_length": 251.20983409881592,
      "epoch": 0.5213965379940484,
      "grad_norm": 0.25513628192429766,
      "kl": 0.08748626708984375,
      "learning_rate": 3.839506172839506e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428906172514,
      "rewards/format_reward_func": 1.0,
      "step": 3110
    },
    {
      "completion_length": 255.2857265472412,
      "epoch": 0.5217318412339159,
      "grad_norm": 0.28046900188641055,
      "kl": 0.3914947509765625,
      "learning_rate": 3.8419753086419754e-07,
      "loss": 0.0004,
      "reward": 1.7464286386966705,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 3112
    },
    {
      "completion_length": 256.214298248291,
      "epoch": 0.5220671444737834,
      "grad_norm": 0.22531422784150337,
      "kl": 0.24646377563476562,
      "learning_rate": 3.8444444444444443e-07,
      "loss": 0.0002,
      "reward": 1.796428605914116,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286170899868,
      "rewards/format_reward_func": 1.0,
      "step": 3114
    },
    {
      "completion_length": 257.6250104904175,
      "epoch": 0.5224024477136511,
      "grad_norm": 0.2730117358782168,
      "kl": 0.4022064208984375,
      "learning_rate": 3.846913580246913e-07,
      "loss": 0.0004,
      "reward": 1.7392857745289803,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.7482143193483353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3116
    },
    {
      "completion_length": 250.9687623977661,
      "epoch": 0.5227377509535186,
      "grad_norm": 0.6576136637774147,
      "kl": 0.7910614013671875,
      "learning_rate": 3.8493827160493826e-07,
      "loss": 0.0008,
      "reward": 1.7821428924798965,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 3118
    },
    {
      "completion_length": 260.790189743042,
      "epoch": 0.5230730541933861,
      "grad_norm": 0.24315733472477172,
      "kl": 0.04315185546875,
      "learning_rate": 3.8518518518518515e-07,
      "loss": 0.0,
      "reward": 1.7857143431901932,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7946428805589676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3120
    },
    {
      "completion_length": 264.5000114440918,
      "epoch": 0.5234083574332536,
      "grad_norm": 0.22778397283566784,
      "kl": 0.7026290893554688,
      "learning_rate": 3.854320987654321e-07,
      "loss": 0.0007,
      "reward": 1.823214329779148,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.8276785910129547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3122
    },
    {
      "completion_length": 262.43751335144043,
      "epoch": 0.5237436606731213,
      "grad_norm": 0.21985933881844663,
      "kl": 0.5026016235351562,
      "learning_rate": 3.8567901234567903e-07,
      "loss": 0.0005,
      "reward": 1.742857202887535,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 3124
    },
    {
      "completion_length": 249.99108219146729,
      "epoch": 0.5240789639129888,
      "grad_norm": 0.33885053896790346,
      "kl": 0.574188232421875,
      "learning_rate": 3.859259259259259e-07,
      "loss": 0.0006,
      "reward": 1.7892857789993286,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857566475868,
      "rewards/format_reward_func": 1.0,
      "step": 3126
    },
    {
      "completion_length": 257.49108505249023,
      "epoch": 0.5244142671528563,
      "grad_norm": 0.2792327924735609,
      "kl": 0.32171630859375,
      "learning_rate": 3.861728395061728e-07,
      "loss": 0.0003,
      "reward": 1.750000074505806,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 3128
    },
    {
      "completion_length": 263.89733505249023,
      "epoch": 0.524749570392724,
      "grad_norm": 0.5251332741259391,
      "kl": 0.3920097351074219,
      "learning_rate": 3.864197530864197e-07,
      "loss": 0.0004,
      "reward": 1.7375000715255737,
      "reward_std": 0.08838834706693888,
      "rewards/equation_reward_func": 0.750892898067832,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3130
    },
    {
      "completion_length": 258.620548248291,
      "epoch": 0.5250848736325915,
      "grad_norm": 0.17825779621215138,
      "kl": 0.44925689697265625,
      "learning_rate": 3.8666666666666664e-07,
      "loss": 0.0004,
      "reward": 1.7982143387198448,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.8116071708500385,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3132
    },
    {
      "completion_length": 259.758939743042,
      "epoch": 0.525420176872459,
      "grad_norm": 0.22491108314728606,
      "kl": 0.089263916015625,
      "learning_rate": 3.869135802469136e-07,
      "loss": 0.0001,
      "reward": 1.7517857924103737,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500238418579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3134
    },
    {
      "completion_length": 265.76787090301514,
      "epoch": 0.5257554801123265,
      "grad_norm": 0.25530360444858247,
      "kl": 0.16037750244140625,
      "learning_rate": 3.8716049382716046e-07,
      "loss": 0.0002,
      "reward": 1.7232143506407738,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7276786155998707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3136
    },
    {
      "completion_length": 266.25894260406494,
      "epoch": 0.5260907833521942,
      "grad_norm": 0.21909712127541298,
      "kl": 0.12729644775390625,
      "learning_rate": 3.874074074074074e-07,
      "loss": 0.0001,
      "reward": 1.7250000983476639,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7250000238418579,
      "rewards/format_reward_func": 1.0,
      "step": 3138
    },
    {
      "completion_length": 254.43304538726807,
      "epoch": 0.5264260865920617,
      "grad_norm": 0.36150859827321014,
      "kl": 0.08214187622070312,
      "learning_rate": 3.8765432098765435e-07,
      "loss": 0.0001,
      "reward": 1.7232143506407738,
      "reward_std": 0.07828682195395231,
      "rewards/equation_reward_func": 0.7366071734577417,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3140
    },
    {
      "completion_length": 264.1160831451416,
      "epoch": 0.5267613898319292,
      "grad_norm": 0.2555141927292338,
      "kl": 0.055820465087890625,
      "learning_rate": 3.879012345679012e-07,
      "loss": 0.0001,
      "reward": 1.698214367032051,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7116071879863739,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3142
    },
    {
      "completion_length": 258.9196548461914,
      "epoch": 0.5270966930717969,
      "grad_norm": 0.1265249629832738,
      "kl": 0.09015274047851562,
      "learning_rate": 3.881481481481481e-07,
      "loss": 0.0001,
      "reward": 1.8392857387661934,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8392857313156128,
      "rewards/format_reward_func": 1.0,
      "step": 3144
    },
    {
      "completion_length": 262.70983600616455,
      "epoch": 0.5274319963116644,
      "grad_norm": 0.2757642721961975,
      "kl": 0.04498291015625,
      "learning_rate": 3.88395061728395e-07,
      "loss": 0.0,
      "reward": 1.7678572237491608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 3146
    },
    {
      "completion_length": 265.4866199493408,
      "epoch": 0.5277672995515319,
      "grad_norm": 0.21547518845810912,
      "kl": 0.05883026123046875,
      "learning_rate": 3.8864197530864195e-07,
      "loss": 0.0001,
      "reward": 1.6821429505944252,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.6821428947150707,
      "rewards/format_reward_func": 1.0,
      "step": 3148
    },
    {
      "completion_length": 262.4687623977661,
      "epoch": 0.5281026027913994,
      "grad_norm": 0.11776663498490876,
      "kl": 0.15918731689453125,
      "learning_rate": 3.888888888888889e-07,
      "loss": 0.0002,
      "reward": 1.8321428894996643,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8321428801864386,
      "rewards/format_reward_func": 1.0,
      "step": 3150
    },
    {
      "completion_length": 259.6116247177124,
      "epoch": 0.5284379060312671,
      "grad_norm": 0.35745340475596804,
      "kl": 0.0491790771484375,
      "learning_rate": 3.891358024691358e-07,
      "loss": 0.0,
      "reward": 1.7142857983708382,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7321428880095482,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3152
    },
    {
      "completion_length": 262.5401906967163,
      "epoch": 0.5287732092711346,
      "grad_norm": 0.21916166918525395,
      "kl": 0.047977447509765625,
      "learning_rate": 3.893827160493827e-07,
      "loss": 0.0,
      "reward": 1.7660714760422707,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7705357521772385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3154
    },
    {
      "completion_length": 264.5000123977661,
      "epoch": 0.5291085125110021,
      "grad_norm": 0.30194854312892355,
      "kl": 0.06650543212890625,
      "learning_rate": 3.8962962962962956e-07,
      "loss": 0.0001,
      "reward": 1.7303572297096252,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7348214648663998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3156
    },
    {
      "completion_length": 260.9910821914673,
      "epoch": 0.5294438157508697,
      "grad_norm": 0.08596848806407578,
      "kl": 0.05260467529296875,
      "learning_rate": 3.898765432098765e-07,
      "loss": 0.0001,
      "reward": 1.8125000670552254,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8169643096625805,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3158
    },
    {
      "completion_length": 259.5000104904175,
      "epoch": 0.5297791189907373,
      "grad_norm": 0.16451427183843081,
      "kl": 0.037384033203125,
      "learning_rate": 3.9012345679012344e-07,
      "loss": 0.0,
      "reward": 1.791071467101574,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7955357432365417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3160
    },
    {
      "completion_length": 267.2187614440918,
      "epoch": 0.5301144222306048,
      "grad_norm": 0.2709270291715736,
      "kl": 0.031284332275390625,
      "learning_rate": 3.9037037037037033e-07,
      "loss": 0.0,
      "reward": 1.8000000566244125,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 3162
    },
    {
      "completion_length": 271.5491180419922,
      "epoch": 0.5304497254704723,
      "grad_norm": 0.3913483799722536,
      "kl": 0.08670806884765625,
      "learning_rate": 3.9061728395061727e-07,
      "loss": 0.0001,
      "reward": 1.7875000536441803,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7919643223285675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3164
    },
    {
      "completion_length": 257.41965675354004,
      "epoch": 0.53078502871034,
      "grad_norm": 0.17074821649440147,
      "kl": 0.03655242919921875,
      "learning_rate": 3.908641975308642e-07,
      "loss": 0.0,
      "reward": 1.7785714864730835,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 3166
    },
    {
      "completion_length": 268.9821538925171,
      "epoch": 0.5311203319502075,
      "grad_norm": 0.19454170410731372,
      "kl": 0.0432891845703125,
      "learning_rate": 3.911111111111111e-07,
      "loss": 0.0,
      "reward": 1.6910715103149414,
      "reward_std": 0.0732360603287816,
      "rewards/equation_reward_func": 0.7044643256813288,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3168
    },
    {
      "completion_length": 259.4464406967163,
      "epoch": 0.531455635190075,
      "grad_norm": 0.2802742714672685,
      "kl": 0.03183746337890625,
      "learning_rate": 3.91358024691358e-07,
      "loss": 0.0,
      "reward": 1.7357143685221672,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143312692642,
      "rewards/format_reward_func": 1.0,
      "step": 3170
    },
    {
      "completion_length": 255.79465579986572,
      "epoch": 0.5317909384299426,
      "grad_norm": 0.24431833873124947,
      "kl": 0.07769012451171875,
      "learning_rate": 3.916049382716049e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7821428943425417,
      "rewards/format_reward_func": 1.0,
      "step": 3172
    },
    {
      "completion_length": 261.43305015563965,
      "epoch": 0.5321262416698102,
      "grad_norm": 0.2838905343856962,
      "kl": 0.0420074462890625,
      "learning_rate": 3.918518518518518e-07,
      "loss": 0.0,
      "reward": 1.7053572162985802,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7098214700818062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3174
    },
    {
      "completion_length": 261.9464406967163,
      "epoch": 0.5324615449096777,
      "grad_norm": 0.383031704099419,
      "kl": 0.052783966064453125,
      "learning_rate": 3.9209876543209876e-07,
      "loss": 0.0001,
      "reward": 1.7660715132951736,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3176
    },
    {
      "completion_length": 253.23661994934082,
      "epoch": 0.5327968481495452,
      "grad_norm": 0.3703123406378154,
      "kl": 0.0786895751953125,
      "learning_rate": 3.9234567901234565e-07,
      "loss": 0.0001,
      "reward": 1.773214340209961,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7776786014437675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3178
    },
    {
      "completion_length": 267.46876525878906,
      "epoch": 0.5331321513894128,
      "grad_norm": 0.2207114618581663,
      "kl": 0.0382232666015625,
      "learning_rate": 3.925925925925926e-07,
      "loss": 0.0,
      "reward": 1.725000061094761,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7339286021888256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3180
    },
    {
      "completion_length": 260.8259086608887,
      "epoch": 0.5334674546292804,
      "grad_norm": 0.3279488891762705,
      "kl": 0.03820037841796875,
      "learning_rate": 3.9283950617283953e-07,
      "loss": 0.0,
      "reward": 1.7464286386966705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 3182
    },
    {
      "completion_length": 253.90625858306885,
      "epoch": 0.5338027578691479,
      "grad_norm": 0.2417637339334101,
      "kl": 0.031524658203125,
      "learning_rate": 3.930864197530864e-07,
      "loss": 0.0,
      "reward": 1.7375000789761543,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7419643253087997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3184
    },
    {
      "completion_length": 255.61608600616455,
      "epoch": 0.5341380611090155,
      "grad_norm": 0.20489918160535506,
      "kl": 0.056060791015625,
      "learning_rate": 3.933333333333333e-07,
      "loss": 0.0001,
      "reward": 1.7875000461935997,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7919643074274063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3186
    },
    {
      "completion_length": 257.433048248291,
      "epoch": 0.534473364348883,
      "grad_norm": 0.25187838214239483,
      "kl": 0.026729583740234375,
      "learning_rate": 3.935802469135802e-07,
      "loss": 0.0,
      "reward": 1.7714286595582962,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7803571820259094,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3188
    },
    {
      "completion_length": 269.12501335144043,
      "epoch": 0.5348086675887506,
      "grad_norm": 0.28126525645205525,
      "kl": 0.05770111083984375,
      "learning_rate": 3.9382716049382714e-07,
      "loss": 0.0001,
      "reward": 1.6446429342031479,
      "reward_std": 0.10859139915555716,
      "rewards/equation_reward_func": 0.666964327916503,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 3190
    },
    {
      "completion_length": 264.58929920196533,
      "epoch": 0.5351439708286181,
      "grad_norm": 0.21792582285765727,
      "kl": 0.02896881103515625,
      "learning_rate": 3.940740740740741e-07,
      "loss": 0.0,
      "reward": 1.7214286401867867,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7214286141097546,
      "rewards/format_reward_func": 1.0,
      "step": 3192
    },
    {
      "completion_length": 261.4419775009155,
      "epoch": 0.5354792740684857,
      "grad_norm": 0.26582125873677714,
      "kl": 0.1292572021484375,
      "learning_rate": 3.9432098765432097e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7607143297791481,
      "rewards/format_reward_func": 1.0,
      "step": 3194
    },
    {
      "completion_length": 254.1830472946167,
      "epoch": 0.5358145773083532,
      "grad_norm": 0.12337232155198748,
      "kl": 0.09566116333007812,
      "learning_rate": 3.945679012345679e-07,
      "loss": 0.0001,
      "reward": 1.7767857611179352,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7812500167638063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3196
    },
    {
      "completion_length": 246.31250858306885,
      "epoch": 0.5361498805482208,
      "grad_norm": 0.13809997347069525,
      "kl": 0.02978515625,
      "learning_rate": 3.948148148148148e-07,
      "loss": 0.0,
      "reward": 1.7107143551111221,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7107143253087997,
      "rewards/format_reward_func": 1.0,
      "step": 3198
    },
    {
      "completion_length": 252.37055015563965,
      "epoch": 0.5364851837880884,
      "grad_norm": 0.2090765586833693,
      "kl": 0.037384033203125,
      "learning_rate": 3.950617283950617e-07,
      "loss": 0.0,
      "reward": 1.7642858028411865,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 3200
    },
    {
      "completion_length": 256.308048248291,
      "epoch": 0.5368204870279559,
      "grad_norm": 0.2867788682863003,
      "kl": 0.027191162109375,
      "learning_rate": 3.9530864197530863e-07,
      "loss": 0.0,
      "reward": 1.7428572326898575,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571619093418,
      "rewards/format_reward_func": 1.0,
      "step": 3202
    },
    {
      "completion_length": 242.46876049041748,
      "epoch": 0.5371557902678235,
      "grad_norm": 0.19248161610849127,
      "kl": 0.0301055908203125,
      "learning_rate": 3.955555555555555e-07,
      "loss": 0.0,
      "reward": 1.7857143431901932,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 3204
    },
    {
      "completion_length": 253.46876049041748,
      "epoch": 0.537491093507691,
      "grad_norm": 0.3401432023399993,
      "kl": 0.047245025634765625,
      "learning_rate": 3.9580246913580246e-07,
      "loss": 0.0,
      "reward": 1.7553572207689285,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7598214633762836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3206
    },
    {
      "completion_length": 251.1205472946167,
      "epoch": 0.5378263967475586,
      "grad_norm": 0.3760622304921827,
      "kl": 0.032135009765625,
      "learning_rate": 3.960493827160494e-07,
      "loss": 0.0,
      "reward": 1.7767857611179352,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.7812500353902578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3208
    },
    {
      "completion_length": 252.9330472946167,
      "epoch": 0.5381616999874261,
      "grad_norm": 0.3604519919978038,
      "kl": 0.10263824462890625,
      "learning_rate": 3.962962962962963e-07,
      "loss": 0.0001,
      "reward": 1.7071429565548897,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.707142885774374,
      "rewards/format_reward_func": 1.0,
      "step": 3210
    },
    {
      "completion_length": 251.75893878936768,
      "epoch": 0.5384970032272937,
      "grad_norm": 0.22457165395344272,
      "kl": 0.15355682373046875,
      "learning_rate": 3.9654320987654323e-07,
      "loss": 0.0002,
      "reward": 1.7464286237955093,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.746428620070219,
      "rewards/format_reward_func": 1.0,
      "step": 3212
    },
    {
      "completion_length": 256.7812623977661,
      "epoch": 0.5388323064671613,
      "grad_norm": 0.21129731942518568,
      "kl": 0.11245346069335938,
      "learning_rate": 3.9679012345679006e-07,
      "loss": 0.0001,
      "reward": 1.762500062584877,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643200933933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3214
    },
    {
      "completion_length": 246.5000114440918,
      "epoch": 0.5391676097070288,
      "grad_norm": 0.18051497905212877,
      "kl": 0.03157806396484375,
      "learning_rate": 3.97037037037037e-07,
      "loss": 0.0,
      "reward": 1.7625000849366188,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643200933933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3216
    },
    {
      "completion_length": 245.3794765472412,
      "epoch": 0.5395029129468963,
      "grad_norm": 0.9203589714968138,
      "kl": 0.261016845703125,
      "learning_rate": 3.9728395061728395e-07,
      "loss": 0.0003,
      "reward": 1.6696429401636124,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.6741071753203869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3218
    },
    {
      "completion_length": 250.7232255935669,
      "epoch": 0.5398382161867639,
      "grad_norm": 0.3017788178994304,
      "kl": 0.03530120849609375,
      "learning_rate": 3.9753086419753083e-07,
      "loss": 0.0,
      "reward": 1.677678644657135,
      "reward_std": 0.07197336759418249,
      "rewards/equation_reward_func": 0.683928593993187,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 3220
    },
    {
      "completion_length": 249.57144165039062,
      "epoch": 0.5401735194266315,
      "grad_norm": 0.5284210803368328,
      "kl": 0.11692047119140625,
      "learning_rate": 3.977777777777778e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143223285675,
      "rewards/format_reward_func": 1.0,
      "step": 3222
    },
    {
      "completion_length": 246.1919765472412,
      "epoch": 0.540508822666499,
      "grad_norm": 0.13747975941449442,
      "kl": 0.24882888793945312,
      "learning_rate": 3.9802469135802466e-07,
      "loss": 0.0002,
      "reward": 1.8142857626080513,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857477068901,
      "rewards/format_reward_func": 1.0,
      "step": 3224
    },
    {
      "completion_length": 255.8080472946167,
      "epoch": 0.5408441259063665,
      "grad_norm": 0.2599070366859934,
      "kl": 0.08871078491210938,
      "learning_rate": 3.982716049382716e-07,
      "loss": 0.0001,
      "reward": 1.7803572043776512,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3226
    },
    {
      "completion_length": 250.17411613464355,
      "epoch": 0.5411794291462341,
      "grad_norm": 0.5749702489024938,
      "kl": 0.1636505126953125,
      "learning_rate": 3.985185185185185e-07,
      "loss": 0.0002,
      "reward": 1.7000000923871994,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7000000309199095,
      "rewards/format_reward_func": 1.0,
      "step": 3228
    },
    {
      "completion_length": 248.71429347991943,
      "epoch": 0.5415147323861017,
      "grad_norm": 0.4718805170337936,
      "kl": 0.0569000244140625,
      "learning_rate": 3.987654320987654e-07,
      "loss": 0.0001,
      "reward": 1.6589286774396896,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.6633928883820772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3230
    },
    {
      "completion_length": 251.4196538925171,
      "epoch": 0.5418500356259692,
      "grad_norm": 0.36286334724653574,
      "kl": 0.04682159423828125,
      "learning_rate": 3.990123456790123e-07,
      "loss": 0.0,
      "reward": 1.796428620815277,
      "reward_std": 0.0858629634603858,
      "rewards/equation_reward_func": 0.7964286133646965,
      "rewards/format_reward_func": 1.0,
      "step": 3232
    },
    {
      "completion_length": 253.1026906967163,
      "epoch": 0.5421853388658368,
      "grad_norm": 0.24718759183417172,
      "kl": 0.08046340942382812,
      "learning_rate": 3.9925925925925926e-07,
      "loss": 0.0001,
      "reward": 1.7875000461935997,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7919643297791481,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3234
    },
    {
      "completion_length": 253.43750953674316,
      "epoch": 0.5425206421057044,
      "grad_norm": 0.14230902666933232,
      "kl": 0.028270721435546875,
      "learning_rate": 3.9950617283950615e-07,
      "loss": 0.0,
      "reward": 1.7857143506407738,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 3236
    },
    {
      "completion_length": 254.06697940826416,
      "epoch": 0.5428559453455719,
      "grad_norm": 0.6630361438819221,
      "kl": 0.07397842407226562,
      "learning_rate": 3.997530864197531e-07,
      "loss": 0.0001,
      "reward": 1.7107143551111221,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7196429036557674,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3238
    },
    {
      "completion_length": 243.4062623977661,
      "epoch": 0.5431912485854394,
      "grad_norm": 0.15745796244051208,
      "kl": 0.023761749267578125,
      "learning_rate": 4e-07,
      "loss": 0.0,
      "reward": 1.8089286088943481,
      "reward_std": 0.017677669413387775,
      "rewards/equation_reward_func": 0.8133928924798965,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3240
    },
    {
      "completion_length": 247.17411994934082,
      "epoch": 0.543526551825307,
      "grad_norm": 0.11801541444411404,
      "kl": 0.045162200927734375,
      "learning_rate": 4.0024691358024687e-07,
      "loss": 0.0,
      "reward": 1.667857214808464,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.6678571905940771,
      "rewards/format_reward_func": 1.0,
      "step": 3242
    },
    {
      "completion_length": 246.5937614440918,
      "epoch": 0.5438618550651746,
      "grad_norm": 0.24052847857381288,
      "kl": 0.038970947265625,
      "learning_rate": 4.004938271604938e-07,
      "loss": 0.0,
      "reward": 1.7142858058214188,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7142857480794191,
      "rewards/format_reward_func": 1.0,
      "step": 3244
    },
    {
      "completion_length": 246.37054824829102,
      "epoch": 0.5441971583050421,
      "grad_norm": 0.20182668701059084,
      "kl": 0.0252532958984375,
      "learning_rate": 4.007407407407407e-07,
      "loss": 0.0,
      "reward": 1.791071467101574,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.795535746961832,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3246
    },
    {
      "completion_length": 248.64733505249023,
      "epoch": 0.5445324615449096,
      "grad_norm": 0.25054728863777276,
      "kl": 0.023464202880859375,
      "learning_rate": 4.0098765432098764e-07,
      "loss": 0.0,
      "reward": 1.7571429312229156,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7660714536905289,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3248
    },
    {
      "completion_length": 252.946439743042,
      "epoch": 0.5448677647847773,
      "grad_norm": 0.2787722432356983,
      "kl": 0.05184173583984375,
      "learning_rate": 4.0123456790123453e-07,
      "loss": 0.0001,
      "reward": 1.7125001102685928,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7169643118977547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3250
    },
    {
      "completion_length": 242.03572463989258,
      "epoch": 0.5452030680246448,
      "grad_norm": 0.18001654363123426,
      "kl": 0.0496063232421875,
      "learning_rate": 4.0148148148148147e-07,
      "loss": 0.0,
      "reward": 1.8392857611179352,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8392857238650322,
      "rewards/format_reward_func": 1.0,
      "step": 3252
    },
    {
      "completion_length": 244.8839406967163,
      "epoch": 0.5455383712645123,
      "grad_norm": 0.3107235931047464,
      "kl": 0.032939910888671875,
      "learning_rate": 4.017283950617284e-07,
      "loss": 0.0,
      "reward": 1.7642857879400253,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 3254
    },
    {
      "completion_length": 254.87054443359375,
      "epoch": 0.5458736745043798,
      "grad_norm": 0.5024745088879808,
      "kl": 0.0463104248046875,
      "learning_rate": 4.0197530864197525e-07,
      "loss": 0.0,
      "reward": 1.7660714983940125,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7705357447266579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3256
    },
    {
      "completion_length": 243.50447368621826,
      "epoch": 0.5462089777442475,
      "grad_norm": 0.19769072115630126,
      "kl": 0.028018951416015625,
      "learning_rate": 4.022222222222222e-07,
      "loss": 0.0,
      "reward": 1.782142922282219,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 3258
    },
    {
      "completion_length": 252.5134038925171,
      "epoch": 0.546544280984115,
      "grad_norm": 0.25804087534525366,
      "kl": 0.037353515625,
      "learning_rate": 4.0246913580246913e-07,
      "loss": 0.0,
      "reward": 1.7875000685453415,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7919643111526966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3260
    },
    {
      "completion_length": 249.3125114440918,
      "epoch": 0.5468795842239825,
      "grad_norm": 0.3788881858860527,
      "kl": 0.036830902099609375,
      "learning_rate": 4.02716049382716e-07,
      "loss": 0.0,
      "reward": 1.7178572192788124,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.717857176437974,
      "rewards/format_reward_func": 1.0,
      "step": 3262
    },
    {
      "completion_length": 249.19643878936768,
      "epoch": 0.5472148874638502,
      "grad_norm": 0.15403221802227582,
      "kl": 0.03723907470703125,
      "learning_rate": 4.0296296296296296e-07,
      "loss": 0.0,
      "reward": 1.757142923772335,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 3264
    },
    {
      "completion_length": 242.4509048461914,
      "epoch": 0.5475501907037177,
      "grad_norm": 0.4151835339606924,
      "kl": 0.03864288330078125,
      "learning_rate": 4.0320987654320985e-07,
      "loss": 0.0,
      "reward": 1.7982143461704254,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.8026785999536514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3266
    },
    {
      "completion_length": 257.71876335144043,
      "epoch": 0.5478854939435852,
      "grad_norm": 0.28616567137191395,
      "kl": 0.03456878662109375,
      "learning_rate": 4.034567901234568e-07,
      "loss": 0.0,
      "reward": 1.6892858296632767,
      "reward_std": 0.08586296439170837,
      "rewards/equation_reward_func": 0.698214303702116,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3268
    },
    {
      "completion_length": 247.76786613464355,
      "epoch": 0.5482207971834527,
      "grad_norm": 0.33138120271755234,
      "kl": 0.027130126953125,
      "learning_rate": 4.0370370370370373e-07,
      "loss": 0.0,
      "reward": 1.807142898440361,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8071428835391998,
      "rewards/format_reward_func": 1.0,
      "step": 3270
    },
    {
      "completion_length": 253.633939743042,
      "epoch": 0.5485561004233204,
      "grad_norm": 0.2025119387803373,
      "kl": 0.07606124877929688,
      "learning_rate": 4.0395061728395057e-07,
      "loss": 0.0001,
      "reward": 1.762500062584877,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3272
    },
    {
      "completion_length": 249.10269165039062,
      "epoch": 0.5488914036631879,
      "grad_norm": 0.3509880306028421,
      "kl": 0.06534576416015625,
      "learning_rate": 4.041975308641975e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143260538578,
      "rewards/format_reward_func": 1.0,
      "step": 3274
    },
    {
      "completion_length": 248.8884048461914,
      "epoch": 0.5492267069030554,
      "grad_norm": 0.3225348638068555,
      "kl": 0.06076812744140625,
      "learning_rate": 4.044444444444444e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.767857164144516,
      "rewards/format_reward_func": 1.0,
      "step": 3276
    },
    {
      "completion_length": 246.52679347991943,
      "epoch": 0.549562010142923,
      "grad_norm": 0.1652321697093135,
      "kl": 0.045623779296875,
      "learning_rate": 4.0469135802469134e-07,
      "loss": 0.0,
      "reward": 1.7678572237491608,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7678571790456772,
      "rewards/format_reward_func": 1.0,
      "step": 3278
    },
    {
      "completion_length": 248.196439743042,
      "epoch": 0.5498973133827906,
      "grad_norm": 0.27022884678968045,
      "kl": 0.03060150146484375,
      "learning_rate": 4.049382716049383e-07,
      "loss": 0.0,
      "reward": 1.8160714954137802,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8205357268452644,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3280
    },
    {
      "completion_length": 257.5267972946167,
      "epoch": 0.5502326166226581,
      "grad_norm": 0.4939771906485696,
      "kl": 0.14651107788085938,
      "learning_rate": 4.0518518518518517e-07,
      "loss": 0.0001,
      "reward": 1.7410714849829674,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7544643189758062,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3282
    },
    {
      "completion_length": 248.93750858306885,
      "epoch": 0.5505679198625256,
      "grad_norm": 0.3348022744898028,
      "kl": 0.032135009765625,
      "learning_rate": 4.054320987654321e-07,
      "loss": 0.0,
      "reward": 1.7125000655651093,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.716964315623045,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3284
    },
    {
      "completion_length": 253.27679538726807,
      "epoch": 0.5509032231023933,
      "grad_norm": 0.4286314695204813,
      "kl": 0.06396102905273438,
      "learning_rate": 4.05679012345679e-07,
      "loss": 0.0001,
      "reward": 1.7125000804662704,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7169643193483353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3286
    },
    {
      "completion_length": 244.12947750091553,
      "epoch": 0.5512385263422608,
      "grad_norm": 0.23906097739214408,
      "kl": 0.027721405029296875,
      "learning_rate": 4.059259259259259e-07,
      "loss": 0.0,
      "reward": 1.7767857983708382,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7812500298023224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3288
    },
    {
      "completion_length": 248.7991189956665,
      "epoch": 0.5515738295821283,
      "grad_norm": 0.3127992260300639,
      "kl": 0.031581878662109375,
      "learning_rate": 4.061728395061728e-07,
      "loss": 0.0,
      "reward": 1.7071429342031479,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7071428820490837,
      "rewards/format_reward_func": 1.0,
      "step": 3290
    },
    {
      "completion_length": 249.4330472946167,
      "epoch": 0.5519091328219959,
      "grad_norm": 0.20008099014907435,
      "kl": 0.0303955078125,
      "learning_rate": 4.064197530864197e-07,
      "loss": 0.0,
      "reward": 1.748214341700077,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.7526785936206579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3292
    },
    {
      "completion_length": 256.7366180419922,
      "epoch": 0.5522444360618635,
      "grad_norm": 0.2468886215141364,
      "kl": 0.029834747314453125,
      "learning_rate": 4.0666666666666666e-07,
      "loss": 0.0,
      "reward": 1.7267858013510704,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.731250025331974,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3294
    },
    {
      "completion_length": 255.7053680419922,
      "epoch": 0.552579739301731,
      "grad_norm": 0.34286991600304806,
      "kl": 0.032001495361328125,
      "learning_rate": 4.069135802469136e-07,
      "loss": 0.0,
      "reward": 1.792857177555561,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 3296
    },
    {
      "completion_length": 246.50001049041748,
      "epoch": 0.5529150425415985,
      "grad_norm": 0.24654089324264858,
      "kl": 0.024234771728515625,
      "learning_rate": 4.071604938271605e-07,
      "loss": 0.0,
      "reward": 1.7500000819563866,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 3298
    },
    {
      "completion_length": 256.11162090301514,
      "epoch": 0.5532503457814661,
      "grad_norm": 0.28781653670523644,
      "kl": 0.028278350830078125,
      "learning_rate": 4.0740740740740737e-07,
      "loss": 0.0,
      "reward": 1.7642857655882835,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7732143141329288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3300
    },
    {
      "completion_length": 249.9642972946167,
      "epoch": 0.5535856490213337,
      "grad_norm": 0.3905942900862037,
      "kl": 0.026691436767578125,
      "learning_rate": 4.0765432098765426e-07,
      "loss": 0.0,
      "reward": 1.778571493923664,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 3302
    },
    {
      "completion_length": 241.17858219146729,
      "epoch": 0.5539209522612012,
      "grad_norm": 0.29323430298274383,
      "kl": 0.028812408447265625,
      "learning_rate": 4.079012345679012e-07,
      "loss": 0.0,
      "reward": 1.814285770058632,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857477068901,
      "rewards/format_reward_func": 1.0,
      "step": 3304
    },
    {
      "completion_length": 242.20983219146729,
      "epoch": 0.5542562555010688,
      "grad_norm": 0.15078015109364112,
      "kl": 0.033367156982421875,
      "learning_rate": 4.0814814814814814e-07,
      "loss": 0.0,
      "reward": 1.8178571835160255,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8178571723401546,
      "rewards/format_reward_func": 1.0,
      "step": 3306
    },
    {
      "completion_length": 265.4196557998657,
      "epoch": 0.5545915587409364,
      "grad_norm": 0.3085516247377834,
      "kl": 0.03081512451171875,
      "learning_rate": 4.0839506172839503e-07,
      "loss": 0.0,
      "reward": 1.7428572103381157,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 3308
    },
    {
      "completion_length": 247.75447463989258,
      "epoch": 0.5549268619808039,
      "grad_norm": 0.3167820411329688,
      "kl": 0.0263214111328125,
      "learning_rate": 4.08641975308642e-07,
      "loss": 0.0,
      "reward": 1.735714390873909,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.7446428947150707,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3310
    },
    {
      "completion_length": 253.93304538726807,
      "epoch": 0.5552621652206714,
      "grad_norm": 0.26818302974849845,
      "kl": 0.02646636962890625,
      "learning_rate": 4.088888888888889e-07,
      "loss": 0.0,
      "reward": 1.7250000685453415,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000443309546,
      "rewards/format_reward_func": 1.0,
      "step": 3312
    },
    {
      "completion_length": 244.9107255935669,
      "epoch": 0.555597468460539,
      "grad_norm": 0.18918188197499258,
      "kl": 0.030887603759765625,
      "learning_rate": 4.0913580246913575e-07,
      "loss": 0.0,
      "reward": 1.7750000655651093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 3314
    },
    {
      "completion_length": 253.24108123779297,
      "epoch": 0.5559327717004066,
      "grad_norm": 0.1587575101234306,
      "kl": 0.03311920166015625,
      "learning_rate": 4.093827160493827e-07,
      "loss": 0.0,
      "reward": 1.7482143267989159,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7526785954833031,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3316
    },
    {
      "completion_length": 243.60715579986572,
      "epoch": 0.5562680749402741,
      "grad_norm": 0.1467960955689759,
      "kl": 0.02486419677734375,
      "learning_rate": 4.096296296296296e-07,
      "loss": 0.0,
      "reward": 1.7714286521077156,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 3318
    },
    {
      "completion_length": 266.9151906967163,
      "epoch": 0.5566033781801417,
      "grad_norm": 0.13892360946264437,
      "kl": 0.045444488525390625,
      "learning_rate": 4.098765432098765e-07,
      "loss": 0.0,
      "reward": 1.7714286595582962,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 3320
    },
    {
      "completion_length": 254.00447463989258,
      "epoch": 0.5569386814200092,
      "grad_norm": 0.24898486187793703,
      "kl": 0.030483245849609375,
      "learning_rate": 4.1012345679012346e-07,
      "loss": 0.0,
      "reward": 1.7946429327130318,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7991071678698063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3322
    },
    {
      "completion_length": 250.87501335144043,
      "epoch": 0.5572739846598768,
      "grad_norm": 0.2500029856545992,
      "kl": 0.03379058837890625,
      "learning_rate": 4.1037037037037035e-07,
      "loss": 0.0,
      "reward": 1.7535714879631996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714507102966,
      "rewards/format_reward_func": 1.0,
      "step": 3324
    },
    {
      "completion_length": 255.415189743042,
      "epoch": 0.5576092878997443,
      "grad_norm": 0.29364029372225736,
      "kl": 0.020549774169921875,
      "learning_rate": 4.106172839506173e-07,
      "loss": 0.0,
      "reward": 1.7285715192556381,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7285714596509933,
      "rewards/format_reward_func": 1.0,
      "step": 3326
    },
    {
      "completion_length": 266.12054920196533,
      "epoch": 0.5579445911396119,
      "grad_norm": 0.25937008559805713,
      "kl": 0.05391693115234375,
      "learning_rate": 4.1086419753086413e-07,
      "loss": 0.0001,
      "reward": 1.6928572207689285,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7017857618629932,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3328
    },
    {
      "completion_length": 254.53126525878906,
      "epoch": 0.5582798943794794,
      "grad_norm": 0.25669521443692506,
      "kl": 0.018245697021484375,
      "learning_rate": 4.1111111111111107e-07,
      "loss": 0.0,
      "reward": 1.805357187986374,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.809821454808116,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3330
    },
    {
      "completion_length": 271.227689743042,
      "epoch": 0.558615197619347,
      "grad_norm": 0.3242229226510621,
      "kl": 0.025936126708984375,
      "learning_rate": 4.11358024691358e-07,
      "loss": 0.0,
      "reward": 1.7857143431901932,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7946428842842579,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3332
    },
    {
      "completion_length": 257.14733123779297,
      "epoch": 0.5589505008592146,
      "grad_norm": 0.16505049686226236,
      "kl": 0.01825714111328125,
      "learning_rate": 4.116049382716049e-07,
      "loss": 0.0,
      "reward": 1.733928643167019,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7383929006755352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3334
    },
    {
      "completion_length": 243.60268878936768,
      "epoch": 0.5592858040990821,
      "grad_norm": 0.24723268697187017,
      "kl": 0.021701812744140625,
      "learning_rate": 4.1185185185185184e-07,
      "loss": 0.0,
      "reward": 1.7571429386734962,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428921073675,
      "rewards/format_reward_func": 1.0,
      "step": 3336
    },
    {
      "completion_length": 253.2500114440918,
      "epoch": 0.5596211073389497,
      "grad_norm": 0.2512410261507857,
      "kl": 0.032398223876953125,
      "learning_rate": 4.120987654320988e-07,
      "loss": 0.0,
      "reward": 1.7517857924103737,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500201165676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3338
    },
    {
      "completion_length": 247.8928680419922,
      "epoch": 0.5599564105788172,
      "grad_norm": 0.20871903585274287,
      "kl": 0.02616119384765625,
      "learning_rate": 4.1234567901234567e-07,
      "loss": 0.0,
      "reward": 1.8214286267757416,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8214285932481289,
      "rewards/format_reward_func": 1.0,
      "step": 3340
    },
    {
      "completion_length": 262.75893783569336,
      "epoch": 0.5602917138186848,
      "grad_norm": 0.3195999706923693,
      "kl": 0.020320892333984375,
      "learning_rate": 4.1259259259259256e-07,
      "loss": 0.0,
      "reward": 1.769642911851406,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3342
    },
    {
      "completion_length": 258.5803699493408,
      "epoch": 0.5606270170585523,
      "grad_norm": 0.12534649378964235,
      "kl": 0.021961212158203125,
      "learning_rate": 4.1283950617283945e-07,
      "loss": 0.0,
      "reward": 1.7589286267757416,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7633928768336773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3344
    },
    {
      "completion_length": 269.0892963409424,
      "epoch": 0.5609623202984199,
      "grad_norm": 0.18819239134083926,
      "kl": 0.022518157958984375,
      "learning_rate": 4.130864197530864e-07,
      "loss": 0.0,
      "reward": 1.7785715088248253,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 3346
    },
    {
      "completion_length": 256.68304443359375,
      "epoch": 0.5612976235382875,
      "grad_norm": 0.17556385032649469,
      "kl": 0.034328460693359375,
      "learning_rate": 4.1333333333333333e-07,
      "loss": 0.0,
      "reward": 1.714285783469677,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.722321467474103,
      "rewards/format_reward_func": 0.9919642880558968,
      "step": 3348
    },
    {
      "completion_length": 258.80804538726807,
      "epoch": 0.561632926778155,
      "grad_norm": 0.19901055099402243,
      "kl": 0.05564117431640625,
      "learning_rate": 4.135802469135802e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8000000193715096,
      "rewards/format_reward_func": 1.0,
      "step": 3350
    },
    {
      "completion_length": 258.51786708831787,
      "epoch": 0.5619682300180225,
      "grad_norm": 0.2753161669446979,
      "kl": 0.25748443603515625,
      "learning_rate": 4.1382716049382716e-07,
      "loss": 0.0003,
      "reward": 1.7946429252624512,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.8080357350409031,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3352
    },
    {
      "completion_length": 268.1294775009155,
      "epoch": 0.5623035332578901,
      "grad_norm": 0.25622783243812475,
      "kl": 0.02396392822265625,
      "learning_rate": 4.140740740740741e-07,
      "loss": 0.0,
      "reward": 1.7250000685453415,
      "reward_std": 0.10606601554900408,
      "rewards/equation_reward_func": 0.7428571581840515,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3354
    },
    {
      "completion_length": 259.7812614440918,
      "epoch": 0.5626388364977577,
      "grad_norm": 0.11022749186091951,
      "kl": 0.02692413330078125,
      "learning_rate": 4.1432098765432094e-07,
      "loss": 0.0,
      "reward": 1.771428644657135,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 3356
    },
    {
      "completion_length": 251.7053689956665,
      "epoch": 0.5629741397376252,
      "grad_norm": 0.25546924480034083,
      "kl": 0.02170562744140625,
      "learning_rate": 4.145679012345679e-07,
      "loss": 0.0,
      "reward": 1.8000000640749931,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.8000000156462193,
      "rewards/format_reward_func": 1.0,
      "step": 3358
    },
    {
      "completion_length": 268.7857246398926,
      "epoch": 0.5633094429774927,
      "grad_norm": 0.26079986354717527,
      "kl": 0.06954193115234375,
      "learning_rate": 4.1481481481481476e-07,
      "loss": 0.0001,
      "reward": 1.7910714820027351,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7955357357859612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3360
    },
    {
      "completion_length": 254.7366180419922,
      "epoch": 0.5636447462173603,
      "grad_norm": 0.10312987099632315,
      "kl": 0.0501708984375,
      "learning_rate": 4.150617283950617e-07,
      "loss": 0.0001,
      "reward": 1.8660714626312256,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8705357313156128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3362
    },
    {
      "completion_length": 271.85715770721436,
      "epoch": 0.5639800494572279,
      "grad_norm": 0.21244909731089875,
      "kl": 0.06446456909179688,
      "learning_rate": 4.1530864197530865e-07,
      "loss": 0.0001,
      "reward": 1.666071504354477,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.67946432903409,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3364
    },
    {
      "completion_length": 263.2009057998657,
      "epoch": 0.5643153526970954,
      "grad_norm": 0.282676138928249,
      "kl": 0.0485687255859375,
      "learning_rate": 4.1555555555555554e-07,
      "loss": 0.0,
      "reward": 1.7142857983708382,
      "reward_std": 0.09091372694820166,
      "rewards/equation_reward_func": 0.723214328289032,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3366
    },
    {
      "completion_length": 249.79465579986572,
      "epoch": 0.564650655936963,
      "grad_norm": 0.2782074907909122,
      "kl": 0.0198974609375,
      "learning_rate": 4.158024691358025e-07,
      "loss": 0.0,
      "reward": 1.782142922282219,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 3368
    },
    {
      "completion_length": 265.97768783569336,
      "epoch": 0.5649859591768306,
      "grad_norm": 0.28559346638349703,
      "kl": 0.0273895263671875,
      "learning_rate": 4.1604938271604937e-07,
      "loss": 0.0,
      "reward": 1.7500000670552254,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000447034836,
      "rewards/format_reward_func": 1.0,
      "step": 3370
    },
    {
      "completion_length": 257.76786708831787,
      "epoch": 0.5653212624166981,
      "grad_norm": 0.318607705266382,
      "kl": 0.022426605224609375,
      "learning_rate": 4.1629629629629625e-07,
      "loss": 0.0,
      "reward": 1.7642858028411865,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7642857450991869,
      "rewards/format_reward_func": 1.0,
      "step": 3372
    },
    {
      "completion_length": 269.2634057998657,
      "epoch": 0.5656565656565656,
      "grad_norm": 0.23426763199945552,
      "kl": 0.02051544189453125,
      "learning_rate": 4.165432098765432e-07,
      "loss": 0.0,
      "reward": 1.7928571924567223,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 3374
    },
    {
      "completion_length": 273.3705463409424,
      "epoch": 0.5659918688964332,
      "grad_norm": 0.23627110930730033,
      "kl": 0.06494903564453125,
      "learning_rate": 4.167901234567901e-07,
      "loss": 0.0001,
      "reward": 1.667857214808464,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.6767857521772385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3376
    },
    {
      "completion_length": 261.133939743042,
      "epoch": 0.5663271721363008,
      "grad_norm": 0.2841826348282849,
      "kl": 0.051708221435546875,
      "learning_rate": 4.17037037037037e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 3378
    },
    {
      "completion_length": 258.04911708831787,
      "epoch": 0.5666624753761683,
      "grad_norm": 0.12243006146358627,
      "kl": 0.0408172607421875,
      "learning_rate": 4.1728395061728397e-07,
      "loss": 0.0,
      "reward": 1.7357143238186836,
      "reward_std": 0.03030457627028227,
      "rewards/equation_reward_func": 0.744642898440361,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3380
    },
    {
      "completion_length": 264.5535831451416,
      "epoch": 0.5669977786160358,
      "grad_norm": 0.185726261550338,
      "kl": 0.03836822509765625,
      "learning_rate": 4.1753086419753085e-07,
      "loss": 0.0,
      "reward": 1.7839286178350449,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7883928902447224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3382
    },
    {
      "completion_length": 268.5580472946167,
      "epoch": 0.5673330818559035,
      "grad_norm": 0.646482019678551,
      "kl": 0.1957550048828125,
      "learning_rate": 4.177777777777778e-07,
      "loss": 0.0002,
      "reward": 1.725000075995922,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7428571581840515,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3384
    },
    {
      "completion_length": 262.4910840988159,
      "epoch": 0.567668385095771,
      "grad_norm": 0.28343271824101846,
      "kl": 0.032367706298828125,
      "learning_rate": 4.1802469135802463e-07,
      "loss": 0.0,
      "reward": 1.769642911851406,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7741071823984385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3386
    },
    {
      "completion_length": 267.2009048461914,
      "epoch": 0.5680036883356385,
      "grad_norm": 0.30762902087198446,
      "kl": 0.22566604614257812,
      "learning_rate": 4.1827160493827157e-07,
      "loss": 0.0002,
      "reward": 1.7607143372297287,
      "reward_std": 0.09596448950469494,
      "rewards/equation_reward_func": 0.769642885774374,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3388
    },
    {
      "completion_length": 261.4509029388428,
      "epoch": 0.568338991575506,
      "grad_norm": 0.11987462910732502,
      "kl": 0.06043243408203125,
      "learning_rate": 4.185185185185185e-07,
      "loss": 0.0001,
      "reward": 1.7285714969038963,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7375000324100256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3390
    },
    {
      "completion_length": 264.91072368621826,
      "epoch": 0.5686742948153737,
      "grad_norm": 0.19351392333550568,
      "kl": 0.05767250061035156,
      "learning_rate": 4.187654320987654e-07,
      "loss": 0.0001,
      "reward": 1.7803572043776512,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214618861675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3392
    },
    {
      "completion_length": 259.0089406967163,
      "epoch": 0.5690095980552412,
      "grad_norm": 0.19611064120910207,
      "kl": 0.3055915832519531,
      "learning_rate": 4.1901234567901234e-07,
      "loss": 0.0003,
      "reward": 1.6803572252392769,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.684821467846632,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3394
    },
    {
      "completion_length": 255.05804824829102,
      "epoch": 0.5693449012951087,
      "grad_norm": 0.23378041111291098,
      "kl": 0.034976959228515625,
      "learning_rate": 4.1925925925925923e-07,
      "loss": 0.0,
      "reward": 1.7642857655882835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857562750578,
      "rewards/format_reward_func": 1.0,
      "step": 3396
    },
    {
      "completion_length": 255.34822463989258,
      "epoch": 0.5696802045349764,
      "grad_norm": 0.20520753045702905,
      "kl": 0.12027740478515625,
      "learning_rate": 4.1950617283950617e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 3398
    },
    {
      "completion_length": 263.2142972946167,
      "epoch": 0.5700155077748439,
      "grad_norm": 0.23000060551566762,
      "kl": 0.09436416625976562,
      "learning_rate": 4.1975308641975306e-07,
      "loss": 0.0001,
      "reward": 1.7571429163217545,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571429014205933,
      "rewards/format_reward_func": 1.0,
      "step": 3400
    },
    {
      "completion_length": 257.0535840988159,
      "epoch": 0.5703508110147114,
      "grad_norm": 0.27011411244907235,
      "kl": 0.02211761474609375,
      "learning_rate": 4.1999999999999995e-07,
      "loss": 0.0,
      "reward": 1.7678572162985802,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 3402
    },
    {
      "completion_length": 260.5580472946167,
      "epoch": 0.5706861142545789,
      "grad_norm": 0.35785158580122634,
      "kl": 0.030055999755859375,
      "learning_rate": 4.202469135802469e-07,
      "loss": 0.0,
      "reward": 1.773214377462864,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.777678593993187,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3404
    },
    {
      "completion_length": 269.73662090301514,
      "epoch": 0.5710214174944466,
      "grad_norm": 0.2128801567734277,
      "kl": 0.246490478515625,
      "learning_rate": 4.2049382716049383e-07,
      "loss": 0.0002,
      "reward": 1.7785715013742447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 3406
    },
    {
      "completion_length": 270.5312623977661,
      "epoch": 0.5713567207343141,
      "grad_norm": 0.21184046708081816,
      "kl": 0.039031982421875,
      "learning_rate": 4.207407407407407e-07,
      "loss": 0.0,
      "reward": 1.7464286610484123,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286163449287,
      "rewards/format_reward_func": 1.0,
      "step": 3408
    },
    {
      "completion_length": 268.50001335144043,
      "epoch": 0.5716920239741816,
      "grad_norm": 0.46570394726523234,
      "kl": 0.2350921630859375,
      "learning_rate": 4.2098765432098766e-07,
      "loss": 0.0002,
      "reward": 1.735267922282219,
      "reward_std": 0.08144354820251465,
      "rewards/equation_reward_func": 0.7473214603960514,
      "rewards/format_reward_func": 0.9879464358091354,
      "step": 3410
    },
    {
      "completion_length": 267.25894260406494,
      "epoch": 0.5720273272140493,
      "grad_norm": 0.2918455764955246,
      "kl": 0.12889862060546875,
      "learning_rate": 4.2123456790123455e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000447034836,
      "rewards/format_reward_func": 1.0,
      "step": 3412
    },
    {
      "completion_length": 262.6294765472412,
      "epoch": 0.5723626304539168,
      "grad_norm": 0.24566651888157798,
      "kl": 0.15311050415039062,
      "learning_rate": 4.2148148148148144e-07,
      "loss": 0.0002,
      "reward": 1.7660714983940125,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3414
    },
    {
      "completion_length": 265.6651906967163,
      "epoch": 0.5726979336937843,
      "grad_norm": 0.3001935923900465,
      "kl": 0.13809585571289062,
      "learning_rate": 4.217283950617284e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7732143215835094,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3416
    },
    {
      "completion_length": 257.58483505249023,
      "epoch": 0.5730332369336518,
      "grad_norm": 0.2690218679515261,
      "kl": 0.030422210693359375,
      "learning_rate": 4.2197530864197527e-07,
      "loss": 0.0,
      "reward": 1.7464286461472511,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7553571704775095,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3418
    },
    {
      "completion_length": 260.55358600616455,
      "epoch": 0.5733685401735195,
      "grad_norm": 0.5002997706686756,
      "kl": 0.168212890625,
      "learning_rate": 4.222222222222222e-07,
      "loss": 0.0002,
      "reward": 1.6875000596046448,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.6919643338769674,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3420
    },
    {
      "completion_length": 254.7410831451416,
      "epoch": 0.573703843413387,
      "grad_norm": 0.2889726297534045,
      "kl": 0.024105072021484375,
      "learning_rate": 4.224691358024691e-07,
      "loss": 0.0,
      "reward": 1.7678572088479996,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7678571902215481,
      "rewards/format_reward_func": 1.0,
      "step": 3422
    },
    {
      "completion_length": 253.49108600616455,
      "epoch": 0.5740391466532545,
      "grad_norm": 0.21100531871607167,
      "kl": 0.02407073974609375,
      "learning_rate": 4.2271604938271604e-07,
      "loss": 0.0,
      "reward": 1.8285714983940125,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8285714499652386,
      "rewards/format_reward_func": 1.0,
      "step": 3424
    },
    {
      "completion_length": 262.3169775009155,
      "epoch": 0.5743744498931221,
      "grad_norm": 0.33953151575700286,
      "kl": 0.038204193115234375,
      "learning_rate": 4.22962962962963e-07,
      "loss": 0.0,
      "reward": 1.7232143580913544,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7276786006987095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3426
    },
    {
      "completion_length": 265.00447845458984,
      "epoch": 0.5747097531329897,
      "grad_norm": 0.33966484859625834,
      "kl": 0.032154083251953125,
      "learning_rate": 4.232098765432098e-07,
      "loss": 0.0,
      "reward": 1.7285714820027351,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7285714633762836,
      "rewards/format_reward_func": 1.0,
      "step": 3428
    },
    {
      "completion_length": 263.45537185668945,
      "epoch": 0.5750450563728572,
      "grad_norm": 0.2174688881241481,
      "kl": 0.09495925903320312,
      "learning_rate": 4.2345679012345676e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7553571686148643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3430
    },
    {
      "completion_length": 267.9419765472412,
      "epoch": 0.5753803596127247,
      "grad_norm": 0.18092360650560677,
      "kl": 0.2818183898925781,
      "learning_rate": 4.237037037037037e-07,
      "loss": 0.0003,
      "reward": 1.757142923772335,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 3432
    },
    {
      "completion_length": 268.40180015563965,
      "epoch": 0.5757156628525923,
      "grad_norm": 0.36947735776015517,
      "kl": 0.067230224609375,
      "learning_rate": 4.239506172839506e-07,
      "loss": 0.0001,
      "reward": 1.7517858073115349,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7562500312924385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3434
    },
    {
      "completion_length": 264.8303737640381,
      "epoch": 0.5760509660924599,
      "grad_norm": 0.18742152942426193,
      "kl": 0.06931304931640625,
      "learning_rate": 4.2419753086419753e-07,
      "loss": 0.0001,
      "reward": 1.8160714954137802,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.8205357491970062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3436
    },
    {
      "completion_length": 264.5312614440918,
      "epoch": 0.5763862693323274,
      "grad_norm": 0.2804896692149197,
      "kl": 0.05013275146484375,
      "learning_rate": 4.244444444444444e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 3438
    },
    {
      "completion_length": 271.4732275009155,
      "epoch": 0.576721572572195,
      "grad_norm": 0.18135283016552467,
      "kl": 0.050182342529296875,
      "learning_rate": 4.2469135802469136e-07,
      "loss": 0.0001,
      "reward": 1.7107143849134445,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7107143141329288,
      "rewards/format_reward_func": 1.0,
      "step": 3440
    },
    {
      "completion_length": 269.2544765472412,
      "epoch": 0.5770568758120626,
      "grad_norm": 0.2010359293113445,
      "kl": 0.035064697265625,
      "learning_rate": 4.2493827160493825e-07,
      "loss": 0.0,
      "reward": 1.7428572177886963,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571842610836,
      "rewards/format_reward_func": 1.0,
      "step": 3442
    },
    {
      "completion_length": 271.80358505249023,
      "epoch": 0.5773921790519301,
      "grad_norm": 0.23671808515711454,
      "kl": 0.025234222412109375,
      "learning_rate": 4.2518518518518513e-07,
      "loss": 0.0,
      "reward": 1.725000075995922,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000275671482,
      "rewards/format_reward_func": 1.0,
      "step": 3444
    },
    {
      "completion_length": 252.01340579986572,
      "epoch": 0.5777274822917976,
      "grad_norm": 0.1996366433463103,
      "kl": 0.024082183837890625,
      "learning_rate": 4.254320987654321e-07,
      "loss": 0.0,
      "reward": 1.785714328289032,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 3446
    },
    {
      "completion_length": 273.3348340988159,
      "epoch": 0.5780627855316652,
      "grad_norm": 0.14140392980556388,
      "kl": 0.033966064453125,
      "learning_rate": 4.2567901234567896e-07,
      "loss": 0.0,
      "reward": 1.803571492433548,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 3448
    },
    {
      "completion_length": 261.0446548461914,
      "epoch": 0.5783980887715328,
      "grad_norm": 0.25123462292319676,
      "kl": 0.0379486083984375,
      "learning_rate": 4.259259259259259e-07,
      "loss": 0.0,
      "reward": 1.7392857670783997,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857298254967,
      "rewards/format_reward_func": 1.0,
      "step": 3450
    },
    {
      "completion_length": 267.0893020629883,
      "epoch": 0.5787333920114003,
      "grad_norm": 0.2699059485478521,
      "kl": 0.11783599853515625,
      "learning_rate": 4.2617283950617285e-07,
      "loss": 0.0001,
      "reward": 1.69821435213089,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7026785928755999,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3452
    },
    {
      "completion_length": 268.49108600616455,
      "epoch": 0.5790686952512679,
      "grad_norm": 0.2084339363468767,
      "kl": 0.033172607421875,
      "learning_rate": 4.2641975308641973e-07,
      "loss": 0.0,
      "reward": 1.758482187986374,
      "reward_std": 0.03851206600666046,
      "rewards/equation_reward_func": 0.7660714592784643,
      "rewards/format_reward_func": 0.9924107193946838,
      "step": 3454
    },
    {
      "completion_length": 259.3839406967163,
      "epoch": 0.5794039984911354,
      "grad_norm": 0.17434411741363384,
      "kl": 0.0621795654296875,
      "learning_rate": 4.266666666666667e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 3456
    },
    {
      "completion_length": 265.0089387893677,
      "epoch": 0.579739301731003,
      "grad_norm": 0.3167726445807968,
      "kl": 0.03658294677734375,
      "learning_rate": 4.2691358024691356e-07,
      "loss": 0.0,
      "reward": 1.7785714715719223,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 3458
    },
    {
      "completion_length": 259.3928699493408,
      "epoch": 0.5800746049708705,
      "grad_norm": 0.2945341184109623,
      "kl": 0.0406341552734375,
      "learning_rate": 4.2716049382716045e-07,
      "loss": 0.0,
      "reward": 1.7821429297327995,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 3460
    },
    {
      "completion_length": 265.3660821914673,
      "epoch": 0.5804099082107381,
      "grad_norm": 0.0889414542345476,
      "kl": 0.028598785400390625,
      "learning_rate": 4.274074074074074e-07,
      "loss": 0.0,
      "reward": 1.7857143431901932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 3462
    },
    {
      "completion_length": 262.91072940826416,
      "epoch": 0.5807452114506056,
      "grad_norm": 0.13428184690834213,
      "kl": 0.03705596923828125,
      "learning_rate": 4.276543209876543e-07,
      "loss": 0.0,
      "reward": 1.7196429371833801,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7241071723401546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3464
    },
    {
      "completion_length": 263.2098340988159,
      "epoch": 0.5810805146904732,
      "grad_norm": 0.2829331040542585,
      "kl": 0.05352783203125,
      "learning_rate": 4.279012345679012e-07,
      "loss": 0.0001,
      "reward": 1.7464286610484123,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.7553571686148643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3466
    },
    {
      "completion_length": 272.1339416503906,
      "epoch": 0.5814158179303408,
      "grad_norm": 0.15916034359450193,
      "kl": 0.05171966552734375,
      "learning_rate": 4.2814814814814816e-07,
      "loss": 0.0001,
      "reward": 1.7196429446339607,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7241071723401546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3468
    },
    {
      "completion_length": 256.48215770721436,
      "epoch": 0.5817511211702083,
      "grad_norm": 0.3651088268384066,
      "kl": 0.0333251953125,
      "learning_rate": 4.2839506172839505e-07,
      "loss": 0.0,
      "reward": 1.7107143700122833,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7196428962051868,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3470
    },
    {
      "completion_length": 259.0134029388428,
      "epoch": 0.5820864244100759,
      "grad_norm": 0.41088836334113304,
      "kl": 0.04825592041015625,
      "learning_rate": 4.2864197530864194e-07,
      "loss": 0.0,
      "reward": 1.7553572282195091,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7598214671015739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3472
    },
    {
      "completion_length": 256.48662090301514,
      "epoch": 0.5824217276499434,
      "grad_norm": 0.14851198065644858,
      "kl": 0.032390594482421875,
      "learning_rate": 4.2888888888888883e-07,
      "loss": 0.0,
      "reward": 1.79464291036129,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7991071790456772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3474
    },
    {
      "completion_length": 258.7901916503906,
      "epoch": 0.582757030889811,
      "grad_norm": 0.1969503633329373,
      "kl": 0.0368194580078125,
      "learning_rate": 4.2913580246913577e-07,
      "loss": 0.0,
      "reward": 1.7089286595582962,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7133928835391998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3476
    },
    {
      "completion_length": 257.8928699493408,
      "epoch": 0.5830923341296785,
      "grad_norm": 0.3246895776320101,
      "kl": 0.12652206420898438,
      "learning_rate": 4.293827160493827e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 3478
    },
    {
      "completion_length": 258.3392963409424,
      "epoch": 0.5834276373695461,
      "grad_norm": 0.28282227891625317,
      "kl": 0.02780914306640625,
      "learning_rate": 4.296296296296296e-07,
      "loss": 0.0,
      "reward": 1.7375000789761543,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7508928887546062,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3480
    },
    {
      "completion_length": 257.0491189956665,
      "epoch": 0.5837629406094136,
      "grad_norm": 0.18441674849853126,
      "kl": 0.02960205078125,
      "learning_rate": 4.2987654320987654e-07,
      "loss": 0.0,
      "reward": 1.7785714864730835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 1.0,
      "step": 3482
    },
    {
      "completion_length": 260.7812614440918,
      "epoch": 0.5840982438492812,
      "grad_norm": 0.23324259920480803,
      "kl": 0.03562164306640625,
      "learning_rate": 4.301234567901235e-07,
      "loss": 0.0,
      "reward": 1.725000075995922,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7250000238418579,
      "rewards/format_reward_func": 1.0,
      "step": 3484
    },
    {
      "completion_length": 263.1651906967163,
      "epoch": 0.5844335470891487,
      "grad_norm": 0.41779317202713057,
      "kl": 0.08310699462890625,
      "learning_rate": 4.303703703703703e-07,
      "loss": 0.0001,
      "reward": 1.6625000908970833,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.6758929044008255,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3486
    },
    {
      "completion_length": 254.05804824829102,
      "epoch": 0.5847688503290163,
      "grad_norm": 0.21896304164973182,
      "kl": 0.16289520263671875,
      "learning_rate": 4.3061728395061726e-07,
      "loss": 0.0002,
      "reward": 1.7589286267757416,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7633928880095482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3488
    },
    {
      "completion_length": 250.95090103149414,
      "epoch": 0.5851041535688839,
      "grad_norm": 0.15230906377968959,
      "kl": 0.1225433349609375,
      "learning_rate": 4.3086419753086415e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857544124126,
      "rewards/format_reward_func": 1.0,
      "step": 3490
    },
    {
      "completion_length": 253.61161708831787,
      "epoch": 0.5854394568087514,
      "grad_norm": 0.13634640897201905,
      "kl": 0.06240081787109375,
      "learning_rate": 4.311111111111111e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.792857164517045,
      "rewards/format_reward_func": 1.0,
      "step": 3492
    },
    {
      "completion_length": 259.2410831451416,
      "epoch": 0.585774760048619,
      "grad_norm": 0.20498559394312932,
      "kl": 0.02706146240234375,
      "learning_rate": 4.3135802469135803e-07,
      "loss": 0.0,
      "reward": 1.8125000298023224,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.8169643245637417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3494
    },
    {
      "completion_length": 249.1384048461914,
      "epoch": 0.5861100632884865,
      "grad_norm": 0.31684540637992265,
      "kl": 0.05062103271484375,
      "learning_rate": 4.316049382716049e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 3496
    },
    {
      "completion_length": 265.8660821914673,
      "epoch": 0.5864453665283541,
      "grad_norm": 0.2820708311134308,
      "kl": 0.08450698852539062,
      "learning_rate": 4.3185185185185186e-07,
      "loss": 0.0001,
      "reward": 1.8250000402331352,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.8339285887777805,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3498
    },
    {
      "completion_length": 258.276798248291,
      "epoch": 0.5867806697682216,
      "grad_norm": 0.18988002197728343,
      "kl": 0.0531158447265625,
      "learning_rate": 4.320987654320987e-07,
      "loss": 0.0001,
      "reward": 1.792857214808464,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.8017857559025288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3500
    },
    {
      "completion_length": 253.03572940826416,
      "epoch": 0.5871159730080892,
      "grad_norm": 0.17893344507554287,
      "kl": 0.02783966064453125,
      "learning_rate": 4.3234567901234564e-07,
      "loss": 0.0,
      "reward": 1.8071429133415222,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428909897804,
      "rewards/format_reward_func": 1.0,
      "step": 3502
    },
    {
      "completion_length": 259.8973321914673,
      "epoch": 0.5874512762479568,
      "grad_norm": 0.22936645917658965,
      "kl": 0.02904510498046875,
      "learning_rate": 4.325925925925926e-07,
      "loss": 0.0,
      "reward": 1.7857143580913544,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 3504
    },
    {
      "completion_length": 260.5357275009155,
      "epoch": 0.5877865794878243,
      "grad_norm": 0.41946924390943746,
      "kl": 0.3809051513671875,
      "learning_rate": 4.3283950617283947e-07,
      "loss": 0.0004,
      "reward": 1.7357143387198448,
      "reward_std": 0.10606601741164923,
      "rewards/equation_reward_func": 0.7625000216066837,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 3506
    },
    {
      "completion_length": 259.0982275009155,
      "epoch": 0.5881218827276918,
      "grad_norm": 0.22516693768991086,
      "kl": 0.03868865966796875,
      "learning_rate": 4.330864197530864e-07,
      "loss": 0.0,
      "reward": 1.76071435213089,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7696428932249546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3508
    },
    {
      "completion_length": 246.68304634094238,
      "epoch": 0.5884571859675594,
      "grad_norm": 0.16891949258278482,
      "kl": 0.02500152587890625,
      "learning_rate": 4.3333333333333335e-07,
      "loss": 0.0,
      "reward": 1.7660715207457542,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3510
    },
    {
      "completion_length": 256.5044746398926,
      "epoch": 0.588792489207427,
      "grad_norm": 0.16682964616175375,
      "kl": 0.03672027587890625,
      "learning_rate": 4.3358024691358024e-07,
      "loss": 0.0,
      "reward": 1.7732143551111221,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7776785865426064,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3512
    },
    {
      "completion_length": 262.71429538726807,
      "epoch": 0.5891277924472945,
      "grad_norm": 0.22590571064361892,
      "kl": 0.036014556884765625,
      "learning_rate": 4.338271604938271e-07,
      "loss": 0.0,
      "reward": 1.7714286297559738,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7803571689873934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3514
    },
    {
      "completion_length": 262.2366199493408,
      "epoch": 0.589463095687162,
      "grad_norm": 0.2227653303159639,
      "kl": 0.05431365966796875,
      "learning_rate": 4.34074074074074e-07,
      "loss": 0.0001,
      "reward": 1.716071493923664,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.7294643130153418,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3516
    },
    {
      "completion_length": 263.89287090301514,
      "epoch": 0.5897983989270297,
      "grad_norm": 0.22453191240129117,
      "kl": 0.161773681640625,
      "learning_rate": 4.3432098765432096e-07,
      "loss": 0.0002,
      "reward": 1.7642857730388641,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 3518
    },
    {
      "completion_length": 261.9509029388428,
      "epoch": 0.5901337021668972,
      "grad_norm": 0.25398918227897027,
      "kl": 0.0648956298828125,
      "learning_rate": 4.345679012345679e-07,
      "loss": 0.0001,
      "reward": 1.7357143834233284,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 3520
    },
    {
      "completion_length": 262.4598340988159,
      "epoch": 0.5904690054067647,
      "grad_norm": 0.2944405516763985,
      "kl": 0.025665283203125,
      "learning_rate": 4.348148148148148e-07,
      "loss": 0.0,
      "reward": 1.7642857730388641,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7732143104076385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3522
    },
    {
      "completion_length": 271.6696586608887,
      "epoch": 0.5908043086466322,
      "grad_norm": 0.24076922620858562,
      "kl": 0.0668487548828125,
      "learning_rate": 4.350617283950617e-07,
      "loss": 0.0001,
      "reward": 1.7482143566012383,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526786103844643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3524
    },
    {
      "completion_length": 262.7812604904175,
      "epoch": 0.5911396118864999,
      "grad_norm": 0.3099566866844783,
      "kl": 0.07646942138671875,
      "learning_rate": 4.3530864197530867e-07,
      "loss": 0.0001,
      "reward": 1.746428668498993,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7642857357859612,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3526
    },
    {
      "completion_length": 258.2232255935669,
      "epoch": 0.5914749151263674,
      "grad_norm": 0.17710092998351407,
      "kl": 0.031280517578125,
      "learning_rate": 4.355555555555555e-07,
      "loss": 0.0,
      "reward": 1.8071429207921028,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8160714544355869,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3528
    },
    {
      "completion_length": 249.54465579986572,
      "epoch": 0.5918102183662349,
      "grad_norm": 0.2899026154248438,
      "kl": 0.0284576416015625,
      "learning_rate": 4.3580246913580244e-07,
      "loss": 0.0,
      "reward": 1.7875000685453415,
      "reward_std": 0.06818529777228832,
      "rewards/equation_reward_func": 0.8008928876370192,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3530
    },
    {
      "completion_length": 255.883939743042,
      "epoch": 0.5921455216061026,
      "grad_norm": 0.15538153919110007,
      "kl": 0.02648162841796875,
      "learning_rate": 4.3604938271604933e-07,
      "loss": 0.0,
      "reward": 1.800000049173832,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 3532
    },
    {
      "completion_length": 263.08929920196533,
      "epoch": 0.5924808248459701,
      "grad_norm": 0.3173318092500211,
      "kl": 0.06317138671875,
      "learning_rate": 4.362962962962963e-07,
      "loss": 0.0001,
      "reward": 1.7875000685453415,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7919643074274063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3534
    },
    {
      "completion_length": 252.59376049041748,
      "epoch": 0.5928161280858376,
      "grad_norm": 0.345473218389504,
      "kl": 0.054718017578125,
      "learning_rate": 4.365432098765432e-07,
      "loss": 0.0001,
      "reward": 1.787500061094761,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7919643148779869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3536
    },
    {
      "completion_length": 266.7723331451416,
      "epoch": 0.5931514313257051,
      "grad_norm": 0.1637067972754746,
      "kl": 0.04750823974609375,
      "learning_rate": 4.367901234567901e-07,
      "loss": 0.0,
      "reward": 1.7357143461704254,
      "reward_std": 0.03030457627028227,
      "rewards/equation_reward_func": 0.7446428872644901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3538
    },
    {
      "completion_length": 259.96875858306885,
      "epoch": 0.5934867345655728,
      "grad_norm": 0.30374336459795204,
      "kl": 0.037841796875,
      "learning_rate": 4.3703703703703704e-07,
      "loss": 0.0,
      "reward": 1.762500062584877,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.766964316368103,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3540
    },
    {
      "completion_length": 257.20537090301514,
      "epoch": 0.5938220378054403,
      "grad_norm": 0.18775558730289618,
      "kl": 0.18558502197265625,
      "learning_rate": 4.372839506172839e-07,
      "loss": 0.0002,
      "reward": 1.7482143566012383,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3542
    },
    {
      "completion_length": 267.7589416503906,
      "epoch": 0.5941573410453078,
      "grad_norm": 0.27798905796014955,
      "kl": 0.0243682861328125,
      "learning_rate": 4.375308641975308e-07,
      "loss": 0.0,
      "reward": 1.7464286237955093,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7553571723401546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3544
    },
    {
      "completion_length": 254.3839406967163,
      "epoch": 0.5944926442851755,
      "grad_norm": 0.16482240300876788,
      "kl": 0.2323150634765625,
      "learning_rate": 4.3777777777777776e-07,
      "loss": 0.0002,
      "reward": 1.7535714954137802,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7535714767873287,
      "rewards/format_reward_func": 1.0,
      "step": 3546
    },
    {
      "completion_length": 269.70090103149414,
      "epoch": 0.594827947525043,
      "grad_norm": 0.21117970471010705,
      "kl": 0.17474365234375,
      "learning_rate": 4.3802469135802465e-07,
      "loss": 0.0002,
      "reward": 1.7589286267757416,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.772321455180645,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3548
    },
    {
      "completion_length": 260.1517972946167,
      "epoch": 0.5951632507649105,
      "grad_norm": 0.2455256251100981,
      "kl": 0.029510498046875,
      "learning_rate": 4.382716049382716e-07,
      "loss": 0.0,
      "reward": 1.8000000715255737,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.817857164889574,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3550
    },
    {
      "completion_length": 267.6428699493408,
      "epoch": 0.595498554004778,
      "grad_norm": 0.21551056799708354,
      "kl": 0.03183746337890625,
      "learning_rate": 4.3851851851851853e-07,
      "loss": 0.0,
      "reward": 1.7196429520845413,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.724107176065445,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3552
    },
    {
      "completion_length": 258.040189743042,
      "epoch": 0.5958338572446457,
      "grad_norm": 0.1783330798048646,
      "kl": 0.14756011962890625,
      "learning_rate": 4.387654320987654e-07,
      "loss": 0.0001,
      "reward": 1.707142949104309,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7071428969502449,
      "rewards/format_reward_func": 1.0,
      "step": 3554
    },
    {
      "completion_length": 249.8259038925171,
      "epoch": 0.5961691604845132,
      "grad_norm": 0.21122598255990901,
      "kl": 0.02812957763671875,
      "learning_rate": 4.3901234567901236e-07,
      "loss": 0.0,
      "reward": 1.773214340209961,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7776786088943481,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3556
    },
    {
      "completion_length": 250.7991132736206,
      "epoch": 0.5965044637243807,
      "grad_norm": 0.1805206880065137,
      "kl": 0.0556182861328125,
      "learning_rate": 4.392592592592592e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 3558
    },
    {
      "completion_length": 250.87054634094238,
      "epoch": 0.5968397669642483,
      "grad_norm": 0.7938875469714126,
      "kl": 0.3453216552734375,
      "learning_rate": 4.3950617283950614e-07,
      "loss": 0.0003,
      "reward": 1.7946429401636124,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.799107164144516,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3560
    },
    {
      "completion_length": 260.9732265472412,
      "epoch": 0.5971750702041159,
      "grad_norm": 0.19210387107786725,
      "kl": 0.04297637939453125,
      "learning_rate": 4.397530864197531e-07,
      "loss": 0.0,
      "reward": 1.7464286386966705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464286144822836,
      "rewards/format_reward_func": 1.0,
      "step": 3562
    },
    {
      "completion_length": 245.0669755935669,
      "epoch": 0.5975103734439834,
      "grad_norm": 0.2801304059499941,
      "kl": 0.0260467529296875,
      "learning_rate": 4.3999999999999997e-07,
      "loss": 0.0,
      "reward": 1.7928571924567223,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 3564
    },
    {
      "completion_length": 271.33037090301514,
      "epoch": 0.5978456766838509,
      "grad_norm": 0.11718352236281365,
      "kl": 0.06449508666992188,
      "learning_rate": 4.402469135802469e-07,
      "loss": 0.0001,
      "reward": 1.741071492433548,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7544643245637417,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3566
    },
    {
      "completion_length": 260.3348321914673,
      "epoch": 0.5981809799237185,
      "grad_norm": 0.3075517351831485,
      "kl": 0.08880615234375,
      "learning_rate": 4.404938271604938e-07,
      "loss": 0.0001,
      "reward": 1.687500074505806,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.7098214663565159,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 3568
    },
    {
      "completion_length": 263.1026906967163,
      "epoch": 0.5985162831635861,
      "grad_norm": 0.267592381406727,
      "kl": 0.07773208618164062,
      "learning_rate": 4.4074074074074074e-07,
      "loss": 0.0001,
      "reward": 1.7357143685221672,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7357143014669418,
      "rewards/format_reward_func": 1.0,
      "step": 3570
    },
    {
      "completion_length": 259.83036708831787,
      "epoch": 0.5988515864034536,
      "grad_norm": 0.16486227865906747,
      "kl": 0.0298004150390625,
      "learning_rate": 4.4098765432098763e-07,
      "loss": 0.0,
      "reward": 1.7589286342263222,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7633928880095482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3572
    },
    {
      "completion_length": 263.4285831451416,
      "epoch": 0.5991868896433212,
      "grad_norm": 0.2593585175782869,
      "kl": 0.02993011474609375,
      "learning_rate": 4.412345679012345e-07,
      "loss": 0.0,
      "reward": 1.717857226729393,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7267857454717159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3574
    },
    {
      "completion_length": 253.36161994934082,
      "epoch": 0.5995221928831888,
      "grad_norm": 0.22218048407968422,
      "kl": 0.029876708984375,
      "learning_rate": 4.4148148148148146e-07,
      "loss": 0.0,
      "reward": 1.8071429133415222,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8071428909897804,
      "rewards/format_reward_func": 1.0,
      "step": 3576
    },
    {
      "completion_length": 255.6785831451416,
      "epoch": 0.5998574961230563,
      "grad_norm": 0.27713844671160104,
      "kl": 0.02791595458984375,
      "learning_rate": 4.417283950617284e-07,
      "loss": 0.0,
      "reward": 1.7214286550879478,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.721428606659174,
      "rewards/format_reward_func": 1.0,
      "step": 3578
    },
    {
      "completion_length": 258.5223331451416,
      "epoch": 0.6001927993629238,
      "grad_norm": 0.2268908937876009,
      "kl": 0.14574432373046875,
      "learning_rate": 4.419753086419753e-07,
      "loss": 0.0001,
      "reward": 1.7839286252856255,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7973214574158192,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3580
    },
    {
      "completion_length": 260.96876430511475,
      "epoch": 0.6005281026027914,
      "grad_norm": 0.008560615600363496,
      "kl": 0.04407501220703125,
      "learning_rate": 4.4222222222222223e-07,
      "loss": 0.0,
      "reward": 1.7892857789993286,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 3582
    },
    {
      "completion_length": 263.3794746398926,
      "epoch": 0.600863405842659,
      "grad_norm": 0.3125592782204781,
      "kl": 0.039703369140625,
      "learning_rate": 4.424691358024691e-07,
      "loss": 0.0,
      "reward": 1.7321429401636124,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7410714589059353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3584
    },
    {
      "completion_length": 265.0401945114136,
      "epoch": 0.6011987090825265,
      "grad_norm": 0.23566926170906846,
      "kl": 0.064239501953125,
      "learning_rate": 4.42716049382716e-07,
      "loss": 0.0001,
      "reward": 1.7482143566012383,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3586
    },
    {
      "completion_length": 260.8169765472412,
      "epoch": 0.6015340123223941,
      "grad_norm": 0.18104321652659505,
      "kl": 0.06757354736328125,
      "learning_rate": 4.4296296296296295e-07,
      "loss": 0.0001,
      "reward": 1.7142857685685158,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857518047094,
      "rewards/format_reward_func": 1.0,
      "step": 3588
    },
    {
      "completion_length": 247.33929824829102,
      "epoch": 0.6018693155622616,
      "grad_norm": 0.2856698746219867,
      "kl": 0.051727294921875,
      "learning_rate": 4.4320987654320984e-07,
      "loss": 0.0001,
      "reward": 1.7250000834465027,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000312924385,
      "rewards/format_reward_func": 1.0,
      "step": 3590
    },
    {
      "completion_length": 258.3259029388428,
      "epoch": 0.6022046188021292,
      "grad_norm": 0.4413945663867954,
      "kl": 0.070465087890625,
      "learning_rate": 4.434567901234568e-07,
      "loss": 0.0001,
      "reward": 1.7428572177886963,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 3592
    },
    {
      "completion_length": 260.0669746398926,
      "epoch": 0.6025399220419967,
      "grad_norm": 0.20788258491713757,
      "kl": 0.076507568359375,
      "learning_rate": 4.4370370370370367e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7642857320606709,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3594
    },
    {
      "completion_length": 257.54019260406494,
      "epoch": 0.6028752252818643,
      "grad_norm": 0.35534726550307105,
      "kl": 0.02875518798828125,
      "learning_rate": 4.439506172839506e-07,
      "loss": 0.0,
      "reward": 1.7821429297327995,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7821428775787354,
      "rewards/format_reward_func": 1.0,
      "step": 3596
    },
    {
      "completion_length": 265.495548248291,
      "epoch": 0.6032105285217318,
      "grad_norm": 0.32475422464455267,
      "kl": 0.25457000732421875,
      "learning_rate": 4.4419753086419755e-07,
      "loss": 0.0003,
      "reward": 1.691071517765522,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.6955357398837805,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3598
    },
    {
      "completion_length": 256.9910821914673,
      "epoch": 0.6035458317615994,
      "grad_norm": 0.3155386853226163,
      "kl": 0.033954620361328125,
      "learning_rate": 4.444444444444444e-07,
      "loss": 0.0,
      "reward": 1.7339286506175995,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7383928932249546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3600
    },
    {
      "completion_length": 249.37501049041748,
      "epoch": 0.603881135001467,
      "grad_norm": 0.3366549119682878,
      "kl": 0.15158843994140625,
      "learning_rate": 4.446913580246913e-07,
      "loss": 0.0002,
      "reward": 1.8053571954369545,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8098214510828257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3602
    },
    {
      "completion_length": 261.1160831451416,
      "epoch": 0.6042164382413345,
      "grad_norm": 0.2374163328003629,
      "kl": 0.10363006591796875,
      "learning_rate": 4.4493827160493827e-07,
      "loss": 0.0001,
      "reward": 1.7964286506175995,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.796428594738245,
      "rewards/format_reward_func": 1.0,
      "step": 3604
    },
    {
      "completion_length": 265.6518039703369,
      "epoch": 0.604551741481202,
      "grad_norm": 0.24493017949672824,
      "kl": 0.33367919921875,
      "learning_rate": 4.4518518518518515e-07,
      "loss": 0.0003,
      "reward": 1.7375001013278961,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7508928775787354,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3606
    },
    {
      "completion_length": 267.93304920196533,
      "epoch": 0.6048870447210696,
      "grad_norm": 0.4602830429080869,
      "kl": 0.5742149353027344,
      "learning_rate": 4.454320987654321e-07,
      "loss": 0.0006,
      "reward": 1.778571479022503,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7875000312924385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3608
    },
    {
      "completion_length": 269.2678699493408,
      "epoch": 0.6052223479609372,
      "grad_norm": 0.3401300966547623,
      "kl": 0.16431427001953125,
      "learning_rate": 4.45679012345679e-07,
      "loss": 0.0002,
      "reward": 1.7714286372065544,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7803571745753288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3610
    },
    {
      "completion_length": 274.7009029388428,
      "epoch": 0.6055576512008047,
      "grad_norm": 0.32991585835717985,
      "kl": 0.12230682373046875,
      "learning_rate": 4.459259259259259e-07,
      "loss": 0.0001,
      "reward": 1.7196429297327995,
      "reward_std": 0.09343910869210958,
      "rewards/equation_reward_func": 0.7330357395112514,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3612
    },
    {
      "completion_length": 270.4866189956665,
      "epoch": 0.6058929544406723,
      "grad_norm": 0.255009936150431,
      "kl": 0.03687286376953125,
      "learning_rate": 4.461728395061728e-07,
      "loss": 0.0,
      "reward": 1.7625000551342964,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.7758928798139095,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3614
    },
    {
      "completion_length": 275.08929920196533,
      "epoch": 0.6062282576805398,
      "grad_norm": 0.18435769219493214,
      "kl": 0.14650726318359375,
      "learning_rate": 4.464197530864197e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7589286118745804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3616
    },
    {
      "completion_length": 275.9955472946167,
      "epoch": 0.6065635609204074,
      "grad_norm": 0.50317108046287,
      "kl": 0.8645095825195312,
      "learning_rate": 4.4666666666666664e-07,
      "loss": 0.0009,
      "reward": 1.7982143312692642,
      "reward_std": 0.09343910869210958,
      "rewards/equation_reward_func": 0.8116071671247482,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3618
    },
    {
      "completion_length": 265.30804538726807,
      "epoch": 0.6068988641602749,
      "grad_norm": 0.2825123505770023,
      "kl": 0.03264617919921875,
      "learning_rate": 4.4691358024691353e-07,
      "loss": 0.0,
      "reward": 1.7464286386966705,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.755357176065445,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3620
    },
    {
      "completion_length": 271.2857246398926,
      "epoch": 0.6072341674001425,
      "grad_norm": 0.18990244417672877,
      "kl": 0.08856964111328125,
      "learning_rate": 4.4716049382716047e-07,
      "loss": 0.0001,
      "reward": 1.7053572162985802,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7187500335276127,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3622
    },
    {
      "completion_length": 255.42411994934082,
      "epoch": 0.6075694706400101,
      "grad_norm": 0.28196388269226436,
      "kl": 0.16127777099609375,
      "learning_rate": 4.474074074074074e-07,
      "loss": 0.0002,
      "reward": 1.8125000521540642,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8169643022119999,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3624
    },
    {
      "completion_length": 273.1428689956665,
      "epoch": 0.6079047738798776,
      "grad_norm": 0.4959674196608172,
      "kl": 1.2173080444335938,
      "learning_rate": 4.476543209876543e-07,
      "loss": 0.0012,
      "reward": 1.714285783469677,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.732142886146903,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3626
    },
    {
      "completion_length": 264.4330472946167,
      "epoch": 0.6082400771197451,
      "grad_norm": 0.18397191856889522,
      "kl": 0.22011566162109375,
      "learning_rate": 4.479012345679012e-07,
      "loss": 0.0002,
      "reward": 1.7714286521077156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 3628
    },
    {
      "completion_length": 265.06697845458984,
      "epoch": 0.6085753803596127,
      "grad_norm": 0.27923133250657345,
      "kl": 0.48724365234375,
      "learning_rate": 4.4814814814814813e-07,
      "loss": 0.0005,
      "reward": 1.7446429207921028,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7580357510596514,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3630
    },
    {
      "completion_length": 263.29912185668945,
      "epoch": 0.6089106835994803,
      "grad_norm": 0.2536893755418448,
      "kl": 0.7797088623046875,
      "learning_rate": 4.48395061728395e-07,
      "loss": 0.0008,
      "reward": 1.8160715252161026,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.820535734295845,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3632
    },
    {
      "completion_length": 263.401798248291,
      "epoch": 0.6092459868393478,
      "grad_norm": 0.18328565704488098,
      "kl": 0.48796844482421875,
      "learning_rate": 4.4864197530864196e-07,
      "loss": 0.0005,
      "reward": 1.7196429297327995,
      "reward_std": 0.0328299580141902,
      "rewards/equation_reward_func": 0.7241071872413158,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3634
    },
    {
      "completion_length": 257.8973331451416,
      "epoch": 0.6095812900792154,
      "grad_norm": 0.4203365404663292,
      "kl": 0.1136932373046875,
      "learning_rate": 4.4888888888888885e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7732142955064774,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3636
    },
    {
      "completion_length": 261.6384038925171,
      "epoch": 0.609916593319083,
      "grad_norm": 0.27518745976881687,
      "kl": 0.206268310546875,
      "learning_rate": 4.491358024691358e-07,
      "loss": 0.0002,
      "reward": 1.6767858192324638,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.681250024586916,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3638
    },
    {
      "completion_length": 258.3660840988159,
      "epoch": 0.6102518965589505,
      "grad_norm": 0.28078115457804403,
      "kl": 0.970550537109375,
      "learning_rate": 4.4938271604938273e-07,
      "loss": 0.001,
      "reward": 1.7750000655651093,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 3640
    },
    {
      "completion_length": 261.7500104904175,
      "epoch": 0.610587199798818,
      "grad_norm": 0.38733232869582146,
      "kl": 1.2825469970703125,
      "learning_rate": 4.496296296296296e-07,
      "loss": 0.0013,
      "reward": 1.7160715013742447,
      "reward_std": 0.09848987404257059,
      "rewards/equation_reward_func": 0.7383928783237934,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 3642
    },
    {
      "completion_length": 254.3928680419922,
      "epoch": 0.6109225030386856,
      "grad_norm": 0.22281246987673115,
      "kl": 0.1315765380859375,
      "learning_rate": 4.498765432098765e-07,
      "loss": 0.0001,
      "reward": 1.7642857655882835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 3644
    },
    {
      "completion_length": 275.2098331451416,
      "epoch": 0.6112578062785532,
      "grad_norm": 0.3146735029654327,
      "kl": 1.4842376708984375,
      "learning_rate": 4.501234567901234e-07,
      "loss": 0.0015,
      "reward": 1.676785796880722,
      "reward_std": 0.11364216171205044,
      "rewards/equation_reward_func": 0.6991071626543999,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 3646
    },
    {
      "completion_length": 255.4241180419922,
      "epoch": 0.6115931095184207,
      "grad_norm": 0.2005355282523996,
      "kl": 0.8999481201171875,
      "learning_rate": 4.5037037037037034e-07,
      "loss": 0.0009,
      "reward": 1.7816964611411095,
      "reward_std": 0.03598668519407511,
      "rewards/equation_reward_func": 0.7937500290572643,
      "rewards/format_reward_func": 0.9879464358091354,
      "step": 3648
    },
    {
      "completion_length": 274.8169755935669,
      "epoch": 0.6119284127582882,
      "grad_norm": 0.14003841077332135,
      "kl": 2.5995941162109375,
      "learning_rate": 4.506172839506173e-07,
      "loss": 0.0026,
      "reward": 1.7392857894301414,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7482143137603998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3650
    },
    {
      "completion_length": 259.9642972946167,
      "epoch": 0.6122637159981559,
      "grad_norm": 0.4061581313184363,
      "kl": 0.16374969482421875,
      "learning_rate": 4.5086419753086417e-07,
      "loss": 0.0002,
      "reward": 1.7892857491970062,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7982143089175224,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3652
    },
    {
      "completion_length": 268.370548248291,
      "epoch": 0.6125990192380234,
      "grad_norm": 0.38859381055339315,
      "kl": 1.5375289916992188,
      "learning_rate": 4.511111111111111e-07,
      "loss": 0.0015,
      "reward": 1.7678572162985802,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7767857424914837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3654
    },
    {
      "completion_length": 253.7812623977661,
      "epoch": 0.6129343224778909,
      "grad_norm": 0.22522029722982645,
      "kl": 0.46714019775390625,
      "learning_rate": 4.5135802469135805e-07,
      "loss": 0.0005,
      "reward": 1.7160715013742447,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7294643335044384,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3656
    },
    {
      "completion_length": 257.25001430511475,
      "epoch": 0.6132696257177584,
      "grad_norm": 0.32075938946960136,
      "kl": 0.14533233642578125,
      "learning_rate": 4.516049382716049e-07,
      "loss": 0.0001,
      "reward": 1.812500074505806,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.8258928768336773,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3658
    },
    {
      "completion_length": 262.0535840988159,
      "epoch": 0.6136049289576261,
      "grad_norm": 0.16659359770005583,
      "kl": 0.3695220947265625,
      "learning_rate": 4.5185185185185183e-07,
      "loss": 0.0004,
      "reward": 1.744642935693264,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3660
    },
    {
      "completion_length": 259.53126430511475,
      "epoch": 0.6139402321974936,
      "grad_norm": 0.3132566015456766,
      "kl": 0.03826141357421875,
      "learning_rate": 4.520987654320987e-07,
      "loss": 0.0,
      "reward": 1.7339286357164383,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7473214641213417,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3662
    },
    {
      "completion_length": 256.51787090301514,
      "epoch": 0.6142755354373611,
      "grad_norm": 0.2757417864081997,
      "kl": 0.06600189208984375,
      "learning_rate": 4.5234567901234566e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 3664
    },
    {
      "completion_length": 245.51340198516846,
      "epoch": 0.6146108386772288,
      "grad_norm": 0.17796460260570873,
      "kl": 0.2266845703125,
      "learning_rate": 4.525925925925926e-07,
      "loss": 0.0002,
      "reward": 1.751785784959793,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500238418579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3666
    },
    {
      "completion_length": 258.04912281036377,
      "epoch": 0.6149461419170963,
      "grad_norm": 0.3287939137579827,
      "kl": 0.10906219482421875,
      "learning_rate": 4.528395061728395e-07,
      "loss": 0.0001,
      "reward": 1.758928656578064,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7633928898721933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3668
    },
    {
      "completion_length": 248.98661994934082,
      "epoch": 0.6152814451569638,
      "grad_norm": 0.12255448020441631,
      "kl": 0.041259765625,
      "learning_rate": 4.5308641975308643e-07,
      "loss": 0.0,
      "reward": 1.7946429029107094,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7991071753203869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3670
    },
    {
      "completion_length": 260.6919746398926,
      "epoch": 0.6156167483968313,
      "grad_norm": 0.26965476202319116,
      "kl": 0.07506561279296875,
      "learning_rate": 4.5333333333333326e-07,
      "loss": 0.0001,
      "reward": 1.7392857670783997,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.748214315623045,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3672
    },
    {
      "completion_length": 253.81697940826416,
      "epoch": 0.615952051636699,
      "grad_norm": 0.22782034526467285,
      "kl": 0.09051513671875,
      "learning_rate": 4.535802469135802e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7875000275671482,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3674
    },
    {
      "completion_length": 254.91519165039062,
      "epoch": 0.6162873548765665,
      "grad_norm": 0.2151209777437395,
      "kl": 0.23415374755859375,
      "learning_rate": 4.5382716049382715e-07,
      "loss": 0.0002,
      "reward": 1.7642857655882835,
      "reward_std": 0.06060915347188711,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3676
    },
    {
      "completion_length": 257.40626335144043,
      "epoch": 0.616622658116434,
      "grad_norm": 0.16959510394683283,
      "kl": 0.04297637939453125,
      "learning_rate": 4.5407407407407403e-07,
      "loss": 0.0,
      "reward": 1.7196429371833801,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7241071723401546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3678
    },
    {
      "completion_length": 264.46876335144043,
      "epoch": 0.6169579613563017,
      "grad_norm": 0.4565196614990558,
      "kl": 0.9423294067382812,
      "learning_rate": 4.54320987654321e-07,
      "loss": 0.0009,
      "reward": 1.6928572207689285,
      "reward_std": 0.09091372694820166,
      "rewards/equation_reward_func": 0.7196428887546062,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 3680
    },
    {
      "completion_length": 249.86161708831787,
      "epoch": 0.6172932645961692,
      "grad_norm": 0.20489827746347586,
      "kl": 0.0494842529296875,
      "learning_rate": 4.545679012345679e-07,
      "loss": 0.0,
      "reward": 1.728571504354477,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714522004128,
      "rewards/format_reward_func": 1.0,
      "step": 3682
    },
    {
      "completion_length": 260.1294765472412,
      "epoch": 0.6176285678360367,
      "grad_norm": 0.8204189816885677,
      "kl": 1.1546859741210938,
      "learning_rate": 4.548148148148148e-07,
      "loss": 0.0012,
      "reward": 1.7500000596046448,
      "reward_std": 0.07071067858487368,
      "rewards/equation_reward_func": 0.767857164144516,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3684
    },
    {
      "completion_length": 250.92858219146729,
      "epoch": 0.6179638710759042,
      "grad_norm": 0.3160806438632435,
      "kl": 0.0600128173828125,
      "learning_rate": 4.550617283950617e-07,
      "loss": 0.0001,
      "reward": 1.7160715088248253,
      "reward_std": 0.0883883461356163,
      "rewards/equation_reward_func": 0.7294643092900515,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3686
    },
    {
      "completion_length": 257.26786708831787,
      "epoch": 0.6182991743157719,
      "grad_norm": 0.36269127757830905,
      "kl": 0.04349517822265625,
      "learning_rate": 4.553086419753086e-07,
      "loss": 0.0,
      "reward": 1.7750000655651093,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 3688
    },
    {
      "completion_length": 253.61608219146729,
      "epoch": 0.6186344775556394,
      "grad_norm": 0.5216867037004947,
      "kl": 1.238555908203125,
      "learning_rate": 4.555555555555555e-07,
      "loss": 0.0012,
      "reward": 1.7321429252624512,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.741071468219161,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3690
    },
    {
      "completion_length": 259.3839387893677,
      "epoch": 0.6189697807955069,
      "grad_norm": 0.8136101314294178,
      "kl": 3.5810470581054688,
      "learning_rate": 4.5580246913580246e-07,
      "loss": 0.0036,
      "reward": 1.7357143834233284,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7446428909897804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3692
    },
    {
      "completion_length": 260.73215675354004,
      "epoch": 0.6193050840353745,
      "grad_norm": 0.2764728454179333,
      "kl": 0.43869781494140625,
      "learning_rate": 4.5604938271604935e-07,
      "loss": 0.0004,
      "reward": 1.7339286506175995,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7383928913623095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3694
    },
    {
      "completion_length": 255.40625953674316,
      "epoch": 0.6196403872752421,
      "grad_norm": 0.27911070506951186,
      "kl": 0.2676239013671875,
      "learning_rate": 4.562962962962963e-07,
      "loss": 0.0003,
      "reward": 1.748214341700077,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3696
    },
    {
      "completion_length": 243.78572368621826,
      "epoch": 0.6199756905151096,
      "grad_norm": 0.23063148507154832,
      "kl": 0.04782867431640625,
      "learning_rate": 4.5654320987654324e-07,
      "loss": 0.0,
      "reward": 1.7517857998609543,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500275671482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3698
    },
    {
      "completion_length": 249.63394260406494,
      "epoch": 0.6203109937549771,
      "grad_norm": 0.250609945184822,
      "kl": 0.130401611328125,
      "learning_rate": 4.5679012345679007e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7696428932249546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3700
    },
    {
      "completion_length": 252.46876430511475,
      "epoch": 0.6206462969948447,
      "grad_norm": 0.3030883228478924,
      "kl": 0.6184768676757812,
      "learning_rate": 4.57037037037037e-07,
      "loss": 0.0006,
      "reward": 1.7785714864730835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 1.0,
      "step": 3702
    },
    {
      "completion_length": 245.9821538925171,
      "epoch": 0.6209816002347123,
      "grad_norm": 0.22697660760153765,
      "kl": 0.03972625732421875,
      "learning_rate": 4.572839506172839e-07,
      "loss": 0.0,
      "reward": 1.7428572177886963,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571693599224,
      "rewards/format_reward_func": 1.0,
      "step": 3704
    },
    {
      "completion_length": 246.85268878936768,
      "epoch": 0.6213169034745798,
      "grad_norm": 0.692975794353227,
      "kl": 1.3213653564453125,
      "learning_rate": 4.5753086419753084e-07,
      "loss": 0.0013,
      "reward": 1.7714286148548126,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7803571857511997,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3706
    },
    {
      "completion_length": 246.72768878936768,
      "epoch": 0.6216522067144474,
      "grad_norm": 0.2097287152304815,
      "kl": 0.0494232177734375,
      "learning_rate": 4.577777777777778e-07,
      "loss": 0.0,
      "reward": 1.7160715162754059,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7205357514321804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3708
    },
    {
      "completion_length": 252.20536708831787,
      "epoch": 0.621987509954315,
      "grad_norm": 0.3316795839811799,
      "kl": 0.046783447265625,
      "learning_rate": 4.5802469135802467e-07,
      "loss": 0.0,
      "reward": 1.6767858117818832,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.6812500283122063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3710
    },
    {
      "completion_length": 247.56251049041748,
      "epoch": 0.6223228131941825,
      "grad_norm": 0.1709766239519439,
      "kl": 0.240081787109375,
      "learning_rate": 4.582716049382716e-07,
      "loss": 0.0002,
      "reward": 1.7843750640749931,
      "reward_std": 0.02209708606824279,
      "rewards/equation_reward_func": 0.7857143059372902,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 3712
    },
    {
      "completion_length": 255.37947750091553,
      "epoch": 0.62265811643405,
      "grad_norm": 0.24275119545561563,
      "kl": 0.20494842529296875,
      "learning_rate": 4.5851851851851845e-07,
      "loss": 0.0002,
      "reward": 1.782142922282219,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7910714596509933,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3714
    },
    {
      "completion_length": 257.3348331451416,
      "epoch": 0.6229934196739176,
      "grad_norm": 0.1281208930031389,
      "kl": 0.1625823974609375,
      "learning_rate": 4.587654320987654e-07,
      "loss": 0.0002,
      "reward": 1.7625000551342964,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669642996042967,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3716
    },
    {
      "completion_length": 257.07590103149414,
      "epoch": 0.6233287229137852,
      "grad_norm": 0.30500522924730594,
      "kl": 0.0452117919921875,
      "learning_rate": 4.5901234567901233e-07,
      "loss": 0.0,
      "reward": 1.7750000655651093,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 3718
    },
    {
      "completion_length": 263.6384048461914,
      "epoch": 0.6236640261536527,
      "grad_norm": 0.21559576261696073,
      "kl": 0.29811859130859375,
      "learning_rate": 4.592592592592592e-07,
      "loss": 0.0003,
      "reward": 1.7428572177886963,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 3720
    },
    {
      "completion_length": 250.7053689956665,
      "epoch": 0.6239993293935203,
      "grad_norm": 0.2677336786068505,
      "kl": 0.135406494140625,
      "learning_rate": 4.5950617283950616e-07,
      "loss": 0.0001,
      "reward": 1.7125000581145287,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.7258928865194321,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3722
    },
    {
      "completion_length": 254.31697750091553,
      "epoch": 0.6243346326333878,
      "grad_norm": 0.20902682284429114,
      "kl": 0.03919219970703125,
      "learning_rate": 4.597530864197531e-07,
      "loss": 0.0,
      "reward": 1.7375000789761543,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7508928887546062,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3724
    },
    {
      "completion_length": 257.21876335144043,
      "epoch": 0.6246699358732554,
      "grad_norm": 0.31779276105932713,
      "kl": 0.034637451171875,
      "learning_rate": 4.6e-07,
      "loss": 0.0,
      "reward": 1.701785795390606,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7062500342726707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3726
    },
    {
      "completion_length": 262.53572368621826,
      "epoch": 0.6250052391131229,
      "grad_norm": 0.2653994448875676,
      "kl": 0.065643310546875,
      "learning_rate": 4.6024691358024693e-07,
      "loss": 0.0001,
      "reward": 1.78035718947649,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7848214693367481,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3728
    },
    {
      "completion_length": 254.70536994934082,
      "epoch": 0.6253405423529905,
      "grad_norm": 0.19647143960852814,
      "kl": 0.039215087890625,
      "learning_rate": 4.6049382716049377e-07,
      "loss": 0.0,
      "reward": 1.7392858117818832,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 1.0,
      "step": 3730
    },
    {
      "completion_length": 252.8794755935669,
      "epoch": 0.625675845592858,
      "grad_norm": 0.2663136311118988,
      "kl": 0.04831695556640625,
      "learning_rate": 4.607407407407407e-07,
      "loss": 0.0,
      "reward": 1.791071467101574,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7955357320606709,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3732
    },
    {
      "completion_length": 261.4866180419922,
      "epoch": 0.6260111488327256,
      "grad_norm": 0.5976991428641791,
      "kl": 0.09447479248046875,
      "learning_rate": 4.6098765432098765e-07,
      "loss": 0.0001,
      "reward": 1.6808036416769028,
      "reward_std": 0.05745242489501834,
      "rewards/equation_reward_func": 0.6910714711993933,
      "rewards/format_reward_func": 0.9897321499884129,
      "step": 3734
    },
    {
      "completion_length": 249.4062623977661,
      "epoch": 0.6263464520725931,
      "grad_norm": 0.20909877453812703,
      "kl": 0.03856658935546875,
      "learning_rate": 4.6123456790123454e-07,
      "loss": 0.0,
      "reward": 1.7517857551574707,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7562500406056643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3736
    },
    {
      "completion_length": 258.77679920196533,
      "epoch": 0.6266817553124607,
      "grad_norm": 0.20655282498276678,
      "kl": 0.04911041259765625,
      "learning_rate": 4.614814814814815e-07,
      "loss": 0.0,
      "reward": 1.7875000685453415,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7919643148779869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3738
    },
    {
      "completion_length": 247.31697463989258,
      "epoch": 0.6270170585523283,
      "grad_norm": 0.2317755787603404,
      "kl": 0.07225799560546875,
      "learning_rate": 4.6172839506172837e-07,
      "loss": 0.0001,
      "reward": 1.7767857685685158,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7812500223517418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3740
    },
    {
      "completion_length": 241.24554824829102,
      "epoch": 0.6273523617921958,
      "grad_norm": 0.24892404972568108,
      "kl": 0.21956634521484375,
      "learning_rate": 4.619753086419753e-07,
      "loss": 0.0002,
      "reward": 1.750000074505806,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7589286044239998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3742
    },
    {
      "completion_length": 250.0357265472412,
      "epoch": 0.6276876650320634,
      "grad_norm": 0.2655812732823818,
      "kl": 0.07500457763671875,
      "learning_rate": 4.622222222222222e-07,
      "loss": 0.0001,
      "reward": 1.7517857626080513,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7562500387430191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3744
    },
    {
      "completion_length": 255.49108505249023,
      "epoch": 0.6280229682719309,
      "grad_norm": 0.37752015835330077,
      "kl": 0.08672332763671875,
      "learning_rate": 4.624691358024691e-07,
      "loss": 0.0001,
      "reward": 1.7910714894533157,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7955357432365417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3746
    },
    {
      "completion_length": 242.4598331451416,
      "epoch": 0.6283582715117985,
      "grad_norm": 0.22297982995630966,
      "kl": 0.07720947265625,
      "learning_rate": 4.62716049382716e-07,
      "loss": 0.0001,
      "reward": 1.7196429297327995,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7241071574389935,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3748
    },
    {
      "completion_length": 247.51340579986572,
      "epoch": 0.628693574751666,
      "grad_norm": 0.2722172443493719,
      "kl": 0.3848876953125,
      "learning_rate": 4.6296296296296297e-07,
      "loss": 0.0004,
      "reward": 1.7553572207689285,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7687500268220901,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3750
    },
    {
      "completion_length": 244.4509048461914,
      "epoch": 0.6290288779915336,
      "grad_norm": 0.23104650402682889,
      "kl": 0.04361724853515625,
      "learning_rate": 4.6320987654320986e-07,
      "loss": 0.0,
      "reward": 1.769642911851406,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3752
    },
    {
      "completion_length": 248.19197845458984,
      "epoch": 0.6293641812314011,
      "grad_norm": 0.2429922187286504,
      "kl": 0.41692352294921875,
      "learning_rate": 4.634567901234568e-07,
      "loss": 0.0004,
      "reward": 1.7375000566244125,
      "reward_std": 0.10859139543026686,
      "rewards/equation_reward_func": 0.7508928887546062,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3754
    },
    {
      "completion_length": 243.48661994934082,
      "epoch": 0.6296994844712687,
      "grad_norm": 0.5067240377764414,
      "kl": 1.2078857421875,
      "learning_rate": 4.637037037037037e-07,
      "loss": 0.0012,
      "reward": 1.8008929044008255,
      "reward_std": 0.049244935158640146,
      "rewards/equation_reward_func": 0.8026785850524902,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 3756
    },
    {
      "completion_length": 254.66965770721436,
      "epoch": 0.6300347877111363,
      "grad_norm": 0.20975011767382296,
      "kl": 0.07564544677734375,
      "learning_rate": 4.639506172839506e-07,
      "loss": 0.0001,
      "reward": 1.7571429163217545,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7660714611411095,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3758
    },
    {
      "completion_length": 262.8035840988159,
      "epoch": 0.6303700909510038,
      "grad_norm": 0.9492404850582921,
      "kl": 3.7341232299804688,
      "learning_rate": 4.641975308641975e-07,
      "loss": 0.0037,
      "reward": 1.7589286118745804,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.7812500186264515,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 3760
    },
    {
      "completion_length": 273.1250123977661,
      "epoch": 0.6307053941908713,
      "grad_norm": 0.4768543484406165,
      "kl": 0.6936798095703125,
      "learning_rate": 4.644444444444444e-07,
      "loss": 0.0007,
      "reward": 1.6446429416537285,
      "reward_std": 0.13889597728848457,
      "rewards/equation_reward_func": 0.6937500238418579,
      "rewards/format_reward_func": 0.9508928805589676,
      "step": 3762
    },
    {
      "completion_length": 253.3169765472412,
      "epoch": 0.6310406974307389,
      "grad_norm": 0.545748958630044,
      "kl": 0.9039535522460938,
      "learning_rate": 4.6469135802469134e-07,
      "loss": 0.0009,
      "reward": 1.6625000536441803,
      "reward_std": 0.09343910869210958,
      "rewards/equation_reward_func": 0.7026786003261805,
      "rewards/format_reward_func": 0.9598214477300644,
      "step": 3764
    },
    {
      "completion_length": 262.6830463409424,
      "epoch": 0.6313760006706065,
      "grad_norm": 0.39882267008820893,
      "kl": 0.4140472412109375,
      "learning_rate": 4.6493827160493823e-07,
      "loss": 0.0004,
      "reward": 1.708928644657135,
      "reward_std": 0.12879444938153028,
      "rewards/equation_reward_func": 0.7491071783006191,
      "rewards/format_reward_func": 0.9598214477300644,
      "step": 3766
    },
    {
      "completion_length": 275.0089416503906,
      "epoch": 0.631711303910474,
      "grad_norm": 0.4620834014351106,
      "kl": 0.78497314453125,
      "learning_rate": 4.651851851851852e-07,
      "loss": 0.0008,
      "reward": 1.6575893387198448,
      "reward_std": 0.20139916171319783,
      "rewards/equation_reward_func": 0.7080357428640127,
      "rewards/format_reward_func": 0.9495535977184772,
      "step": 3768
    },
    {
      "completion_length": 261.3437623977661,
      "epoch": 0.6320466071503416,
      "grad_norm": 0.40476944953139327,
      "kl": 2.376129150390625,
      "learning_rate": 4.654320987654321e-07,
      "loss": 0.0024,
      "reward": 1.6946428939700127,
      "reward_std": 0.10859139915555716,
      "rewards/equation_reward_func": 0.734821442514658,
      "rewards/format_reward_func": 0.9598214477300644,
      "step": 3770
    },
    {
      "completion_length": 248.98215579986572,
      "epoch": 0.6323819103902092,
      "grad_norm": 0.18001335211723607,
      "kl": 1.412933349609375,
      "learning_rate": 4.6567901234567895e-07,
      "loss": 0.0014,
      "reward": 1.7330357730388641,
      "reward_std": 0.08460027631372213,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9803571552038193,
      "step": 3772
    },
    {
      "completion_length": 263.4330463409424,
      "epoch": 0.6327172136300767,
      "grad_norm": 0.3190957941370173,
      "kl": 0.8990097045898438,
      "learning_rate": 4.659259259259259e-07,
      "loss": 0.0009,
      "reward": 1.68392863124609,
      "reward_std": 0.0833375845104456,
      "rewards/equation_reward_func": 0.706250037997961,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 3774
    },
    {
      "completion_length": 254.8794755935669,
      "epoch": 0.6330525168699442,
      "grad_norm": 0.31257213744887996,
      "kl": 0.89825439453125,
      "learning_rate": 4.6617283950617283e-07,
      "loss": 0.0009,
      "reward": 1.7276786416769028,
      "reward_std": 0.07197337062098086,
      "rewards/equation_reward_func": 0.7437500357627869,
      "rewards/format_reward_func": 0.9839285798370838,
      "step": 3776
    },
    {
      "completion_length": 258.2276916503906,
      "epoch": 0.6333878201098118,
      "grad_norm": 0.7055444014800766,
      "kl": 8.225692749023438,
      "learning_rate": 4.664197530864197e-07,
      "loss": 0.0082,
      "reward": 1.6964286342263222,
      "reward_std": 0.10606601648032665,
      "rewards/equation_reward_func": 0.7232143171131611,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 3778
    },
    {
      "completion_length": 265.3482275009155,
      "epoch": 0.6337231233496794,
      "grad_norm": 0.9790034779343154,
      "kl": 4.100105285644531,
      "learning_rate": 4.6666666666666666e-07,
      "loss": 0.0041,
      "reward": 1.657142959535122,
      "reward_std": 0.1616244027391076,
      "rewards/equation_reward_func": 0.6928571723401546,
      "rewards/format_reward_func": 0.9642857313156128,
      "step": 3780
    },
    {
      "completion_length": 256.03572845458984,
      "epoch": 0.6340584265895469,
      "grad_norm": 0.29187955878990074,
      "kl": 0.35442352294921875,
      "learning_rate": 4.6691358024691355e-07,
      "loss": 0.0004,
      "reward": 1.6660715118050575,
      "reward_std": 0.10859139915555716,
      "rewards/equation_reward_func": 0.6883928924798965,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 3782
    },
    {
      "completion_length": 252.56697750091553,
      "epoch": 0.6343937298294144,
      "grad_norm": 0.41425872484802523,
      "kl": 0.2930908203125,
      "learning_rate": 4.671604938271605e-07,
      "loss": 0.0003,
      "reward": 1.7535714730620384,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7625000216066837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3784
    },
    {
      "completion_length": 244.29018878936768,
      "epoch": 0.6347290330692821,
      "grad_norm": 0.4734383263821688,
      "kl": 0.1266326904296875,
      "learning_rate": 4.674074074074074e-07,
      "loss": 0.0001,
      "reward": 1.6875000819563866,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.6919643208384514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3786
    },
    {
      "completion_length": 255.27679634094238,
      "epoch": 0.6350643363091496,
      "grad_norm": 0.5475088235706634,
      "kl": 0.07224273681640625,
      "learning_rate": 4.6765432098765427e-07,
      "loss": 0.0001,
      "reward": 1.657142959535122,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.6741071790456772,
      "rewards/format_reward_func": 0.9830357283353806,
      "step": 3788
    },
    {
      "completion_length": 241.01340293884277,
      "epoch": 0.6353996395490171,
      "grad_norm": 0.20196700802888495,
      "kl": 0.046722412109375,
      "learning_rate": 4.679012345679012e-07,
      "loss": 0.0,
      "reward": 1.7357143759727478,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 3790
    },
    {
      "completion_length": 260.00893783569336,
      "epoch": 0.6357349427888846,
      "grad_norm": 0.6002225994492422,
      "kl": 0.1472930908203125,
      "learning_rate": 4.681481481481481e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7446428835391998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3792
    },
    {
      "completion_length": 253.6026906967163,
      "epoch": 0.6360702460287523,
      "grad_norm": 0.23931617298852403,
      "kl": 0.0593109130859375,
      "learning_rate": 4.6839506172839504e-07,
      "loss": 0.0001,
      "reward": 1.741071492433548,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.745535746216774,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3794
    },
    {
      "completion_length": 250.92411708831787,
      "epoch": 0.6364055492686198,
      "grad_norm": 0.2321628612327699,
      "kl": 0.057647705078125,
      "learning_rate": 4.68641975308642e-07,
      "loss": 0.0001,
      "reward": 1.7214286401867867,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7303571663796902,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3796
    },
    {
      "completion_length": 253.66965103149414,
      "epoch": 0.6367408525084873,
      "grad_norm": 0.34411481195861043,
      "kl": 0.0452880859375,
      "learning_rate": 4.6888888888888887e-07,
      "loss": 0.0,
      "reward": 1.7250000685453415,
      "reward_std": 0.08586296439170837,
      "rewards/equation_reward_func": 0.7339286170899868,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3798
    },
    {
      "completion_length": 249.4017972946167,
      "epoch": 0.637076155748355,
      "grad_norm": 0.3156697298150423,
      "kl": 0.0444183349609375,
      "learning_rate": 4.6913580246913576e-07,
      "loss": 0.0,
      "reward": 1.8125000596046448,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.8169643133878708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3800
    },
    {
      "completion_length": 251.35268688201904,
      "epoch": 0.6374114589882225,
      "grad_norm": 0.39588853568629256,
      "kl": 0.0497589111328125,
      "learning_rate": 4.693827160493827e-07,
      "loss": 0.0,
      "reward": 1.7366072237491608,
      "reward_std": 0.06944798771291971,
      "rewards/equation_reward_func": 0.7473214641213417,
      "rewards/format_reward_func": 0.9892857223749161,
      "step": 3802
    },
    {
      "completion_length": 269.97769260406494,
      "epoch": 0.63774676222809,
      "grad_norm": 0.2850366769440957,
      "kl": 0.05326080322265625,
      "learning_rate": 4.696296296296296e-07,
      "loss": 0.0001,
      "reward": 1.692857250571251,
      "reward_std": 0.11111677810549736,
      "rewards/equation_reward_func": 0.7107143178582191,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3804
    },
    {
      "completion_length": 244.63840293884277,
      "epoch": 0.6380820654679575,
      "grad_norm": 0.27088312156428124,
      "kl": 0.0478973388671875,
      "learning_rate": 4.6987654320987653e-07,
      "loss": 0.0,
      "reward": 1.7982143461704254,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8026785925030708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3806
    },
    {
      "completion_length": 255.5223331451416,
      "epoch": 0.6384173687078252,
      "grad_norm": 0.15589935664162494,
      "kl": 0.05022430419921875,
      "learning_rate": 4.701234567901234e-07,
      "loss": 0.0001,
      "reward": 1.6910715103149414,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.6955357454717159,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3808
    },
    {
      "completion_length": 260.58483123779297,
      "epoch": 0.6387526719476927,
      "grad_norm": 0.46828900118421973,
      "kl": 0.0674285888671875,
      "learning_rate": 4.7037037037037036e-07,
      "loss": 0.0001,
      "reward": 1.7767857760190964,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.7991071566939354,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 3810
    },
    {
      "completion_length": 243.7053689956665,
      "epoch": 0.6390879751875602,
      "grad_norm": 0.1797720332442718,
      "kl": 0.05417633056640625,
      "learning_rate": 4.706172839506173e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 3812
    },
    {
      "completion_length": 252.46429634094238,
      "epoch": 0.6394232784274279,
      "grad_norm": 1.0245365579568664,
      "kl": 0.049041748046875,
      "learning_rate": 4.7086419753086414e-07,
      "loss": 0.0,
      "reward": 1.7839286252856255,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7883928939700127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3814
    },
    {
      "completion_length": 248.0803680419922,
      "epoch": 0.6397585816672954,
      "grad_norm": 0.1690612877408516,
      "kl": 0.04561614990234375,
      "learning_rate": 4.711111111111111e-07,
      "loss": 0.0,
      "reward": 1.8089286237955093,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.813392885029316,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3816
    },
    {
      "completion_length": 242.28125953674316,
      "epoch": 0.6400938849071629,
      "grad_norm": 0.1799869429711767,
      "kl": 0.04736328125,
      "learning_rate": 4.7135802469135797e-07,
      "loss": 0.0,
      "reward": 1.8000000566244125,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.8089286014437675,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3818
    },
    {
      "completion_length": 247.89286994934082,
      "epoch": 0.6404291881470304,
      "grad_norm": 0.2382277237406641,
      "kl": 0.04730224609375,
      "learning_rate": 4.716049382716049e-07,
      "loss": 0.0,
      "reward": 1.7089286744594574,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7133928835391998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3820
    },
    {
      "completion_length": 253.94643878936768,
      "epoch": 0.6407644913868981,
      "grad_norm": 0.28596620542102036,
      "kl": 0.0505523681640625,
      "learning_rate": 4.7185185185185185e-07,
      "loss": 0.0001,
      "reward": 1.7571429163217545,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7660714574158192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3822
    },
    {
      "completion_length": 247.9732255935669,
      "epoch": 0.6410997946267656,
      "grad_norm": 0.11720032512746174,
      "kl": 0.05326080322265625,
      "learning_rate": 4.7209876543209874e-07,
      "loss": 0.0001,
      "reward": 1.744642935693264,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7580357398837805,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3824
    },
    {
      "completion_length": 263.95090675354004,
      "epoch": 0.6414350978666331,
      "grad_norm": 0.17071112444417788,
      "kl": 0.06207275390625,
      "learning_rate": 4.723456790123457e-07,
      "loss": 0.0001,
      "reward": 1.7303572073578835,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7437500283122063,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3826
    },
    {
      "completion_length": 259.48661613464355,
      "epoch": 0.6417704011065007,
      "grad_norm": 0.2601322695367908,
      "kl": 0.06282806396484375,
      "learning_rate": 4.725925925925926e-07,
      "loss": 0.0001,
      "reward": 1.7633929252624512,
      "reward_std": 0.10227794293314219,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9750000089406967,
      "step": 3828
    },
    {
      "completion_length": 258.39733600616455,
      "epoch": 0.6421057043463683,
      "grad_norm": 0.22779481358853262,
      "kl": 0.04955291748046875,
      "learning_rate": 4.7283950617283945e-07,
      "loss": 0.0,
      "reward": 1.7535714879631996,
      "reward_std": 0.07576144207268953,
      "rewards/equation_reward_func": 0.771428594365716,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3830
    },
    {
      "completion_length": 250.21876430511475,
      "epoch": 0.6424410075862358,
      "grad_norm": 0.43527318397332027,
      "kl": 0.05060577392578125,
      "learning_rate": 4.730864197530864e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3832
    },
    {
      "completion_length": 262.4598340988159,
      "epoch": 0.6427763108261033,
      "grad_norm": 0.21911360098756538,
      "kl": 0.06902313232421875,
      "learning_rate": 4.733333333333333e-07,
      "loss": 0.0001,
      "reward": 1.6392858177423477,
      "reward_std": 0.09596449136734009,
      "rewards/equation_reward_func": 0.6660714633762836,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 3834
    },
    {
      "completion_length": 273.68304920196533,
      "epoch": 0.643111614065971,
      "grad_norm": 1.615608755048004,
      "kl": 0.154632568359375,
      "learning_rate": 4.735802469135802e-07,
      "loss": 0.0002,
      "reward": 1.6790179163217545,
      "reward_std": 0.12058696104213595,
      "rewards/equation_reward_func": 0.7294643055647612,
      "rewards/format_reward_func": 0.949553593993187,
      "step": 3836
    },
    {
      "completion_length": 264.53572273254395,
      "epoch": 0.6434469173058385,
      "grad_norm": 0.2778056266442193,
      "kl": 0.07929229736328125,
      "learning_rate": 4.7382716049382717e-07,
      "loss": 0.0001,
      "reward": 1.680357240140438,
      "reward_std": 0.13384521193802357,
      "rewards/equation_reward_func": 0.7294643186032772,
      "rewards/format_reward_func": 0.9508928768336773,
      "step": 3838
    },
    {
      "completion_length": 267.4062623977661,
      "epoch": 0.643782220545706,
      "grad_norm": 0.18156955523379578,
      "kl": 0.06781005859375,
      "learning_rate": 4.7407407407407405e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.08586296625435352,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 0.9642857275903225,
      "step": 3840
    },
    {
      "completion_length": 260.8839406967163,
      "epoch": 0.6441175237855736,
      "grad_norm": 0.31977780335247097,
      "kl": 0.153656005859375,
      "learning_rate": 4.74320987654321e-07,
      "loss": 0.0002,
      "reward": 1.7285714894533157,
      "reward_std": 0.07071068044751883,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3842
    },
    {
      "completion_length": 261.85269355773926,
      "epoch": 0.6444528270254412,
      "grad_norm": 0.3417284451794752,
      "kl": 0.09317779541015625,
      "learning_rate": 4.745679012345679e-07,
      "loss": 0.0001,
      "reward": 1.7500000596046448,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7678571790456772,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 3844
    },
    {
      "completion_length": 246.83036994934082,
      "epoch": 0.6447881302653087,
      "grad_norm": 0.17477561853506282,
      "kl": 0.06887054443359375,
      "learning_rate": 4.7481481481481477e-07,
      "loss": 0.0001,
      "reward": 1.7625000327825546,
      "reward_std": 0.0833375845104456,
      "rewards/equation_reward_func": 0.7758928872644901,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 3846
    },
    {
      "completion_length": 255.99554634094238,
      "epoch": 0.6451234335051762,
      "grad_norm": 0.44253711066314283,
      "kl": 0.1705474853515625,
      "learning_rate": 4.750617283950617e-07,
      "loss": 0.0002,
      "reward": 1.7821429073810577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428794413805,
      "rewards/format_reward_func": 1.0,
      "step": 3848
    },
    {
      "completion_length": 248.18750953674316,
      "epoch": 0.6454587367450438,
      "grad_norm": 0.2387987965129388,
      "kl": 0.067352294921875,
      "learning_rate": 4.753086419753086e-07,
      "loss": 0.0001,
      "reward": 1.7642858028411865,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7732143215835094,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3850
    },
    {
      "completion_length": 251.83483409881592,
      "epoch": 0.6457940399849114,
      "grad_norm": 0.18727904381625438,
      "kl": 0.0418701171875,
      "learning_rate": 4.7555555555555554e-07,
      "loss": 0.0,
      "reward": 1.7571429312229156,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7660714574158192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3852
    },
    {
      "completion_length": 253.39733219146729,
      "epoch": 0.6461293432247789,
      "grad_norm": 0.18840059404995516,
      "kl": 0.23563385009765625,
      "learning_rate": 4.758024691358025e-07,
      "loss": 0.0002,
      "reward": 1.7321429252624512,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428973227739,
      "rewards/format_reward_func": 1.0,
      "step": 3854
    },
    {
      "completion_length": 259.3571557998657,
      "epoch": 0.6464646464646465,
      "grad_norm": 0.24891815306132903,
      "kl": 0.1553192138671875,
      "learning_rate": 4.7604938271604937e-07,
      "loss": 0.0002,
      "reward": 1.7517857551574707,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7562500536441803,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3856
    },
    {
      "completion_length": 249.96875953674316,
      "epoch": 0.646799949704514,
      "grad_norm": 0.1939251343094505,
      "kl": 0.10715484619140625,
      "learning_rate": 4.7629629629629626e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.73214291036129,
      "rewards/format_reward_func": 1.0,
      "step": 3858
    },
    {
      "completion_length": 264.3616189956665,
      "epoch": 0.6471352529443816,
      "grad_norm": 0.31045812888398727,
      "kl": 0.240264892578125,
      "learning_rate": 4.7654320987654315e-07,
      "loss": 0.0002,
      "reward": 1.7089286521077156,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7133929040282965,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3860
    },
    {
      "completion_length": 248.1741189956665,
      "epoch": 0.6474705561842491,
      "grad_norm": 0.29967271897328407,
      "kl": 0.142913818359375,
      "learning_rate": 4.767901234567901e-07,
      "loss": 0.0001,
      "reward": 1.7553572058677673,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.759821455925703,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3862
    },
    {
      "completion_length": 244.29465293884277,
      "epoch": 0.6478058594241167,
      "grad_norm": 0.34126496872835815,
      "kl": 0.0392608642578125,
      "learning_rate": 4.77037037037037e-07,
      "loss": 0.0,
      "reward": 1.7178572192788124,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7178571820259094,
      "rewards/format_reward_func": 1.0,
      "step": 3864
    },
    {
      "completion_length": 253.07144165039062,
      "epoch": 0.6481411626639842,
      "grad_norm": 5.56472514014859,
      "kl": 5.6567535400390625,
      "learning_rate": 4.772839506172839e-07,
      "loss": 0.0057,
      "reward": 1.775000050663948,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.783928606659174,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3866
    },
    {
      "completion_length": 260.7812614440918,
      "epoch": 0.6484764659038518,
      "grad_norm": 0.2560663927017658,
      "kl": 0.22957611083984375,
      "learning_rate": 4.775308641975309e-07,
      "loss": 0.0002,
      "reward": 1.7428571954369545,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7517857402563095,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3868
    },
    {
      "completion_length": 254.96876335144043,
      "epoch": 0.6488117691437193,
      "grad_norm": 0.21644056151909313,
      "kl": 0.15191650390625,
      "learning_rate": 4.777777777777778e-07,
      "loss": 0.0002,
      "reward": 1.764285795390606,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857357859612,
      "rewards/format_reward_func": 1.0,
      "step": 3870
    },
    {
      "completion_length": 254.66072463989258,
      "epoch": 0.6491470723835869,
      "grad_norm": 0.346480276669036,
      "kl": 0.04920196533203125,
      "learning_rate": 4.780246913580246e-07,
      "loss": 0.0,
      "reward": 1.748214341700077,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3872
    },
    {
      "completion_length": 267.8482265472412,
      "epoch": 0.6494823756234545,
      "grad_norm": 0.29899998204476824,
      "kl": 0.03923797607421875,
      "learning_rate": 4.782716049382716e-07,
      "loss": 0.0,
      "reward": 1.7321429550647736,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7321428917348385,
      "rewards/format_reward_func": 1.0,
      "step": 3874
    },
    {
      "completion_length": 258.6026906967163,
      "epoch": 0.649817678863322,
      "grad_norm": 0.08752132121753611,
      "kl": 0.0488739013671875,
      "learning_rate": 4.785185185185185e-07,
      "loss": 0.0,
      "reward": 1.7267858013510704,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7312500216066837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3876
    },
    {
      "completion_length": 261.4687662124634,
      "epoch": 0.6501529821031896,
      "grad_norm": 0.32540356794934916,
      "kl": 0.0558624267578125,
      "learning_rate": 4.787654320987654e-07,
      "loss": 0.0001,
      "reward": 1.7428572326898575,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7517857421189547,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3878
    },
    {
      "completion_length": 239.008939743042,
      "epoch": 0.6504882853430571,
      "grad_norm": 0.23190866512432148,
      "kl": 0.072418212890625,
      "learning_rate": 4.790123456790123e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 3880
    },
    {
      "completion_length": 254.6875123977661,
      "epoch": 0.6508235885829247,
      "grad_norm": 0.6270072702489933,
      "kl": 0.17505645751953125,
      "learning_rate": 4.792592592592592e-07,
      "loss": 0.0002,
      "reward": 1.726785808801651,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.731250025331974,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3882
    },
    {
      "completion_length": 260.0134048461914,
      "epoch": 0.6511588918227922,
      "grad_norm": 0.23875093940047093,
      "kl": 0.08642578125,
      "learning_rate": 4.795061728395062e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7553571797907352,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3884
    },
    {
      "completion_length": 245.8928689956665,
      "epoch": 0.6514941950626598,
      "grad_norm": 0.15408522982548276,
      "kl": 0.04544830322265625,
      "learning_rate": 4.79753086419753e-07,
      "loss": 0.0,
      "reward": 1.7160715237259865,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7205357477068901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3886
    },
    {
      "completion_length": 260.94644355773926,
      "epoch": 0.6518294983025273,
      "grad_norm": 0.1560889557012829,
      "kl": 0.0451812744140625,
      "learning_rate": 4.8e-07,
      "loss": 0.0,
      "reward": 1.7142857760190964,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857443541288,
      "rewards/format_reward_func": 1.0,
      "step": 3888
    },
    {
      "completion_length": 250.03125953674316,
      "epoch": 0.6521648015423949,
      "grad_norm": 0.1550896631696827,
      "kl": 0.04175567626953125,
      "learning_rate": 4.802469135802469e-07,
      "loss": 0.0,
      "reward": 1.74821437895298,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7526785992085934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3890
    },
    {
      "completion_length": 244.3660831451416,
      "epoch": 0.6525001047822625,
      "grad_norm": 0.3224570837771038,
      "kl": 0.05951690673828125,
      "learning_rate": 4.804938271604938e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7821428999304771,
      "rewards/format_reward_func": 1.0,
      "step": 3892
    },
    {
      "completion_length": 233.0625114440918,
      "epoch": 0.65283540802213,
      "grad_norm": 0.23181470359460926,
      "kl": 0.04352569580078125,
      "learning_rate": 4.807407407407407e-07,
      "loss": 0.0,
      "reward": 1.7535714879631996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714488476515,
      "rewards/format_reward_func": 1.0,
      "step": 3894
    },
    {
      "completion_length": 247.50447273254395,
      "epoch": 0.6531707112619975,
      "grad_norm": 0.4284565414178115,
      "kl": 0.0639801025390625,
      "learning_rate": 4.809876543209876e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 3896
    },
    {
      "completion_length": 254.9419755935669,
      "epoch": 0.6535060145018651,
      "grad_norm": 0.2229631953694362,
      "kl": 0.09334564208984375,
      "learning_rate": 4.812345679012346e-07,
      "loss": 0.0001,
      "reward": 1.7285714969038963,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.728571455925703,
      "rewards/format_reward_func": 1.0,
      "step": 3898
    },
    {
      "completion_length": 244.6071548461914,
      "epoch": 0.6538413177417327,
      "grad_norm": 0.1938204139192448,
      "kl": 0.06211090087890625,
      "learning_rate": 4.814814814814814e-07,
      "loss": 0.0001,
      "reward": 1.733928643167019,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7383928801864386,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3900
    },
    {
      "completion_length": 257.8392963409424,
      "epoch": 0.6541766209816002,
      "grad_norm": 0.21032308816890635,
      "kl": 0.29449462890625,
      "learning_rate": 4.817283950617283e-07,
      "loss": 0.0003,
      "reward": 1.7571429088711739,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.757142897695303,
      "rewards/format_reward_func": 1.0,
      "step": 3902
    },
    {
      "completion_length": 247.4509038925171,
      "epoch": 0.6545119242214678,
      "grad_norm": 0.3046733976307481,
      "kl": 0.14829254150390625,
      "learning_rate": 4.819753086419753e-07,
      "loss": 0.0001,
      "reward": 1.8089286163449287,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8133928775787354,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3904
    },
    {
      "completion_length": 246.89733695983887,
      "epoch": 0.6548472274613354,
      "grad_norm": 0.39941927413480616,
      "kl": 0.2657928466796875,
      "learning_rate": 4.822222222222222e-07,
      "loss": 0.0003,
      "reward": 1.7285714969038963,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714671015739,
      "rewards/format_reward_func": 1.0,
      "step": 3906
    },
    {
      "completion_length": 251.4241180419922,
      "epoch": 0.6551825307012029,
      "grad_norm": 0.16084090620029584,
      "kl": 0.05809783935546875,
      "learning_rate": 4.824691358024692e-07,
      "loss": 0.0001,
      "reward": 1.755357213318348,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7598214522004128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3908
    },
    {
      "completion_length": 253.85269165039062,
      "epoch": 0.6555178339410704,
      "grad_norm": 0.19861177988293185,
      "kl": 0.0594329833984375,
      "learning_rate": 4.82716049382716e-07,
      "loss": 0.0001,
      "reward": 1.7642858028411865,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 3910
    },
    {
      "completion_length": 246.08929634094238,
      "epoch": 0.655853137180938,
      "grad_norm": 0.24120335029514853,
      "kl": 0.047882080078125,
      "learning_rate": 4.829629629629629e-07,
      "loss": 0.0,
      "reward": 1.707142949104309,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7071428913623095,
      "rewards/format_reward_func": 1.0,
      "step": 3912
    },
    {
      "completion_length": 245.1428689956665,
      "epoch": 0.6561884404208056,
      "grad_norm": 0.30892198013938815,
      "kl": 0.05008697509765625,
      "learning_rate": 4.832098765432099e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 3914
    },
    {
      "completion_length": 256.74554920196533,
      "epoch": 0.6565237436606731,
      "grad_norm": 0.2089904775804934,
      "kl": 0.05450439453125,
      "learning_rate": 4.834567901234567e-07,
      "loss": 0.0001,
      "reward": 1.7750000432133675,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7839286141097546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3916
    },
    {
      "completion_length": 250.4196548461914,
      "epoch": 0.6568590469005406,
      "grad_norm": 0.26592058060680673,
      "kl": 0.0504913330078125,
      "learning_rate": 4.837037037037037e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7607143074274063,
      "rewards/format_reward_func": 1.0,
      "step": 3918
    },
    {
      "completion_length": 249.6785831451416,
      "epoch": 0.6571943501404083,
      "grad_norm": 0.040893189818078074,
      "kl": 0.04575347900390625,
      "learning_rate": 4.839506172839506e-07,
      "loss": 0.0,
      "reward": 1.782142922282219,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7910714522004128,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3920
    },
    {
      "completion_length": 259.4509057998657,
      "epoch": 0.6575296533802758,
      "grad_norm": 0.13055749625194799,
      "kl": 0.18926239013671875,
      "learning_rate": 4.841975308641975e-07,
      "loss": 0.0002,
      "reward": 1.7000000700354576,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7089285999536514,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3922
    },
    {
      "completion_length": 257.4285821914673,
      "epoch": 0.6578649566201433,
      "grad_norm": 0.2842193792194698,
      "kl": 0.5179595947265625,
      "learning_rate": 4.844444444444445e-07,
      "loss": 0.0005,
      "reward": 1.7035714983940125,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.712500037625432,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3924
    },
    {
      "completion_length": 257.8616199493408,
      "epoch": 0.6582002598600109,
      "grad_norm": 0.21372307018495712,
      "kl": 0.04400634765625,
      "learning_rate": 4.846913580246913e-07,
      "loss": 0.0,
      "reward": 1.6982143595814705,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.702678607776761,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3926
    },
    {
      "completion_length": 250.1741189956665,
      "epoch": 0.6585355630998785,
      "grad_norm": 0.2179283235343074,
      "kl": 0.0731353759765625,
      "learning_rate": 4.849382716049383e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 1.0,
      "step": 3928
    },
    {
      "completion_length": 255.47769165039062,
      "epoch": 0.658870866339746,
      "grad_norm": 0.20049233410239328,
      "kl": 0.0641937255859375,
      "learning_rate": 4.851851851851852e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 3930
    },
    {
      "completion_length": 248.6294755935669,
      "epoch": 0.6592061695796135,
      "grad_norm": 0.41126545761337224,
      "kl": 0.062530517578125,
      "learning_rate": 4.85432098765432e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7589285969734192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3932
    },
    {
      "completion_length": 243.2500114440918,
      "epoch": 0.6595414728194812,
      "grad_norm": 0.3831697730735232,
      "kl": 0.14061737060546875,
      "learning_rate": 4.85679012345679e-07,
      "loss": 0.0001,
      "reward": 1.7517857998609543,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500350177288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3934
    },
    {
      "completion_length": 244.65625953674316,
      "epoch": 0.6598767760593487,
      "grad_norm": 0.19433989843152025,
      "kl": 0.05853271484375,
      "learning_rate": 4.859259259259259e-07,
      "loss": 0.0001,
      "reward": 1.723214365541935,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7276786118745804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3936
    },
    {
      "completion_length": 247.46429824829102,
      "epoch": 0.6602120792992162,
      "grad_norm": 0.27797701510862327,
      "kl": 0.07159423828125,
      "learning_rate": 4.861728395061729e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 3938
    },
    {
      "completion_length": 245.87054824829102,
      "epoch": 0.6605473825390837,
      "grad_norm": 0.3248319516246433,
      "kl": 0.05426025390625,
      "learning_rate": 4.864197530864198e-07,
      "loss": 0.0001,
      "reward": 1.7000000923871994,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7000000365078449,
      "rewards/format_reward_func": 1.0,
      "step": 3940
    },
    {
      "completion_length": 256.95536613464355,
      "epoch": 0.6608826857789514,
      "grad_norm": 0.21364917257066787,
      "kl": 0.9769058227539062,
      "learning_rate": 4.866666666666666e-07,
      "loss": 0.001,
      "reward": 1.7607143446803093,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143260538578,
      "rewards/format_reward_func": 1.0,
      "step": 3942
    },
    {
      "completion_length": 242.3303680419922,
      "epoch": 0.6612179890188189,
      "grad_norm": 0.6023822317409888,
      "kl": 1.2734832763671875,
      "learning_rate": 4.869135802469136e-07,
      "loss": 0.0013,
      "reward": 1.7776786386966705,
      "reward_std": 0.08207489224150777,
      "rewards/equation_reward_func": 0.7839285992085934,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 3944
    },
    {
      "completion_length": 247.9821538925171,
      "epoch": 0.6615532922586864,
      "grad_norm": 0.21272371324088035,
      "kl": 0.08065032958984375,
      "learning_rate": 4.871604938271604e-07,
      "loss": 0.0001,
      "reward": 1.8071429133415222,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8071428872644901,
      "rewards/format_reward_func": 1.0,
      "step": 3946
    },
    {
      "completion_length": 255.55804824829102,
      "epoch": 0.661888595498554,
      "grad_norm": 0.2834225170622278,
      "kl": 0.0803070068359375,
      "learning_rate": 4.874074074074073e-07,
      "loss": 0.0001,
      "reward": 1.7767857760190964,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7812500335276127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3948
    },
    {
      "completion_length": 249.44197845458984,
      "epoch": 0.6622238987384216,
      "grad_norm": 0.16218911032112202,
      "kl": 0.06241607666015625,
      "learning_rate": 4.876543209876543e-07,
      "loss": 0.0001,
      "reward": 1.814285770058632,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857439815998,
      "rewards/format_reward_func": 1.0,
      "step": 3950
    },
    {
      "completion_length": 241.6071548461914,
      "epoch": 0.6625592019782891,
      "grad_norm": 0.24746330003861983,
      "kl": 0.06597900390625,
      "learning_rate": 4.879012345679012e-07,
      "loss": 0.0001,
      "reward": 1.8058036267757416,
      "reward_std": 0.03219861118122935,
      "rewards/equation_reward_func": 0.8071428760886192,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 3952
    },
    {
      "completion_length": 256.14733505249023,
      "epoch": 0.6628945052181566,
      "grad_norm": 0.26130082812710603,
      "kl": 0.0624542236328125,
      "learning_rate": 4.881481481481482e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 3954
    },
    {
      "completion_length": 238.16519165039062,
      "epoch": 0.6632298084580243,
      "grad_norm": 0.5290773175802812,
      "kl": 0.268890380859375,
      "learning_rate": 4.883950617283951e-07,
      "loss": 0.0003,
      "reward": 1.7696429267525673,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3956
    },
    {
      "completion_length": 247.38840675354004,
      "epoch": 0.6635651116978918,
      "grad_norm": 0.18398423923323337,
      "kl": 0.072296142578125,
      "learning_rate": 4.88641975308642e-07,
      "loss": 0.0001,
      "reward": 1.7178572192788124,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7178571745753288,
      "rewards/format_reward_func": 1.0,
      "step": 3958
    },
    {
      "completion_length": 238.4776906967163,
      "epoch": 0.6639004149377593,
      "grad_norm": 0.9794345870204398,
      "kl": 0.08960723876953125,
      "learning_rate": 4.888888888888889e-07,
      "loss": 0.0001,
      "reward": 1.742857240140438,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7428571656346321,
      "rewards/format_reward_func": 1.0,
      "step": 3960
    },
    {
      "completion_length": 239.95983409881592,
      "epoch": 0.6642357181776269,
      "grad_norm": 0.27360081177453655,
      "kl": 0.092071533203125,
      "learning_rate": 4.891358024691357e-07,
      "loss": 0.0001,
      "reward": 1.7964286357164383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 3962
    },
    {
      "completion_length": 251.196439743042,
      "epoch": 0.6645710214174945,
      "grad_norm": 0.31778832714402766,
      "kl": 0.05353546142578125,
      "learning_rate": 4.893827160493827e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321428880095482,
      "rewards/format_reward_func": 1.0,
      "step": 3964
    },
    {
      "completion_length": 242.4687623977661,
      "epoch": 0.664906324657362,
      "grad_norm": 0.17589893133387763,
      "kl": 0.10141754150390625,
      "learning_rate": 4.896296296296296e-07,
      "loss": 0.0001,
      "reward": 1.7714286521077156,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 3966
    },
    {
      "completion_length": 233.05357933044434,
      "epoch": 0.6652416278972295,
      "grad_norm": 0.21849168483193168,
      "kl": 0.052154541015625,
      "learning_rate": 4.898765432098765e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571660071611,
      "rewards/format_reward_func": 1.0,
      "step": 3968
    },
    {
      "completion_length": 245.81697273254395,
      "epoch": 0.6655769311370971,
      "grad_norm": 0.17133072243592265,
      "kl": 0.060943603515625,
      "learning_rate": 4.901234567901235e-07,
      "loss": 0.0001,
      "reward": 1.801785759627819,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8062500283122063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3970
    },
    {
      "completion_length": 230.18304920196533,
      "epoch": 0.6659122343769647,
      "grad_norm": 0.28829507057273934,
      "kl": 0.063507080078125,
      "learning_rate": 4.903703703703703e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 3972
    },
    {
      "completion_length": 245.37500953674316,
      "epoch": 0.6662475376168322,
      "grad_norm": 0.38017604329471877,
      "kl": 0.047882080078125,
      "learning_rate": 4.906172839506173e-07,
      "loss": 0.0,
      "reward": 1.773214340209961,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7776786051690578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3974
    },
    {
      "completion_length": 243.52233123779297,
      "epoch": 0.6665828408566998,
      "grad_norm": 0.15516051295757663,
      "kl": 0.05037689208984375,
      "learning_rate": 4.908641975308642e-07,
      "loss": 0.0001,
      "reward": 1.716071531176567,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7205357439815998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 3976
    },
    {
      "completion_length": 238.6875123977661,
      "epoch": 0.6669181440965674,
      "grad_norm": 0.24182324189802537,
      "kl": 0.04228973388671875,
      "learning_rate": 4.91111111111111e-07,
      "loss": 0.0,
      "reward": 1.7285714820027351,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7375000454485416,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3978
    },
    {
      "completion_length": 234.69643688201904,
      "epoch": 0.6672534473364349,
      "grad_norm": 0.2698467568318753,
      "kl": 0.0405731201171875,
      "learning_rate": 4.91358024691358e-07,
      "loss": 0.0,
      "reward": 1.7500000819563866,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 3980
    },
    {
      "completion_length": 236.25893783569336,
      "epoch": 0.6675887505763024,
      "grad_norm": 0.23898061740767884,
      "kl": 0.0444488525390625,
      "learning_rate": 4.916049382716049e-07,
      "loss": 0.0,
      "reward": 1.7750000581145287,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 3982
    },
    {
      "completion_length": 246.40626049041748,
      "epoch": 0.66792405381617,
      "grad_norm": 0.20696305255706168,
      "kl": 0.05846405029296875,
      "learning_rate": 4.918518518518519e-07,
      "loss": 0.0001,
      "reward": 1.7321429401636124,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321428824216127,
      "rewards/format_reward_func": 1.0,
      "step": 3984
    },
    {
      "completion_length": 248.3214406967163,
      "epoch": 0.6682593570560376,
      "grad_norm": 0.21276370391828747,
      "kl": 0.04009246826171875,
      "learning_rate": 4.920987654320987e-07,
      "loss": 0.0,
      "reward": 1.8107143342494965,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 3986
    },
    {
      "completion_length": 254.3259038925171,
      "epoch": 0.6685946602959051,
      "grad_norm": 0.10395895819260763,
      "kl": 0.0443267822265625,
      "learning_rate": 4.923456790123456e-07,
      "loss": 0.0,
      "reward": 1.68571437895298,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.6857143174856901,
      "rewards/format_reward_func": 1.0,
      "step": 3988
    },
    {
      "completion_length": 233.83929824829102,
      "epoch": 0.6689299635357726,
      "grad_norm": 0.3865026619681927,
      "kl": 0.055450439453125,
      "learning_rate": 4.925925925925926e-07,
      "loss": 0.0001,
      "reward": 1.8285714983940125,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.8375000171363354,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 3990
    },
    {
      "completion_length": 247.30358219146729,
      "epoch": 0.6692652667756402,
      "grad_norm": 0.1731516426047963,
      "kl": 0.0422210693359375,
      "learning_rate": 4.928395061728395e-07,
      "loss": 0.0,
      "reward": 1.8142857626080513,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857514321804,
      "rewards/format_reward_func": 1.0,
      "step": 3992
    },
    {
      "completion_length": 242.26340579986572,
      "epoch": 0.6696005700155078,
      "grad_norm": 0.3130980713610786,
      "kl": 0.0402679443359375,
      "learning_rate": 4.930864197530864e-07,
      "loss": 0.0,
      "reward": 1.7785715013742447,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 3994
    },
    {
      "completion_length": 240.4776906967163,
      "epoch": 0.6699358732553753,
      "grad_norm": 0.18480171777833662,
      "kl": 0.04180145263671875,
      "learning_rate": 4.933333333333333e-07,
      "loss": 0.0,
      "reward": 1.7464286386966705,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 3996
    },
    {
      "completion_length": 239.3214406967163,
      "epoch": 0.6702711764952429,
      "grad_norm": 0.2287792806277298,
      "kl": 0.05632781982421875,
      "learning_rate": 4.935802469135802e-07,
      "loss": 0.0001,
      "reward": 1.8071428909897804,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428872644901,
      "rewards/format_reward_func": 1.0,
      "step": 3998
    },
    {
      "completion_length": 233.94643783569336,
      "epoch": 0.6706064797351105,
      "grad_norm": 0.05918938448798421,
      "kl": 0.038482666015625,
      "learning_rate": 4.938271604938271e-07,
      "loss": 0.0,
      "reward": 1.7642857730388641,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7732143104076385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4000
    },
    {
      "completion_length": 253.99554920196533,
      "epoch": 0.670941782974978,
      "grad_norm": 0.1558838101555109,
      "kl": 0.039703369140625,
      "learning_rate": 4.94074074074074e-07,
      "loss": 0.0,
      "reward": 1.730357214808464,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7348214518278837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4002
    },
    {
      "completion_length": 237.4062623977661,
      "epoch": 0.6712770862148455,
      "grad_norm": 0.23672466601064102,
      "kl": 0.03436279296875,
      "learning_rate": 4.94320987654321e-07,
      "loss": 0.0,
      "reward": 1.755357213318348,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214596509933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4004
    },
    {
      "completion_length": 235.45983028411865,
      "epoch": 0.6716123894547131,
      "grad_norm": 0.23630688175269002,
      "kl": 0.04888153076171875,
      "learning_rate": 4.945679012345679e-07,
      "loss": 0.0,
      "reward": 1.7625000774860382,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7669643275439739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4006
    },
    {
      "completion_length": 234.4107255935669,
      "epoch": 0.6719476926945807,
      "grad_norm": 0.1509615780935135,
      "kl": 0.0428009033203125,
      "learning_rate": 4.948148148148148e-07,
      "loss": 0.0,
      "reward": 1.7142857909202576,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857424914837,
      "rewards/format_reward_func": 1.0,
      "step": 4008
    },
    {
      "completion_length": 248.5669755935669,
      "epoch": 0.6722829959344482,
      "grad_norm": 0.24903427922550173,
      "kl": 0.03661346435546875,
      "learning_rate": 4.950617283950617e-07,
      "loss": 0.0,
      "reward": 1.7750000804662704,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7839285917580128,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4010
    },
    {
      "completion_length": 239.47322463989258,
      "epoch": 0.6726182991743158,
      "grad_norm": 0.3243591425526177,
      "kl": 0.0417327880859375,
      "learning_rate": 4.953086419753086e-07,
      "loss": 0.0,
      "reward": 1.7803572118282318,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7848214469850063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4012
    },
    {
      "completion_length": 245.58483028411865,
      "epoch": 0.6729536024141833,
      "grad_norm": 0.19513335100841234,
      "kl": 0.055511474609375,
      "learning_rate": 4.955555555555556e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 4014
    },
    {
      "completion_length": 232.09822273254395,
      "epoch": 0.6732889056540509,
      "grad_norm": 0.2645182804473894,
      "kl": 0.0370941162109375,
      "learning_rate": 4.958024691358024e-07,
      "loss": 0.0,
      "reward": 1.8321428894996643,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8321428894996643,
      "rewards/format_reward_func": 1.0,
      "step": 4016
    },
    {
      "completion_length": 232.3392972946167,
      "epoch": 0.6736242088939184,
      "grad_norm": 0.2761508130046895,
      "kl": 0.04465484619140625,
      "learning_rate": 4.960493827160493e-07,
      "loss": 0.0,
      "reward": 1.7642857804894447,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 4018
    },
    {
      "completion_length": 244.05358123779297,
      "epoch": 0.673959512133786,
      "grad_norm": 0.29287932155581453,
      "kl": 0.085784912109375,
      "learning_rate": 4.962962962962963e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 4020
    },
    {
      "completion_length": 240.5223331451416,
      "epoch": 0.6742948153736535,
      "grad_norm": 0.224555054040133,
      "kl": 0.03826904296875,
      "learning_rate": 4.965432098765432e-07,
      "loss": 0.0,
      "reward": 1.7928571999073029,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 4022
    },
    {
      "completion_length": 234.50001335144043,
      "epoch": 0.6746301186135211,
      "grad_norm": 0.24682979395578233,
      "kl": 0.037628173828125,
      "learning_rate": 4.967901234567901e-07,
      "loss": 0.0,
      "reward": 1.7964286506175995,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 4024
    },
    {
      "completion_length": 233.65179634094238,
      "epoch": 0.6749654218533887,
      "grad_norm": 0.1787319429847436,
      "kl": 0.04058074951171875,
      "learning_rate": 4.97037037037037e-07,
      "loss": 0.0,
      "reward": 1.7892857566475868,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 4026
    },
    {
      "completion_length": 252.76340675354004,
      "epoch": 0.6753007250932562,
      "grad_norm": 0.39406210419938975,
      "kl": 0.0468902587890625,
      "learning_rate": 4.972839506172839e-07,
      "loss": 0.0,
      "reward": 1.7428572103381157,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7428571656346321,
      "rewards/format_reward_func": 1.0,
      "step": 4028
    },
    {
      "completion_length": 242.0937614440918,
      "epoch": 0.6756360283331238,
      "grad_norm": 0.4016790575300383,
      "kl": 0.043731689453125,
      "learning_rate": 4.975308641975308e-07,
      "loss": 0.0,
      "reward": 1.764285758137703,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857544124126,
      "rewards/format_reward_func": 1.0,
      "step": 4030
    },
    {
      "completion_length": 244.1384048461914,
      "epoch": 0.6759713315729913,
      "grad_norm": 0.33463115537827764,
      "kl": 0.04061126708984375,
      "learning_rate": 4.977777777777777e-07,
      "loss": 0.0,
      "reward": 1.7857143431901932,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7946428768336773,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4032
    },
    {
      "completion_length": 252.40179443359375,
      "epoch": 0.6763066348128589,
      "grad_norm": 0.16931830679373155,
      "kl": 0.0411834716796875,
      "learning_rate": 4.980246913580247e-07,
      "loss": 0.0,
      "reward": 1.8017857521772385,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8062500357627869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4034
    },
    {
      "completion_length": 241.4687614440918,
      "epoch": 0.6766419380527264,
      "grad_norm": 0.3814898253396864,
      "kl": 0.0712890625,
      "learning_rate": 4.982716049382716e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 4036
    },
    {
      "completion_length": 247.8750123977661,
      "epoch": 0.676977241292594,
      "grad_norm": 0.16397861442788428,
      "kl": 0.0486602783203125,
      "learning_rate": 4.985185185185185e-07,
      "loss": 0.0,
      "reward": 1.7410715147852898,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7544643171131611,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 4038
    },
    {
      "completion_length": 252.11608409881592,
      "epoch": 0.6773125445324616,
      "grad_norm": 0.6544560134450959,
      "kl": 0.1189422607421875,
      "learning_rate": 4.987654320987654e-07,
      "loss": 0.0001,
      "reward": 1.7446429207921028,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7482143267989159,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 4040
    },
    {
      "completion_length": 245.38394165039062,
      "epoch": 0.6776478477723291,
      "grad_norm": 0.2508024090704286,
      "kl": 0.0443267822265625,
      "learning_rate": 4.990123456790123e-07,
      "loss": 0.0,
      "reward": 1.7250000834465027,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000350177288,
      "rewards/format_reward_func": 1.0,
      "step": 4042
    },
    {
      "completion_length": 251.5044765472412,
      "epoch": 0.6779831510121966,
      "grad_norm": 0.2267515588216866,
      "kl": 0.03842926025390625,
      "learning_rate": 4.992592592592593e-07,
      "loss": 0.0,
      "reward": 1.7410714849829674,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7455357573926449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4044
    },
    {
      "completion_length": 248.20090579986572,
      "epoch": 0.6783184542520642,
      "grad_norm": 0.24181546344204055,
      "kl": 0.04335784912109375,
      "learning_rate": 4.995061728395061e-07,
      "loss": 0.0,
      "reward": 1.7535715103149414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 4046
    },
    {
      "completion_length": 243.383939743042,
      "epoch": 0.6786537574919318,
      "grad_norm": 0.1358614247171582,
      "kl": 0.03560638427734375,
      "learning_rate": 4.99753086419753e-07,
      "loss": 0.0,
      "reward": 1.7392858043313026,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 1.0,
      "step": 4048
    },
    {
      "completion_length": 253.8303689956665,
      "epoch": 0.6789890607317993,
      "grad_norm": 0.17895015083217394,
      "kl": 0.0395355224609375,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 1.7450893595814705,
      "reward_std": 0.037249373737722635,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 4050
    },
    {
      "completion_length": 255.852689743042,
      "epoch": 0.6793243639716668,
      "grad_norm": 0.28552055154012596,
      "kl": 0.038604736328125,
      "learning_rate": 4.999999997122036e-07,
      "loss": 0.0,
      "reward": 1.7285715192556381,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.728571442887187,
      "rewards/format_reward_func": 1.0,
      "step": 4052
    },
    {
      "completion_length": 243.11608123779297,
      "epoch": 0.6796596672115345,
      "grad_norm": 0.4019371586263685,
      "kl": 0.04984283447265625,
      "learning_rate": 4.999999988488147e-07,
      "loss": 0.0,
      "reward": 1.682142972946167,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.6821428947150707,
      "rewards/format_reward_func": 1.0,
      "step": 4054
    },
    {
      "completion_length": 255.87947845458984,
      "epoch": 0.679994970451402,
      "grad_norm": 0.21289432083989482,
      "kl": 0.04183197021484375,
      "learning_rate": 4.999999974098329e-07,
      "loss": 0.0,
      "reward": 1.7607143744826317,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.760714303702116,
      "rewards/format_reward_func": 1.0,
      "step": 4056
    },
    {
      "completion_length": 247.18304538726807,
      "epoch": 0.6803302736912695,
      "grad_norm": 0.5173896017748014,
      "kl": 0.05936431884765625,
      "learning_rate": 4.999999953952585e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 4058
    },
    {
      "completion_length": 254.18751335144043,
      "epoch": 0.680665576931137,
      "grad_norm": 0.21114889600510645,
      "kl": 0.03630828857421875,
      "learning_rate": 4.999999928050913e-07,
      "loss": 0.0,
      "reward": 1.6946429163217545,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.6991071812808514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4060
    },
    {
      "completion_length": 263.2276916503906,
      "epoch": 0.6810008801710047,
      "grad_norm": 0.7612326196090565,
      "kl": 0.044281005859375,
      "learning_rate": 4.999999896393315e-07,
      "loss": 0.0,
      "reward": 1.7892857864499092,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.789285734295845,
      "rewards/format_reward_func": 1.0,
      "step": 4062
    },
    {
      "completion_length": 249.9509038925171,
      "epoch": 0.6813361834108722,
      "grad_norm": 0.3232769892257542,
      "kl": 0.0448150634765625,
      "learning_rate": 4.99999985897979e-07,
      "loss": 0.0,
      "reward": 1.7696429193019867,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4064
    },
    {
      "completion_length": 250.33483409881592,
      "epoch": 0.6816714866507397,
      "grad_norm": 0.2483480478469988,
      "kl": 0.03652191162109375,
      "learning_rate": 4.99999981581034e-07,
      "loss": 0.0,
      "reward": 1.76071435213089,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 4066
    },
    {
      "completion_length": 254.665189743042,
      "epoch": 0.6820067898906074,
      "grad_norm": 0.18635947460012328,
      "kl": 0.03978729248046875,
      "learning_rate": 4.999999766884962e-07,
      "loss": 0.0,
      "reward": 1.739285796880722,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 1.0,
      "step": 4068
    },
    {
      "completion_length": 251.08483123779297,
      "epoch": 0.6823420931304749,
      "grad_norm": 0.16341314696001333,
      "kl": 0.03818511962890625,
      "learning_rate": 4.999999712203659e-07,
      "loss": 0.0,
      "reward": 1.7321429252624512,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7321428805589676,
      "rewards/format_reward_func": 1.0,
      "step": 4070
    },
    {
      "completion_length": 248.5803689956665,
      "epoch": 0.6826773963703424,
      "grad_norm": 0.22882997275793895,
      "kl": 0.03765106201171875,
      "learning_rate": 4.999999651766429e-07,
      "loss": 0.0,
      "reward": 1.8285714909434319,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8285714499652386,
      "rewards/format_reward_func": 1.0,
      "step": 4072
    },
    {
      "completion_length": 247.45983028411865,
      "epoch": 0.6830126996102099,
      "grad_norm": 0.30347725721095986,
      "kl": 0.03338623046875,
      "learning_rate": 4.999999585573273e-07,
      "loss": 0.0,
      "reward": 1.769642911851406,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.774107176810503,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4074
    },
    {
      "completion_length": 243.571439743042,
      "epoch": 0.6833480028500776,
      "grad_norm": 0.09929448446690448,
      "kl": 0.0332489013671875,
      "learning_rate": 4.99999951362419e-07,
      "loss": 0.0,
      "reward": 1.7821429148316383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428999304771,
      "rewards/format_reward_func": 1.0,
      "step": 4076
    },
    {
      "completion_length": 248.22322463989258,
      "epoch": 0.6836833060899451,
      "grad_norm": 0.23674906761103792,
      "kl": 0.03179168701171875,
      "learning_rate": 4.999999435919181e-07,
      "loss": 0.0,
      "reward": 1.8196429163217545,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8241071738302708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4078
    },
    {
      "completion_length": 249.00001430511475,
      "epoch": 0.6840186093298126,
      "grad_norm": 0.2803521117329266,
      "kl": 0.03696441650390625,
      "learning_rate": 4.999999352458248e-07,
      "loss": 0.0,
      "reward": 1.7571429461240768,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.757142873480916,
      "rewards/format_reward_func": 1.0,
      "step": 4080
    },
    {
      "completion_length": 255.54465293884277,
      "epoch": 0.6843539125696803,
      "grad_norm": 0.18009514947831426,
      "kl": 0.08721160888671875,
      "learning_rate": 4.999999263241389e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7803571708500385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4082
    },
    {
      "completion_length": 255.66072463989258,
      "epoch": 0.6846892158095478,
      "grad_norm": 0.2365843923361676,
      "kl": 0.04498291015625,
      "learning_rate": 4.999999168268604e-07,
      "loss": 0.0,
      "reward": 1.7642857730388641,
      "reward_std": 0.09091372601687908,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 4084
    },
    {
      "completion_length": 246.0134048461914,
      "epoch": 0.6850245190494153,
      "grad_norm": 0.4343459100999908,
      "kl": 0.03946685791015625,
      "learning_rate": 4.999999067539895e-07,
      "loss": 0.0,
      "reward": 1.7660714909434319,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4086
    },
    {
      "completion_length": 255.75001335144043,
      "epoch": 0.6853598222892828,
      "grad_norm": 0.22611496179012103,
      "kl": 0.03882598876953125,
      "learning_rate": 4.99999896105526e-07,
      "loss": 0.0,
      "reward": 1.7589286491274834,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.763392873108387,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4088
    },
    {
      "completion_length": 245.85268878936768,
      "epoch": 0.6856951255291505,
      "grad_norm": 0.1820744807846706,
      "kl": 0.055633544921875,
      "learning_rate": 4.999998848814701e-07,
      "loss": 0.0001,
      "reward": 1.742857202887535,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 4090
    },
    {
      "completion_length": 245.1785831451416,
      "epoch": 0.686030428769018,
      "grad_norm": 0.2699339435142035,
      "kl": 0.0371856689453125,
      "learning_rate": 4.999998730818219e-07,
      "loss": 0.0,
      "reward": 1.7767857760190964,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7812500409781933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4092
    },
    {
      "completion_length": 250.8794755935669,
      "epoch": 0.6863657320088855,
      "grad_norm": 0.18753002933538254,
      "kl": 0.06398773193359375,
      "learning_rate": 4.999998607065812e-07,
      "loss": 0.0001,
      "reward": 1.7285714820027351,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714726895094,
      "rewards/format_reward_func": 1.0,
      "step": 4094
    },
    {
      "completion_length": 250.4509038925171,
      "epoch": 0.6867010352487531,
      "grad_norm": 0.23536728398067072,
      "kl": 0.0369110107421875,
      "learning_rate": 4.999998477557482e-07,
      "loss": 0.0,
      "reward": 1.7250000834465027,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7250000275671482,
      "rewards/format_reward_func": 1.0,
      "step": 4096
    },
    {
      "completion_length": 249.23215770721436,
      "epoch": 0.6870363384886207,
      "grad_norm": 0.2694385693254488,
      "kl": 0.04195404052734375,
      "learning_rate": 4.999998342293226e-07,
      "loss": 0.0,
      "reward": 1.7125000655651093,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7169643379747868,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4098
    },
    {
      "completion_length": 237.4642972946167,
      "epoch": 0.6873716417284882,
      "grad_norm": 0.20593764322751265,
      "kl": 0.05100250244140625,
      "learning_rate": 4.999998201273049e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7982143014669418,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4100
    },
    {
      "completion_length": 243.3303689956665,
      "epoch": 0.6877069449683557,
      "grad_norm": 0.2052447336531401,
      "kl": 0.03682708740234375,
      "learning_rate": 4.999998054496949e-07,
      "loss": 0.0,
      "reward": 1.7892857864499092,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857491970062,
      "rewards/format_reward_func": 1.0,
      "step": 4102
    },
    {
      "completion_length": 248.73215293884277,
      "epoch": 0.6880422482082234,
      "grad_norm": 0.19766473428498785,
      "kl": 0.0457916259765625,
      "learning_rate": 4.999997901964926e-07,
      "loss": 0.0,
      "reward": 1.7357143461704254,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.735714316368103,
      "rewards/format_reward_func": 1.0,
      "step": 4104
    },
    {
      "completion_length": 252.6250123977661,
      "epoch": 0.6883775514480909,
      "grad_norm": 0.2928526606003355,
      "kl": 0.04019927978515625,
      "learning_rate": 4.999997743676982e-07,
      "loss": 0.0,
      "reward": 1.7214286625385284,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7303571812808514,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4106
    },
    {
      "completion_length": 237.3973331451416,
      "epoch": 0.6887128546879584,
      "grad_norm": 0.11667197186742236,
      "kl": 0.034725189208984375,
      "learning_rate": 4.999997579633115e-07,
      "loss": 0.0,
      "reward": 1.8178571909666061,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.8267857357859612,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4108
    },
    {
      "completion_length": 244.5446538925171,
      "epoch": 0.689048157927826,
      "grad_norm": 0.2764998637390368,
      "kl": 0.06385040283203125,
      "learning_rate": 4.999997409833327e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 4110
    },
    {
      "completion_length": 244.508939743042,
      "epoch": 0.6893834611676936,
      "grad_norm": 0.2871939727209474,
      "kl": 0.038726806640625,
      "learning_rate": 4.999997234277618e-07,
      "loss": 0.0,
      "reward": 1.725000075995922,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000312924385,
      "rewards/format_reward_func": 1.0,
      "step": 4112
    },
    {
      "completion_length": 240.6428680419922,
      "epoch": 0.6897187644075611,
      "grad_norm": 0.26143479711552936,
      "kl": 0.0357818603515625,
      "learning_rate": 4.999997052965989e-07,
      "loss": 0.0,
      "reward": 1.767857201397419,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.767857177183032,
      "rewards/format_reward_func": 1.0,
      "step": 4114
    },
    {
      "completion_length": 251.58929824829102,
      "epoch": 0.6900540676474286,
      "grad_norm": 0.3161348591284228,
      "kl": 0.09564971923828125,
      "learning_rate": 4.99999686589844e-07,
      "loss": 0.0001,
      "reward": 1.7107143551111221,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7107143215835094,
      "rewards/format_reward_func": 1.0,
      "step": 4116
    },
    {
      "completion_length": 247.52232933044434,
      "epoch": 0.6903893708872962,
      "grad_norm": 0.45746669207629176,
      "kl": 0.18476104736328125,
      "learning_rate": 4.999996673074971e-07,
      "loss": 0.0002,
      "reward": 1.7696429193019867,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7741071600466967,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4118
    },
    {
      "completion_length": 249.89733123779297,
      "epoch": 0.6907246741271638,
      "grad_norm": 0.275198361958013,
      "kl": 0.0482330322265625,
      "learning_rate": 4.999996474495583e-07,
      "loss": 0.0,
      "reward": 1.796428643167019,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7964285910129547,
      "rewards/format_reward_func": 1.0,
      "step": 4120
    },
    {
      "completion_length": 250.66072940826416,
      "epoch": 0.6910599773670313,
      "grad_norm": 0.1616753946632268,
      "kl": 0.0462493896484375,
      "learning_rate": 4.999996270160275e-07,
      "loss": 0.0,
      "reward": 1.7839286103844643,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4122
    },
    {
      "completion_length": 242.08929920196533,
      "epoch": 0.6913952806068988,
      "grad_norm": 0.28224358954519824,
      "kl": 0.03643035888671875,
      "learning_rate": 4.999996060069049e-07,
      "loss": 0.0,
      "reward": 1.739285796880722,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857521772385,
      "rewards/format_reward_func": 1.0,
      "step": 4124
    },
    {
      "completion_length": 239.62054538726807,
      "epoch": 0.6917305838467664,
      "grad_norm": 0.2959435225661601,
      "kl": 0.32910919189453125,
      "learning_rate": 4.999995844221906e-07,
      "loss": 0.0003,
      "reward": 1.7642857506871223,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 4126
    },
    {
      "completion_length": 256.1160840988159,
      "epoch": 0.692065887086634,
      "grad_norm": 0.16159897450649705,
      "kl": 0.04415130615234375,
      "learning_rate": 4.999995622618846e-07,
      "loss": 0.0,
      "reward": 1.7196428999304771,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.724107176065445,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4128
    },
    {
      "completion_length": 241.51787185668945,
      "epoch": 0.6924011903265015,
      "grad_norm": 0.16545788625877894,
      "kl": 0.0444793701171875,
      "learning_rate": 4.999995395259868e-07,
      "loss": 0.0,
      "reward": 1.7250000685453415,
      "reward_std": 0.025253813713788986,
      "rewards/equation_reward_func": 0.7339285910129547,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4130
    },
    {
      "completion_length": 245.4241180419922,
      "epoch": 0.6927364935663691,
      "grad_norm": 0.1501842352304599,
      "kl": 0.6236724853515625,
      "learning_rate": 4.999995162144974e-07,
      "loss": 0.0006,
      "reward": 1.79464291036129,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.8080357499420643,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 4132
    },
    {
      "completion_length": 248.04911708831787,
      "epoch": 0.6930717968062367,
      "grad_norm": 0.12794145815186433,
      "kl": 0.04578399658203125,
      "learning_rate": 4.999994923274164e-07,
      "loss": 0.0,
      "reward": 1.8071429133415222,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428798139095,
      "rewards/format_reward_func": 1.0,
      "step": 4134
    },
    {
      "completion_length": 253.52233219146729,
      "epoch": 0.6934071000461042,
      "grad_norm": 0.24779154668910555,
      "kl": 0.340484619140625,
      "learning_rate": 4.999994678647439e-07,
      "loss": 0.0003,
      "reward": 1.817857213318348,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8178571611642838,
      "rewards/format_reward_func": 1.0,
      "step": 4136
    },
    {
      "completion_length": 252.89733409881592,
      "epoch": 0.6937424032859717,
      "grad_norm": 0.06910974345915508,
      "kl": 0.4642486572265625,
      "learning_rate": 4.999994428264799e-07,
      "loss": 0.0005,
      "reward": 1.7196429371833801,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7241071704775095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4138
    },
    {
      "completion_length": 245.62501430511475,
      "epoch": 0.6940777065258393,
      "grad_norm": 0.24250637427227054,
      "kl": 0.04045867919921875,
      "learning_rate": 4.999994172126245e-07,
      "loss": 0.0,
      "reward": 1.7607143595814705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 4140
    },
    {
      "completion_length": 252.32143783569336,
      "epoch": 0.6944130097657069,
      "grad_norm": 0.21264235248878327,
      "kl": 0.040924072265625,
      "learning_rate": 4.999993910231778e-07,
      "loss": 0.0,
      "reward": 1.7875000685453415,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.791964303702116,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4142
    },
    {
      "completion_length": 256.4776887893677,
      "epoch": 0.6947483130055744,
      "grad_norm": 0.18883792480654485,
      "kl": 0.0831146240234375,
      "learning_rate": 4.999993642581397e-07,
      "loss": 0.0001,
      "reward": 1.7803571969270706,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.793750025331974,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 4144
    },
    {
      "completion_length": 256.1026906967163,
      "epoch": 0.695083616245442,
      "grad_norm": 0.4030051813383355,
      "kl": 0.08502197265625,
      "learning_rate": 4.999993369175105e-07,
      "loss": 0.0001,
      "reward": 1.7750000953674316,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7839285954833031,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4146
    },
    {
      "completion_length": 250.8705472946167,
      "epoch": 0.6954189194853095,
      "grad_norm": 0.25329291872625403,
      "kl": 0.055938720703125,
      "learning_rate": 4.999993090012901e-07,
      "loss": 0.0001,
      "reward": 1.723214365541935,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7276785969734192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4148
    },
    {
      "completion_length": 261.0134029388428,
      "epoch": 0.6957542227251771,
      "grad_norm": 0.28460622009753217,
      "kl": 0.0702972412109375,
      "learning_rate": 4.999992805094786e-07,
      "loss": 0.0001,
      "reward": 1.764285758137703,
      "reward_std": 0.11111677717417479,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 4150
    },
    {
      "completion_length": 260.727689743042,
      "epoch": 0.6960895259650446,
      "grad_norm": 0.24220639520506346,
      "kl": 0.05754852294921875,
      "learning_rate": 4.99999251442076e-07,
      "loss": 0.0001,
      "reward": 1.7517857775092125,
      "reward_std": 0.08838834706693888,
      "rewards/equation_reward_func": 0.7651786059141159,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 4152
    },
    {
      "completion_length": 253.7053680419922,
      "epoch": 0.6964248292049122,
      "grad_norm": 0.20059391948110178,
      "kl": 0.0672760009765625,
      "learning_rate": 4.999992217990825e-07,
      "loss": 0.0001,
      "reward": 1.7571429163217545,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 4154
    },
    {
      "completion_length": 257.23661708831787,
      "epoch": 0.6967601324447797,
      "grad_norm": 0.17352793125955543,
      "kl": 0.09174346923828125,
      "learning_rate": 4.99999191580498e-07,
      "loss": 0.0001,
      "reward": 1.7660714909434319,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7705357521772385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4156
    },
    {
      "completion_length": 247.43304538726807,
      "epoch": 0.6970954356846473,
      "grad_norm": 0.4455503723352921,
      "kl": 0.055694580078125,
      "learning_rate": 4.999991607863228e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 4158
    },
    {
      "completion_length": 249.290189743042,
      "epoch": 0.6974307389245149,
      "grad_norm": 0.207070863422582,
      "kl": 0.05780029296875,
      "learning_rate": 4.999991294165567e-07,
      "loss": 0.0001,
      "reward": 1.7482143566012383,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7526785843074322,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4160
    },
    {
      "completion_length": 243.20536613464355,
      "epoch": 0.6977660421643824,
      "grad_norm": 0.30572698157206896,
      "kl": 0.05780029296875,
      "learning_rate": 4.999990974712001e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 4162
    },
    {
      "completion_length": 250.08483219146729,
      "epoch": 0.69810134540425,
      "grad_norm": 0.2265343376034305,
      "kl": 0.0560302734375,
      "learning_rate": 4.999990649502528e-07,
      "loss": 0.0001,
      "reward": 1.7500000894069672,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 4164
    },
    {
      "completion_length": 250.10715579986572,
      "epoch": 0.6984366486441175,
      "grad_norm": 0.204340719128157,
      "kl": 0.04798126220703125,
      "learning_rate": 4.999990318537149e-07,
      "loss": 0.0,
      "reward": 1.7571429014205933,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 4166
    },
    {
      "completion_length": 261.6071548461914,
      "epoch": 0.6987719518839851,
      "grad_norm": 0.2263993650313919,
      "kl": 0.0576019287109375,
      "learning_rate": 4.999989981815865e-07,
      "loss": 0.0001,
      "reward": 1.7660715132951736,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7705357372760773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4168
    },
    {
      "completion_length": 258.6785840988159,
      "epoch": 0.6991072551238526,
      "grad_norm": 0.5688840890288541,
      "kl": 0.08251953125,
      "learning_rate": 4.999989639338678e-07,
      "loss": 0.0001,
      "reward": 1.6928572356700897,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.6928571872413158,
      "rewards/format_reward_func": 1.0,
      "step": 4170
    },
    {
      "completion_length": 251.9866189956665,
      "epoch": 0.6994425583637202,
      "grad_norm": 0.26599224494018464,
      "kl": 0.06134796142578125,
      "learning_rate": 4.999989291105588e-07,
      "loss": 0.0001,
      "reward": 1.714285783469677,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7232143376022577,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4172
    },
    {
      "completion_length": 243.78572463989258,
      "epoch": 0.6997778616035878,
      "grad_norm": 0.2615405086321927,
      "kl": 0.0572052001953125,
      "learning_rate": 4.999988937116595e-07,
      "loss": 0.0001,
      "reward": 1.744642935693264,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7491071708500385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4174
    },
    {
      "completion_length": 246.68751335144043,
      "epoch": 0.7001131648434553,
      "grad_norm": 0.27148599882970154,
      "kl": 0.060150146484375,
      "learning_rate": 4.999988577371702e-07,
      "loss": 0.0001,
      "reward": 1.7089286372065544,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7133928909897804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4176
    },
    {
      "completion_length": 246.977689743042,
      "epoch": 0.7004484680833228,
      "grad_norm": 0.28624932053045216,
      "kl": 0.0514373779296875,
      "learning_rate": 4.999988211870907e-07,
      "loss": 0.0001,
      "reward": 1.817857213318348,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.8178571611642838,
      "rewards/format_reward_func": 1.0,
      "step": 4178
    },
    {
      "completion_length": 247.38840579986572,
      "epoch": 0.7007837713231904,
      "grad_norm": 0.23016414642039873,
      "kl": 0.06423187255859375,
      "learning_rate": 4.999987840614212e-07,
      "loss": 0.0001,
      "reward": 1.7660714909434319,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7705357372760773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4180
    },
    {
      "completion_length": 244.47322845458984,
      "epoch": 0.701119074563058,
      "grad_norm": 0.35210518296275867,
      "kl": 0.0606842041015625,
      "learning_rate": 4.999987463601619e-07,
      "loss": 0.0001,
      "reward": 1.775000050663948,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 4182
    },
    {
      "completion_length": 235.9598331451416,
      "epoch": 0.7014543778029255,
      "grad_norm": 0.19201963590977644,
      "kl": 0.0580291748046875,
      "learning_rate": 4.999987080833128e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714507102966,
      "rewards/format_reward_func": 1.0,
      "step": 4184
    },
    {
      "completion_length": 240.67411518096924,
      "epoch": 0.701789681042793,
      "grad_norm": 0.13459371154887387,
      "kl": 0.04982757568359375,
      "learning_rate": 4.999986692308739e-07,
      "loss": 0.0,
      "reward": 1.7392857745289803,
      "reward_std": 0.025253813713788986,
      "rewards/equation_reward_func": 0.748214315623045,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4186
    },
    {
      "completion_length": 248.06697750091553,
      "epoch": 0.7021249842826607,
      "grad_norm": 0.3074579710955277,
      "kl": 0.05918121337890625,
      "learning_rate": 4.999986298028454e-07,
      "loss": 0.0001,
      "reward": 1.733928643167019,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.738392885774374,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4188
    },
    {
      "completion_length": 246.7500114440918,
      "epoch": 0.7024602875225282,
      "grad_norm": 0.3280048381695006,
      "kl": 0.0514984130859375,
      "learning_rate": 4.999985897992274e-07,
      "loss": 0.0001,
      "reward": 1.7214286252856255,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7214286029338837,
      "rewards/format_reward_func": 1.0,
      "step": 4190
    },
    {
      "completion_length": 237.38840293884277,
      "epoch": 0.7027955907623957,
      "grad_norm": 0.2873894479997041,
      "kl": 0.0552215576171875,
      "learning_rate": 4.999985492200199e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7500000111758709,
      "rewards/format_reward_func": 1.0,
      "step": 4192
    },
    {
      "completion_length": 235.9821538925171,
      "epoch": 0.7031308940022633,
      "grad_norm": 0.2801387774210403,
      "kl": 0.0472564697265625,
      "learning_rate": 4.999985080652232e-07,
      "loss": 0.0,
      "reward": 1.7357143461704254,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7446428835391998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4194
    },
    {
      "completion_length": 241.6071548461914,
      "epoch": 0.7034661972421309,
      "grad_norm": 0.2864929528982983,
      "kl": 0.0567474365234375,
      "learning_rate": 4.99998466334837e-07,
      "loss": 0.0001,
      "reward": 1.7285714969038963,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.7375000268220901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4196
    },
    {
      "completion_length": 233.7232255935669,
      "epoch": 0.7038015004819984,
      "grad_norm": 0.27030146938564104,
      "kl": 0.0601806640625,
      "learning_rate": 4.999984240288618e-07,
      "loss": 0.0001,
      "reward": 1.792857214808464,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7928571682423353,
      "rewards/format_reward_func": 1.0,
      "step": 4198
    },
    {
      "completion_length": 229.2187614440918,
      "epoch": 0.7041368037218659,
      "grad_norm": 0.12054623921859627,
      "kl": 0.061126708984375,
      "learning_rate": 4.999983811472975e-07,
      "loss": 0.0001,
      "reward": 1.7285715118050575,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7285714671015739,
      "rewards/format_reward_func": 1.0,
      "step": 4200
    },
    {
      "completion_length": 234.73215293884277,
      "epoch": 0.7044721069617336,
      "grad_norm": 0.19157688673168496,
      "kl": 0.05464935302734375,
      "learning_rate": 4.999983376901442e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.800000037997961,
      "rewards/format_reward_func": 1.0,
      "step": 4202
    },
    {
      "completion_length": 232.62500953674316,
      "epoch": 0.7048074102016011,
      "grad_norm": 0.31481143427392383,
      "kl": 0.05657958984375,
      "learning_rate": 4.99998293657402e-07,
      "loss": 0.0001,
      "reward": 1.7767857909202576,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7812500223517418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4204
    },
    {
      "completion_length": 241.8125114440918,
      "epoch": 0.7051427134414686,
      "grad_norm": 0.7161200664730465,
      "kl": 0.06140899658203125,
      "learning_rate": 4.999982490490711e-07,
      "loss": 0.0001,
      "reward": 1.7500001043081284,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 4206
    },
    {
      "completion_length": 237.290189743042,
      "epoch": 0.7054780166813361,
      "grad_norm": 0.2976173951165174,
      "kl": 0.074066162109375,
      "learning_rate": 4.999982038651515e-07,
      "loss": 0.0001,
      "reward": 1.7142858058214188,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7142857313156128,
      "rewards/format_reward_func": 1.0,
      "step": 4208
    },
    {
      "completion_length": 237.55804634094238,
      "epoch": 0.7058133199212038,
      "grad_norm": 1.294270106977747,
      "kl": 0.0904998779296875,
      "learning_rate": 4.999981581056434e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 4210
    },
    {
      "completion_length": 267.2544746398926,
      "epoch": 0.7061486231610713,
      "grad_norm": 0.6168138104071692,
      "kl": 0.399688720703125,
      "learning_rate": 4.999981117705468e-07,
      "loss": 0.0004,
      "reward": 1.526785783469677,
      "reward_std": 0.33082496002316475,
      "rewards/equation_reward_func": 0.6741071753203869,
      "rewards/format_reward_func": 0.8526786118745804,
      "step": 4212
    },
    {
      "completion_length": 260.12500858306885,
      "epoch": 0.7064839264009388,
      "grad_norm": 0.31882764649050893,
      "kl": 0.3432769775390625,
      "learning_rate": 4.999980648598619e-07,
      "loss": 0.0003,
      "reward": 1.548214353621006,
      "reward_std": 0.154048265889287,
      "rewards/equation_reward_func": 0.6419643182307482,
      "rewards/format_reward_func": 0.9062500335276127,
      "step": 4214
    },
    {
      "completion_length": 246.56697750091553,
      "epoch": 0.7068192296408065,
      "grad_norm": 0.24098316483255156,
      "kl": 0.2245941162109375,
      "learning_rate": 4.999980173735887e-07,
      "loss": 0.0002,
      "reward": 1.7535714879631996,
      "reward_std": 0.09596449136734009,
      "rewards/equation_reward_func": 0.7714286111295223,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 4216
    },
    {
      "completion_length": 234.25893783569336,
      "epoch": 0.707154532880674,
      "grad_norm": 0.316137458068654,
      "kl": 0.201934814453125,
      "learning_rate": 4.999979693117275e-07,
      "loss": 0.0002,
      "reward": 1.7839286178350449,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4218
    },
    {
      "completion_length": 237.37947368621826,
      "epoch": 0.7074898361205415,
      "grad_norm": 0.08814150042085736,
      "kl": 0.1511077880859375,
      "learning_rate": 4.999979206742782e-07,
      "loss": 0.0002,
      "reward": 1.7214286625385284,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7214286010712385,
      "rewards/format_reward_func": 1.0,
      "step": 4220
    },
    {
      "completion_length": 238.258939743042,
      "epoch": 0.707825139360409,
      "grad_norm": 0.18570467030693735,
      "kl": 0.11859130859375,
      "learning_rate": 4.99997871461241e-07,
      "loss": 0.0001,
      "reward": 1.7178572118282318,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7178571987897158,
      "rewards/format_reward_func": 1.0,
      "step": 4222
    },
    {
      "completion_length": 231.2009048461914,
      "epoch": 0.7081604426002767,
      "grad_norm": 0.2520463039518999,
      "kl": 0.0881805419921875,
      "learning_rate": 4.99997821672616e-07,
      "loss": 0.0001,
      "reward": 1.755357213318348,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214484751225,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4224
    },
    {
      "completion_length": 221.31250953674316,
      "epoch": 0.7084957458401442,
      "grad_norm": 0.40034280305593867,
      "kl": 0.0886993408203125,
      "learning_rate": 4.999977713084033e-07,
      "loss": 0.0001,
      "reward": 1.7839286178350449,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7883928939700127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4226
    },
    {
      "completion_length": 232.2857255935669,
      "epoch": 0.7088310490800117,
      "grad_norm": 0.18969120135227732,
      "kl": 0.161529541015625,
      "learning_rate": 4.99997720368603e-07,
      "loss": 0.0002,
      "reward": 1.7821429073810577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 4228
    },
    {
      "completion_length": 224.8750114440918,
      "epoch": 0.7091663523198793,
      "grad_norm": 0.17195233399392781,
      "kl": 0.1042633056640625,
      "learning_rate": 4.999976688532153e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428962051868,
      "rewards/format_reward_func": 1.0,
      "step": 4230
    },
    {
      "completion_length": 216.96429538726807,
      "epoch": 0.7095016555597469,
      "grad_norm": 0.26720859900169286,
      "kl": 0.0937652587890625,
      "learning_rate": 4.999976167622403e-07,
      "loss": 0.0001,
      "reward": 1.7892858013510704,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857305705547,
      "rewards/format_reward_func": 1.0,
      "step": 4232
    },
    {
      "completion_length": 230.7321548461914,
      "epoch": 0.7098369587996144,
      "grad_norm": 0.15169793509915727,
      "kl": 0.101806640625,
      "learning_rate": 4.99997564095678e-07,
      "loss": 0.0001,
      "reward": 1.739285796880722,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857521772385,
      "rewards/format_reward_func": 1.0,
      "step": 4234
    },
    {
      "completion_length": 224.22768783569336,
      "epoch": 0.7101722620394819,
      "grad_norm": 0.0039682440196801665,
      "kl": 0.0744781494140625,
      "learning_rate": 4.999975108535288e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 4236
    },
    {
      "completion_length": 231.23215293884277,
      "epoch": 0.7105075652793496,
      "grad_norm": 0.25855297377161357,
      "kl": 0.1085968017578125,
      "learning_rate": 4.999974570357925e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.757142897695303,
      "rewards/format_reward_func": 1.0,
      "step": 4238
    },
    {
      "completion_length": 221.85715293884277,
      "epoch": 0.7108428685192171,
      "grad_norm": 0.33264525079129287,
      "kl": 0.124664306640625,
      "learning_rate": 4.999974026424694e-07,
      "loss": 0.0001,
      "reward": 1.7665179520845413,
      "reward_std": 0.037249373737722635,
      "rewards/equation_reward_func": 0.7678571604192257,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 4240
    },
    {
      "completion_length": 230.91072273254395,
      "epoch": 0.7111781717590846,
      "grad_norm": 0.30380484394856,
      "kl": 0.088958740234375,
      "learning_rate": 4.999973476735596e-07,
      "loss": 0.0001,
      "reward": 1.7142858058214188,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7142857499420643,
      "rewards/format_reward_func": 1.0,
      "step": 4242
    },
    {
      "completion_length": 209.61161518096924,
      "epoch": 0.7115134749989521,
      "grad_norm": 0.4488191314283875,
      "kl": 0.10036468505859375,
      "learning_rate": 4.999972921290632e-07,
      "loss": 0.0001,
      "reward": 1.7589286416769028,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928917348385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4244
    },
    {
      "completion_length": 215.65626049041748,
      "epoch": 0.7118487782388198,
      "grad_norm": 0.2875484664537562,
      "kl": 0.10089874267578125,
      "learning_rate": 4.999972360089804e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 4246
    },
    {
      "completion_length": 224.06697368621826,
      "epoch": 0.7121840814786873,
      "grad_norm": 0.25228779296360854,
      "kl": 0.1021575927734375,
      "learning_rate": 4.99997179313311e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7857142947614193,
      "rewards/format_reward_func": 1.0,
      "step": 4248
    },
    {
      "completion_length": 221.2991189956665,
      "epoch": 0.7125193847185548,
      "grad_norm": 0.21487976932657415,
      "kl": 0.101104736328125,
      "learning_rate": 4.999971220420557e-07,
      "loss": 0.0001,
      "reward": 1.7589286267757416,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7633928824216127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4250
    },
    {
      "completion_length": 221.16518878936768,
      "epoch": 0.7128546879584224,
      "grad_norm": 0.17815512314744297,
      "kl": 0.1637115478515625,
      "learning_rate": 4.999970641952142e-07,
      "loss": 0.0002,
      "reward": 1.7464286163449287,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7464286349713802,
      "rewards/format_reward_func": 1.0,
      "step": 4252
    },
    {
      "completion_length": 229.12054634094238,
      "epoch": 0.71318999119829,
      "grad_norm": 0.28107228216462243,
      "kl": 0.1016998291015625,
      "learning_rate": 4.999970057727869e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 4254
    },
    {
      "completion_length": 224.0803680419922,
      "epoch": 0.7135252944381575,
      "grad_norm": 0.2769542511971505,
      "kl": 0.09307861328125,
      "learning_rate": 4.999969467747736e-07,
      "loss": 0.0001,
      "reward": 1.6785715073347092,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.678571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 4256
    },
    {
      "completion_length": 220.14286708831787,
      "epoch": 0.713860597678025,
      "grad_norm": 0.3152799908343859,
      "kl": 0.1051177978515625,
      "learning_rate": 4.999968872011748e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 4258
    },
    {
      "completion_length": 218.1741180419922,
      "epoch": 0.7141959009178926,
      "grad_norm": 0.3424092259120418,
      "kl": 0.09275054931640625,
      "learning_rate": 4.999968270519905e-07,
      "loss": 0.0001,
      "reward": 1.787500061094761,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7919643148779869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4260
    },
    {
      "completion_length": 217.66965198516846,
      "epoch": 0.7145312041577602,
      "grad_norm": 0.24849842953591986,
      "kl": 0.1216278076171875,
      "learning_rate": 4.999967663272207e-07,
      "loss": 0.0001,
      "reward": 1.817857176065445,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8178571574389935,
      "rewards/format_reward_func": 1.0,
      "step": 4262
    },
    {
      "completion_length": 216.36161518096924,
      "epoch": 0.7148665073976277,
      "grad_norm": 0.1771942860689893,
      "kl": 0.1179046630859375,
      "learning_rate": 4.999967050268657e-07,
      "loss": 0.0001,
      "reward": 1.7785714715719223,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 1.0,
      "step": 4264
    },
    {
      "completion_length": 227.24108028411865,
      "epoch": 0.7152018106374953,
      "grad_norm": 0.23422982109973303,
      "kl": 0.180267333984375,
      "learning_rate": 4.999966431509255e-07,
      "loss": 0.0002,
      "reward": 1.7892857789993286,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857491970062,
      "rewards/format_reward_func": 1.0,
      "step": 4266
    },
    {
      "completion_length": 222.05357933044434,
      "epoch": 0.7155371138773629,
      "grad_norm": 0.2085439923439692,
      "kl": 0.128448486328125,
      "learning_rate": 4.999965806994005e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 4268
    },
    {
      "completion_length": 222.42411994934082,
      "epoch": 0.7158724171172304,
      "grad_norm": 0.2230259143420745,
      "kl": 0.12346649169921875,
      "learning_rate": 4.999965176722905e-07,
      "loss": 0.0001,
      "reward": 1.7964286357164383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 4270
    },
    {
      "completion_length": 228.90625858306885,
      "epoch": 0.7162077203570979,
      "grad_norm": 0.14893081654121634,
      "kl": 0.11907958984375,
      "learning_rate": 4.999964540695959e-07,
      "loss": 0.0001,
      "reward": 1.7214286625385284,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7214285954833031,
      "rewards/format_reward_func": 1.0,
      "step": 4272
    },
    {
      "completion_length": 225.11608123779297,
      "epoch": 0.7165430235969655,
      "grad_norm": 0.28803787165074113,
      "kl": 0.1007537841796875,
      "learning_rate": 4.999963898913168e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 4274
    },
    {
      "completion_length": 228.2232255935669,
      "epoch": 0.7168783268368331,
      "grad_norm": 0.3384674104032473,
      "kl": 0.105621337890625,
      "learning_rate": 4.999963251374533e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428917348385,
      "rewards/format_reward_func": 1.0,
      "step": 4276
    },
    {
      "completion_length": 226.52679634094238,
      "epoch": 0.7172136300767006,
      "grad_norm": 0.2255294947243617,
      "kl": 0.087432861328125,
      "learning_rate": 4.999962598080055e-07,
      "loss": 0.0001,
      "reward": 1.6875000894069672,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.6919643133878708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4278
    },
    {
      "completion_length": 229.29911708831787,
      "epoch": 0.7175489333165682,
      "grad_norm": 0.17192694908242845,
      "kl": 0.1046295166015625,
      "learning_rate": 4.999961939029738e-07,
      "loss": 0.0001,
      "reward": 1.753571517765522,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714469850063,
      "rewards/format_reward_func": 1.0,
      "step": 4280
    },
    {
      "completion_length": 228.3571538925171,
      "epoch": 0.7178842365564357,
      "grad_norm": 0.2238045079730679,
      "kl": 0.1636505126953125,
      "learning_rate": 4.999961274223581e-07,
      "loss": 0.0002,
      "reward": 1.8071428835391998,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428909897804,
      "rewards/format_reward_func": 1.0,
      "step": 4282
    },
    {
      "completion_length": 229.05358028411865,
      "epoch": 0.7182195397963033,
      "grad_norm": 0.1545800715199412,
      "kl": 0.1165924072265625,
      "learning_rate": 4.999960603661585e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571812808514,
      "rewards/format_reward_func": 1.0,
      "step": 4284
    },
    {
      "completion_length": 221.13393878936768,
      "epoch": 0.7185548430361708,
      "grad_norm": 0.16588131399199002,
      "kl": 0.10345458984375,
      "learning_rate": 4.999959927343754e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714440047741,
      "rewards/format_reward_func": 1.0,
      "step": 4286
    },
    {
      "completion_length": 227.6250114440918,
      "epoch": 0.7188901462760384,
      "grad_norm": 0.23818363949622876,
      "kl": 0.1177215576171875,
      "learning_rate": 4.999959245270088e-07,
      "loss": 0.0001,
      "reward": 1.7571429461240768,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428921073675,
      "rewards/format_reward_func": 1.0,
      "step": 4288
    },
    {
      "completion_length": 223.75000762939453,
      "epoch": 0.719225449515906,
      "grad_norm": 0.30625169067460517,
      "kl": 0.1202545166015625,
      "learning_rate": 4.999958557440589e-07,
      "loss": 0.0001,
      "reward": 1.7446429282426834,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7491071783006191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4290
    },
    {
      "completion_length": 223.0803680419922,
      "epoch": 0.7195607527557735,
      "grad_norm": 0.3376410563758561,
      "kl": 0.11126708984375,
      "learning_rate": 4.999957863855257e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857525497675,
      "rewards/format_reward_func": 1.0,
      "step": 4292
    },
    {
      "completion_length": 220.6741180419922,
      "epoch": 0.7198960559956411,
      "grad_norm": 0.1431025431630177,
      "kl": 0.0991058349609375,
      "learning_rate": 4.999957164514097e-07,
      "loss": 0.0001,
      "reward": 1.7571428939700127,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 4294
    },
    {
      "completion_length": 218.46429538726807,
      "epoch": 0.7202313592355086,
      "grad_norm": 0.20456145213274624,
      "kl": 0.118072509765625,
      "learning_rate": 4.999956459417109e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 4296
    },
    {
      "completion_length": 220.79911422729492,
      "epoch": 0.7205666624753762,
      "grad_norm": 0.34602565657175177,
      "kl": 0.10986328125,
      "learning_rate": 4.999955748564293e-07,
      "loss": 0.0001,
      "reward": 1.7357143685221672,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.735714316368103,
      "rewards/format_reward_func": 1.0,
      "step": 4298
    },
    {
      "completion_length": 226.42858219146729,
      "epoch": 0.7209019657152437,
      "grad_norm": 0.30351734013178966,
      "kl": 0.1095733642578125,
      "learning_rate": 4.999955031955653e-07,
      "loss": 0.0001,
      "reward": 1.7607143372297287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 4300
    },
    {
      "completion_length": 224.81697177886963,
      "epoch": 0.7212372689551113,
      "grad_norm": 0.22631733827458714,
      "kl": 0.1202392578125,
      "learning_rate": 4.999954309591188e-07,
      "loss": 0.0001,
      "reward": 1.7482143491506577,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7526786103844643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4302
    },
    {
      "completion_length": 222.08929538726807,
      "epoch": 0.7215725721949788,
      "grad_norm": 0.19482129354219505,
      "kl": 0.1111297607421875,
      "learning_rate": 4.999953581470903e-07,
      "loss": 0.0001,
      "reward": 1.81428574770689,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857421189547,
      "rewards/format_reward_func": 1.0,
      "step": 4304
    },
    {
      "completion_length": 228.1116189956665,
      "epoch": 0.7219078754348464,
      "grad_norm": 0.15421096995458153,
      "kl": 0.106353759765625,
      "learning_rate": 4.999952847594796e-07,
      "loss": 0.0001,
      "reward": 1.821428619325161,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8214286006987095,
      "rewards/format_reward_func": 1.0,
      "step": 4306
    },
    {
      "completion_length": 222.49108028411865,
      "epoch": 0.722243178674714,
      "grad_norm": 0.2832644661349313,
      "kl": 0.1219482421875,
      "learning_rate": 4.999952107962873e-07,
      "loss": 0.0001,
      "reward": 1.760714367032051,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 4308
    },
    {
      "completion_length": 228.3526906967163,
      "epoch": 0.7225784819145815,
      "grad_norm": 0.2739123316787506,
      "kl": 0.1038055419921875,
      "learning_rate": 4.999951362575131e-07,
      "loss": 0.0001,
      "reward": 1.7660714834928513,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.770535746589303,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4310
    },
    {
      "completion_length": 224.91965293884277,
      "epoch": 0.722913785154449,
      "grad_norm": 0.33445050842398866,
      "kl": 0.131072998046875,
      "learning_rate": 4.999950611431576e-07,
      "loss": 0.0001,
      "reward": 1.7464286610484123,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 4312
    },
    {
      "completion_length": 226.31697463989258,
      "epoch": 0.7232490883943166,
      "grad_norm": 0.21341353048079031,
      "kl": 0.12420654296875,
      "learning_rate": 4.999949854532206e-07,
      "loss": 0.0001,
      "reward": 1.7160715162754059,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7205357402563095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4314
    },
    {
      "completion_length": 229.93304443359375,
      "epoch": 0.7235843916341842,
      "grad_norm": 0.30601848106418333,
      "kl": 0.1174163818359375,
      "learning_rate": 4.999949091877026e-07,
      "loss": 0.0001,
      "reward": 1.7589286640286446,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7633928842842579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4316
    },
    {
      "completion_length": 232.65179920196533,
      "epoch": 0.7239196948740517,
      "grad_norm": 0.1687238975285787,
      "kl": 0.10968017578125,
      "learning_rate": 4.999948323466035e-07,
      "loss": 0.0001,
      "reward": 1.6839286535978317,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.6883928887546062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4318
    },
    {
      "completion_length": 224.08483123779297,
      "epoch": 0.7242549981139192,
      "grad_norm": 0.004285123511293161,
      "kl": 0.11199951171875,
      "learning_rate": 4.999947549299238e-07,
      "loss": 0.0001,
      "reward": 1.782142885029316,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7910714726895094,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4320
    },
    {
      "completion_length": 224.9196538925171,
      "epoch": 0.7245903013537869,
      "grad_norm": 0.19576979486681245,
      "kl": 0.107696533203125,
      "learning_rate": 4.999946769376633e-07,
      "loss": 0.0001,
      "reward": 1.7357143312692642,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 4322
    },
    {
      "completion_length": 221.56697177886963,
      "epoch": 0.7249256045936544,
      "grad_norm": 0.28962875146992173,
      "kl": 0.10064697265625,
      "learning_rate": 4.999945983698224e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714715719223,
      "rewards/format_reward_func": 1.0,
      "step": 4324
    },
    {
      "completion_length": 233.17411708831787,
      "epoch": 0.7252609078335219,
      "grad_norm": 0.46336086157682704,
      "kl": 0.10491943359375,
      "learning_rate": 4.999945192264013e-07,
      "loss": 0.0001,
      "reward": 1.7665179446339607,
      "reward_std": 0.06755394907668233,
      "rewards/equation_reward_func": 0.767857164144516,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 4326
    },
    {
      "completion_length": 230.23661613464355,
      "epoch": 0.7255962110733895,
      "grad_norm": 0.2822536367827795,
      "kl": 0.1027679443359375,
      "learning_rate": 4.999944395074001e-07,
      "loss": 0.0001,
      "reward": 1.796428643167019,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 4328
    },
    {
      "completion_length": 233.15626049041748,
      "epoch": 0.7259315143132571,
      "grad_norm": 0.19606233734620573,
      "kl": 0.0909881591796875,
      "learning_rate": 4.99994359212819e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 4330
    },
    {
      "completion_length": 229.98215293884277,
      "epoch": 0.7262668175531246,
      "grad_norm": 0.3140268857004368,
      "kl": 0.103546142578125,
      "learning_rate": 4.999942783426583e-07,
      "loss": 0.0001,
      "reward": 1.7303572297096252,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7348214648663998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4332
    },
    {
      "completion_length": 241.1294765472412,
      "epoch": 0.7266021207929921,
      "grad_norm": 0.0038540600931844153,
      "kl": 0.0908050537109375,
      "learning_rate": 4.999941968969179e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7535714693367481,
      "rewards/format_reward_func": 1.0,
      "step": 4334
    },
    {
      "completion_length": 237.35268783569336,
      "epoch": 0.7269374240328598,
      "grad_norm": 0.28605535444102753,
      "kl": 0.094390869140625,
      "learning_rate": 4.999941148755983e-07,
      "loss": 0.0001,
      "reward": 1.8165179342031479,
      "reward_std": 0.0473508988507092,
      "rewards/equation_reward_func": 0.817857164889574,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 4336
    },
    {
      "completion_length": 248.6830472946167,
      "epoch": 0.7272727272727273,
      "grad_norm": 0.14850043176738945,
      "kl": 0.0977783203125,
      "learning_rate": 4.999940322786994e-07,
      "loss": 0.0001,
      "reward": 1.8232143372297287,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8276785910129547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4338
    },
    {
      "completion_length": 244.63840293884277,
      "epoch": 0.7276080305125948,
      "grad_norm": 0.3221783517308769,
      "kl": 0.1132965087890625,
      "learning_rate": 4.999939491062217e-07,
      "loss": 0.0001,
      "reward": 1.7500000894069672,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 4340
    },
    {
      "completion_length": 241.77679634094238,
      "epoch": 0.7279433337524623,
      "grad_norm": 0.182633691742018,
      "kl": 0.111907958984375,
      "learning_rate": 4.999938653581652e-07,
      "loss": 0.0001,
      "reward": 1.7107143625617027,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7107143122702837,
      "rewards/format_reward_func": 1.0,
      "step": 4342
    },
    {
      "completion_length": 248.05358028411865,
      "epoch": 0.72827863699233,
      "grad_norm": 0.18817809176508976,
      "kl": 0.1070709228515625,
      "learning_rate": 4.999937810345301e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571850061417,
      "rewards/format_reward_func": 1.0,
      "step": 4344
    },
    {
      "completion_length": 243.54018878936768,
      "epoch": 0.7286139402321975,
      "grad_norm": 0.22139500314548327,
      "kl": 0.121734619140625,
      "learning_rate": 4.999936961353166e-07,
      "loss": 0.0001,
      "reward": 1.714285783469677,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857536673546,
      "rewards/format_reward_func": 1.0,
      "step": 4346
    },
    {
      "completion_length": 238.27233409881592,
      "epoch": 0.728949243472065,
      "grad_norm": 0.2784243303823706,
      "kl": 0.097625732421875,
      "learning_rate": 4.999936106605251e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 1.0,
      "step": 4348
    },
    {
      "completion_length": 243.17858219146729,
      "epoch": 0.7292845467119327,
      "grad_norm": 0.23200086785651602,
      "kl": 0.102752685546875,
      "learning_rate": 4.999935246101554e-07,
      "loss": 0.0001,
      "reward": 1.8214286044239998,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8214286081492901,
      "rewards/format_reward_func": 1.0,
      "step": 4350
    },
    {
      "completion_length": 243.65626049041748,
      "epoch": 0.7296198499518002,
      "grad_norm": 0.2619354559845573,
      "kl": 0.103515625,
      "learning_rate": 4.99993437984208e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143223285675,
      "rewards/format_reward_func": 1.0,
      "step": 4352
    },
    {
      "completion_length": 244.821439743042,
      "epoch": 0.7299551531916677,
      "grad_norm": 0.38513346945983323,
      "kl": 0.101348876953125,
      "learning_rate": 4.99993350782683e-07,
      "loss": 0.0001,
      "reward": 1.8232143372297287,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.827678594738245,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4354
    },
    {
      "completion_length": 229.3884038925171,
      "epoch": 0.7302904564315352,
      "grad_norm": 0.12321027482845619,
      "kl": 0.107513427734375,
      "learning_rate": 4.999932630055807e-07,
      "loss": 0.0001,
      "reward": 1.8250000327825546,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8250000327825546,
      "rewards/format_reward_func": 1.0,
      "step": 4356
    },
    {
      "completion_length": 242.5446548461914,
      "epoch": 0.7306257596714029,
      "grad_norm": 0.30145710161379585,
      "kl": 0.223480224609375,
      "learning_rate": 4.99993174652901e-07,
      "loss": 0.0002,
      "reward": 1.728571504354477,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7375000454485416,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4358
    },
    {
      "completion_length": 236.34822273254395,
      "epoch": 0.7309610629112704,
      "grad_norm": 0.3049816173707025,
      "kl": 0.103057861328125,
      "learning_rate": 4.999930857246445e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714507102966,
      "rewards/format_reward_func": 1.0,
      "step": 4360
    },
    {
      "completion_length": 245.19644260406494,
      "epoch": 0.7312963661511379,
      "grad_norm": 0.262340460674097,
      "kl": 0.0929107666015625,
      "learning_rate": 4.999929962208113e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 4362
    },
    {
      "completion_length": 238.6339406967163,
      "epoch": 0.7316316693910055,
      "grad_norm": 0.21572310070899314,
      "kl": 0.097137451171875,
      "learning_rate": 4.999929061414014e-07,
      "loss": 0.0001,
      "reward": 1.7446429058909416,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7491071838885546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4364
    },
    {
      "completion_length": 236.9866180419922,
      "epoch": 0.7319669726308731,
      "grad_norm": 0.20339241801146096,
      "kl": 0.1023712158203125,
      "learning_rate": 4.999928154864152e-07,
      "loss": 0.0001,
      "reward": 1.7558036297559738,
      "reward_std": 0.04230013629421592,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 4366
    },
    {
      "completion_length": 239.3705472946167,
      "epoch": 0.7323022758707406,
      "grad_norm": 0.055339300543254144,
      "kl": 0.09466552734375,
      "learning_rate": 4.999927242558527e-07,
      "loss": 0.0001,
      "reward": 1.719642922282219,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7241071742027998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4368
    },
    {
      "completion_length": 237.75001049041748,
      "epoch": 0.7326375791106081,
      "grad_norm": 0.2988596314576556,
      "kl": 0.091400146484375,
      "learning_rate": 4.999926324497145e-07,
      "loss": 0.0001,
      "reward": 1.7321429252624512,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321428954601288,
      "rewards/format_reward_func": 1.0,
      "step": 4370
    },
    {
      "completion_length": 234.7321538925171,
      "epoch": 0.7329728823504758,
      "grad_norm": 0.12317099081226329,
      "kl": 0.095947265625,
      "learning_rate": 4.999925400680004e-07,
      "loss": 0.0001,
      "reward": 1.7392857819795609,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 1.0,
      "step": 4372
    },
    {
      "completion_length": 231.85268783569336,
      "epoch": 0.7333081855903433,
      "grad_norm": 0.2941982386958749,
      "kl": 0.09271240234375,
      "learning_rate": 4.999924471107108e-07,
      "loss": 0.0001,
      "reward": 1.807142898440361,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428872644901,
      "rewards/format_reward_func": 1.0,
      "step": 4374
    },
    {
      "completion_length": 229.54018783569336,
      "epoch": 0.7336434888302108,
      "grad_norm": 0.24718352959271161,
      "kl": 0.0890960693359375,
      "learning_rate": 4.99992353577846e-07,
      "loss": 0.0001,
      "reward": 1.7982143461704254,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.8026785925030708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4376
    },
    {
      "completion_length": 237.3794755935669,
      "epoch": 0.7339787920700783,
      "grad_norm": 0.21844899000323725,
      "kl": 0.0857391357421875,
      "learning_rate": 4.999922594694059e-07,
      "loss": 0.0001,
      "reward": 1.7982143312692642,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8026786036789417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4378
    },
    {
      "completion_length": 228.34822463989258,
      "epoch": 0.734314095309946,
      "grad_norm": 0.31139955126338215,
      "kl": 0.0860137939453125,
      "learning_rate": 4.999921647853911e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 4380
    },
    {
      "completion_length": 236.80804538726807,
      "epoch": 0.7346493985498135,
      "grad_norm": 0.25288666916101726,
      "kl": 0.0904083251953125,
      "learning_rate": 4.999920695258016e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 4382
    },
    {
      "completion_length": 237.9509038925171,
      "epoch": 0.734984701789681,
      "grad_norm": 0.15349708899045258,
      "kl": 0.0903472900390625,
      "learning_rate": 4.999919736906377e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714507102966,
      "rewards/format_reward_func": 1.0,
      "step": 4384
    },
    {
      "completion_length": 233.35715198516846,
      "epoch": 0.7353200050295486,
      "grad_norm": 0.20999069531459416,
      "kl": 0.0869598388671875,
      "learning_rate": 4.999918772798995e-07,
      "loss": 0.0001,
      "reward": 1.6910715103149414,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.6955357491970062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4386
    },
    {
      "completion_length": 225.27679538726807,
      "epoch": 0.7356553082694162,
      "grad_norm": 0.21644641902809753,
      "kl": 0.0892486572265625,
      "learning_rate": 4.999917802935875e-07,
      "loss": 0.0001,
      "reward": 1.7607143744826317,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 4388
    },
    {
      "completion_length": 242.883939743042,
      "epoch": 0.7359906115092837,
      "grad_norm": 0.2097634616773623,
      "kl": 0.10491943359375,
      "learning_rate": 4.999916827317016e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 4390
    },
    {
      "completion_length": 234.321439743042,
      "epoch": 0.7363259147491512,
      "grad_norm": 0.301361903974932,
      "kl": 0.0983428955078125,
      "learning_rate": 4.99991584594242e-07,
      "loss": 0.0001,
      "reward": 1.7267857491970062,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7312500327825546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4392
    },
    {
      "completion_length": 240.28572463989258,
      "epoch": 0.7366612179890188,
      "grad_norm": 0.2971735765340939,
      "kl": 0.106536865234375,
      "learning_rate": 4.999914858812094e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 4394
    },
    {
      "completion_length": 236.8035831451416,
      "epoch": 0.7369965212288864,
      "grad_norm": 0.24862167834729915,
      "kl": 0.1034393310546875,
      "learning_rate": 4.999913865926035e-07,
      "loss": 0.0001,
      "reward": 1.7839286252856255,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4396
    },
    {
      "completion_length": 236.5892972946167,
      "epoch": 0.7373318244687539,
      "grad_norm": 0.17730660559017838,
      "kl": 0.1024017333984375,
      "learning_rate": 4.999912867284247e-07,
      "loss": 0.0001,
      "reward": 1.7589286267757416,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7633928842842579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4398
    },
    {
      "completion_length": 235.58929538726807,
      "epoch": 0.7376671277086215,
      "grad_norm": 0.19618760815384542,
      "kl": 0.0888214111328125,
      "learning_rate": 4.999911862886735e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285925030708,
      "rewards/format_reward_func": 1.0,
      "step": 4400
    },
    {
      "completion_length": 241.71429824829102,
      "epoch": 0.738002430948489,
      "grad_norm": 0.23361102730300204,
      "kl": 0.097503662109375,
      "learning_rate": 4.999910852733496e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.09596449043601751,
      "rewards/equation_reward_func": 0.7857143059372902,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 4402
    },
    {
      "completion_length": 230.92858123779297,
      "epoch": 0.7383377341883566,
      "grad_norm": 0.13161910784321273,
      "kl": 0.102325439453125,
      "learning_rate": 4.999909836824538e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7642857544124126,
      "rewards/format_reward_func": 1.0,
      "step": 4404
    },
    {
      "completion_length": 238.383939743042,
      "epoch": 0.7386730374282241,
      "grad_norm": 0.11006102889865772,
      "kl": 0.088531494140625,
      "learning_rate": 4.999908815159859e-07,
      "loss": 0.0001,
      "reward": 1.741071492433548,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7455357387661934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4406
    },
    {
      "completion_length": 242.133939743042,
      "epoch": 0.7390083406680917,
      "grad_norm": 0.21817612106604028,
      "kl": 0.0933990478515625,
      "learning_rate": 4.999907787739464e-07,
      "loss": 0.0001,
      "reward": 1.7571429535746574,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428753435612,
      "rewards/format_reward_func": 1.0,
      "step": 4408
    },
    {
      "completion_length": 240.34375858306885,
      "epoch": 0.7393436439079593,
      "grad_norm": 0.15413740306942578,
      "kl": 0.0780181884765625,
      "learning_rate": 4.999906754563354e-07,
      "loss": 0.0001,
      "reward": 1.7446429282426834,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4410
    },
    {
      "completion_length": 231.25893783569336,
      "epoch": 0.7396789471478268,
      "grad_norm": 0.14813305726517229,
      "kl": 0.089752197265625,
      "learning_rate": 4.999905715631532e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.785714328289032,
      "rewards/format_reward_func": 1.0,
      "step": 4412
    },
    {
      "completion_length": 236.66072273254395,
      "epoch": 0.7400142503876944,
      "grad_norm": 0.46415342718216857,
      "kl": 0.08685302734375,
      "learning_rate": 4.999904670943999e-07,
      "loss": 0.0001,
      "reward": 1.7892857640981674,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7892857231199741,
      "rewards/format_reward_func": 1.0,
      "step": 4414
    },
    {
      "completion_length": 231.31697463989258,
      "epoch": 0.7403495536275619,
      "grad_norm": 0.2764112079171355,
      "kl": 0.081451416015625,
      "learning_rate": 4.999903620500759e-07,
      "loss": 0.0001,
      "reward": 1.805357202887535,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.8098214566707611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4416
    },
    {
      "completion_length": 230.25447368621826,
      "epoch": 0.7406848568674295,
      "grad_norm": 0.11346544928980262,
      "kl": 0.0750732421875,
      "learning_rate": 4.999902564301815e-07,
      "loss": 0.0001,
      "reward": 1.8035714626312256,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8035714626312256,
      "rewards/format_reward_func": 1.0,
      "step": 4418
    },
    {
      "completion_length": 227.22768878936768,
      "epoch": 0.741020160107297,
      "grad_norm": 0.23948562808508556,
      "kl": 0.0700836181640625,
      "learning_rate": 4.999901502347168e-07,
      "loss": 0.0001,
      "reward": 1.8107143342494965,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.810714315623045,
      "rewards/format_reward_func": 1.0,
      "step": 4420
    },
    {
      "completion_length": 230.86608409881592,
      "epoch": 0.7413554633471646,
      "grad_norm": 0.2489915856046395,
      "kl": 0.0712127685546875,
      "learning_rate": 4.99990043463682e-07,
      "loss": 0.0001,
      "reward": 1.6915179416537285,
      "reward_std": 0.06250318652018905,
      "rewards/equation_reward_func": 0.6928571667522192,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 4422
    },
    {
      "completion_length": 228.29911613464355,
      "epoch": 0.7416907665870321,
      "grad_norm": 0.1812571795709786,
      "kl": 0.077301025390625,
      "learning_rate": 4.999899361170775e-07,
      "loss": 0.0001,
      "reward": 1.7290179431438446,
      "reward_std": 0.0599778073374182,
      "rewards/equation_reward_func": 0.7348214685916901,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 4424
    },
    {
      "completion_length": 230.34375858306885,
      "epoch": 0.7420260698268997,
      "grad_norm": 0.26999021740137746,
      "kl": 0.0660858154296875,
      "learning_rate": 4.999898281949035e-07,
      "loss": 0.0001,
      "reward": 1.7000000849366188,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.708928631618619,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4426
    },
    {
      "completion_length": 225.80358028411865,
      "epoch": 0.7423613730667673,
      "grad_norm": 0.30457239085627186,
      "kl": 0.0674896240234375,
      "learning_rate": 4.999897196971602e-07,
      "loss": 0.0001,
      "reward": 1.7214286550879478,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7214286103844643,
      "rewards/format_reward_func": 1.0,
      "step": 4428
    },
    {
      "completion_length": 224.40626049041748,
      "epoch": 0.7426966763066348,
      "grad_norm": 0.21223745143930695,
      "kl": 0.0684814453125,
      "learning_rate": 4.999896106238479e-07,
      "loss": 0.0001,
      "reward": 1.8035714700818062,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 4430
    },
    {
      "completion_length": 227.92411708831787,
      "epoch": 0.7430319795465024,
      "grad_norm": 0.26564997925657297,
      "kl": 0.06092071533203125,
      "learning_rate": 4.999895009749667e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714693367481,
      "rewards/format_reward_func": 1.0,
      "step": 4432
    },
    {
      "completion_length": 218.03126049041748,
      "epoch": 0.7433672827863699,
      "grad_norm": 0.23386483907548092,
      "kl": 0.068023681640625,
      "learning_rate": 4.999893907505172e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 4434
    },
    {
      "completion_length": 227.23661518096924,
      "epoch": 0.7437025860262375,
      "grad_norm": 0.2580465030816322,
      "kl": 0.065032958984375,
      "learning_rate": 4.999892799504992e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 4436
    },
    {
      "completion_length": 218.92411613464355,
      "epoch": 0.744037889266105,
      "grad_norm": 0.3291190988547193,
      "kl": 0.087890625,
      "learning_rate": 4.999891685749133e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 4438
    },
    {
      "completion_length": 230.83483219146729,
      "epoch": 0.7443731925059726,
      "grad_norm": 0.20748276719564904,
      "kl": 0.0592041015625,
      "learning_rate": 4.999890566237597e-07,
      "loss": 0.0001,
      "reward": 1.8232143446803093,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.827678594738245,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4440
    },
    {
      "completion_length": 223.0892972946167,
      "epoch": 0.7447084957458402,
      "grad_norm": 0.3388534696826737,
      "kl": 0.067169189453125,
      "learning_rate": 4.999889440970385e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714562982321,
      "rewards/format_reward_func": 1.0,
      "step": 4442
    },
    {
      "completion_length": 222.49554634094238,
      "epoch": 0.7450437989857077,
      "grad_norm": 0.2706600096126691,
      "kl": 0.0620880126953125,
      "learning_rate": 4.999888309947501e-07,
      "loss": 0.0001,
      "reward": 1.7375000640749931,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643253087997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4444
    },
    {
      "completion_length": 221.94197368621826,
      "epoch": 0.7453791022255752,
      "grad_norm": 0.24547408044411922,
      "kl": 0.06488037109375,
      "learning_rate": 4.999887173168947e-07,
      "loss": 0.0001,
      "reward": 1.7285714969038963,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7285714633762836,
      "rewards/format_reward_func": 1.0,
      "step": 4446
    },
    {
      "completion_length": 214.54018783569336,
      "epoch": 0.7457144054654428,
      "grad_norm": 0.1853125626531355,
      "kl": 0.0753631591796875,
      "learning_rate": 4.999886030634727e-07,
      "loss": 0.0001,
      "reward": 1.7571429163217545,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 1.0,
      "step": 4448
    },
    {
      "completion_length": 221.7053680419922,
      "epoch": 0.7460497087053104,
      "grad_norm": 0.19931767396673297,
      "kl": 0.0742950439453125,
      "learning_rate": 4.999884882344842e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 4450
    },
    {
      "completion_length": 217.12947273254395,
      "epoch": 0.7463850119451779,
      "grad_norm": 0.1636954950166591,
      "kl": 0.06884765625,
      "learning_rate": 4.999883728299294e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 4452
    },
    {
      "completion_length": 215.13840103149414,
      "epoch": 0.7467203151850454,
      "grad_norm": 0.2250329654524265,
      "kl": 0.07806396484375,
      "learning_rate": 4.999882568498087e-07,
      "loss": 0.0001,
      "reward": 1.7107143476605415,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7107143308967352,
      "rewards/format_reward_func": 1.0,
      "step": 4454
    },
    {
      "completion_length": 213.21429443359375,
      "epoch": 0.7470556184249131,
      "grad_norm": 0.2505331278808723,
      "kl": 0.0777130126953125,
      "learning_rate": 4.999881402941225e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.792857164517045,
      "rewards/format_reward_func": 1.0,
      "step": 4456
    },
    {
      "completion_length": 222.19197463989258,
      "epoch": 0.7473909216647806,
      "grad_norm": 0.16466956778998829,
      "kl": 0.0843505859375,
      "learning_rate": 4.999880231628707e-07,
      "loss": 0.0001,
      "reward": 1.74910718947649,
      "reward_std": 0.05177031829953194,
      "rewards/equation_reward_func": 0.7553571872413158,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 4458
    },
    {
      "completion_length": 221.11161613464355,
      "epoch": 0.7477262249046481,
      "grad_norm": 0.2396358314615966,
      "kl": 0.0821380615234375,
      "learning_rate": 4.999879054560539e-07,
      "loss": 0.0001,
      "reward": 1.74642863124609,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 4460
    },
    {
      "completion_length": 227.1651906967163,
      "epoch": 0.7480615281445157,
      "grad_norm": 0.33908675843157704,
      "kl": 0.0939178466796875,
      "learning_rate": 4.999877871736723e-07,
      "loss": 0.0001,
      "reward": 1.7892857939004898,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.789285734295845,
      "rewards/format_reward_func": 1.0,
      "step": 4462
    },
    {
      "completion_length": 216.33929538726807,
      "epoch": 0.7483968313843833,
      "grad_norm": 0.14003622530992152,
      "kl": 0.09588623046875,
      "learning_rate": 4.999876683157261e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 4464
    },
    {
      "completion_length": 218.0446538925171,
      "epoch": 0.7487321346242508,
      "grad_norm": 0.2560114087268138,
      "kl": 0.09271240234375,
      "learning_rate": 4.999875488822155e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 4466
    },
    {
      "completion_length": 224.3437614440918,
      "epoch": 0.7490674378641183,
      "grad_norm": 0.31758820202141924,
      "kl": 0.07741546630859375,
      "learning_rate": 4.999874288731409e-07,
      "loss": 0.0001,
      "reward": 1.7232143729925156,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7276786118745804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4468
    },
    {
      "completion_length": 213.69643878936768,
      "epoch": 0.749402741103986,
      "grad_norm": 0.44047391205664077,
      "kl": 0.089691162109375,
      "learning_rate": 4.999873082885027e-07,
      "loss": 0.0001,
      "reward": 1.7142857983708382,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7142857387661934,
      "rewards/format_reward_func": 1.0,
      "step": 4470
    },
    {
      "completion_length": 207.86608123779297,
      "epoch": 0.7497380443438535,
      "grad_norm": 0.4861281163334542,
      "kl": 0.09527587890625,
      "learning_rate": 4.999871871283008e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 4472
    },
    {
      "completion_length": 209.946439743042,
      "epoch": 0.750073347583721,
      "grad_norm": 0.19385061512985016,
      "kl": 0.1087646484375,
      "learning_rate": 4.999870653925359e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 4474
    },
    {
      "completion_length": 214.26786518096924,
      "epoch": 0.7504086508235885,
      "grad_norm": 0.15678270263892813,
      "kl": 0.098419189453125,
      "learning_rate": 4.99986943081208e-07,
      "loss": 0.0001,
      "reward": 1.737500086426735,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.741964315995574,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4476
    },
    {
      "completion_length": 203.92411613464355,
      "epoch": 0.7507439540634562,
      "grad_norm": 0.2732266849605991,
      "kl": 0.11285400390625,
      "learning_rate": 4.999868201943175e-07,
      "loss": 0.0001,
      "reward": 1.7089286595582962,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7133928798139095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4478
    },
    {
      "completion_length": 215.11161613464355,
      "epoch": 0.7510792573033237,
      "grad_norm": 0.17807125877809504,
      "kl": 0.111358642578125,
      "learning_rate": 4.999866967318645e-07,
      "loss": 0.0001,
      "reward": 1.7071429342031479,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7071428913623095,
      "rewards/format_reward_func": 1.0,
      "step": 4480
    },
    {
      "completion_length": 216.14286708831787,
      "epoch": 0.7514145605431912,
      "grad_norm": 0.25479605394237403,
      "kl": 0.1113739013671875,
      "learning_rate": 4.999865726938497e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 4482
    },
    {
      "completion_length": 208.52233028411865,
      "epoch": 0.7517498637830589,
      "grad_norm": 0.3540568480128976,
      "kl": 0.128082275390625,
      "learning_rate": 4.999864480802729e-07,
      "loss": 0.0001,
      "reward": 1.7142858281731606,
      "reward_std": 0.07071067858487368,
      "rewards/equation_reward_func": 0.7232143059372902,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4484
    },
    {
      "completion_length": 222.41072273254395,
      "epoch": 0.7520851670229264,
      "grad_norm": 0.28369640416103614,
      "kl": 0.104766845703125,
      "learning_rate": 4.999863228911347e-07,
      "loss": 0.0001,
      "reward": 1.7214286476373672,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7294643130153418,
      "rewards/format_reward_func": 0.9919642955064774,
      "step": 4486
    },
    {
      "completion_length": 213.75893783569336,
      "epoch": 0.7524204702627939,
      "grad_norm": 0.0049135323066809514,
      "kl": 0.1141357421875,
      "learning_rate": 4.999861971264353e-07,
      "loss": 0.0001,
      "reward": 1.6982143744826317,
      "reward_std": 0.012626906856894493,
      "rewards/equation_reward_func": 0.7026786096394062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4488
    },
    {
      "completion_length": 221.0803680419922,
      "epoch": 0.7527557735026614,
      "grad_norm": 0.20714617745583425,
      "kl": 0.104827880859375,
      "learning_rate": 4.999860707861751e-07,
      "loss": 0.0001,
      "reward": 1.7607143446803093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143223285675,
      "rewards/format_reward_func": 1.0,
      "step": 4490
    },
    {
      "completion_length": 216.79911708831787,
      "epoch": 0.7530910767425291,
      "grad_norm": 0.3164850402145241,
      "kl": 0.0961151123046875,
      "learning_rate": 4.999859438703541e-07,
      "loss": 0.0001,
      "reward": 1.8017857670783997,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.8062500357627869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4492
    },
    {
      "completion_length": 216.8303680419922,
      "epoch": 0.7534263799823966,
      "grad_norm": 0.41792000510914323,
      "kl": 0.1044158935546875,
      "learning_rate": 4.999858163789728e-07,
      "loss": 0.0001,
      "reward": 1.7732143551111221,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7776785902678967,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4494
    },
    {
      "completion_length": 220.0446548461914,
      "epoch": 0.7537616832222641,
      "grad_norm": 0.3411173294484826,
      "kl": 0.0836029052734375,
      "learning_rate": 4.999856883120314e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 4496
    },
    {
      "completion_length": 220.37500858306885,
      "epoch": 0.7540969864621316,
      "grad_norm": 0.2682306203706297,
      "kl": 0.0904388427734375,
      "learning_rate": 4.999855596695304e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000316649675,
      "rewards/format_reward_func": 1.0,
      "step": 4498
    },
    {
      "completion_length": 219.66072273254395,
      "epoch": 0.7544322897019993,
      "grad_norm": 0.25123051432297105,
      "kl": 0.0814666748046875,
      "learning_rate": 4.999854304514699e-07,
      "loss": 0.0001,
      "reward": 1.7767857760190964,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7812500335276127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4500
    },
    {
      "completion_length": 221.74554538726807,
      "epoch": 0.7547675929418668,
      "grad_norm": 0.25857901554901946,
      "kl": 0.0801544189453125,
      "learning_rate": 4.999853006578503e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714285925030708,
      "rewards/format_reward_func": 1.0,
      "step": 4502
    },
    {
      "completion_length": 227.44643878936768,
      "epoch": 0.7551028961817343,
      "grad_norm": 0.37851301366513695,
      "kl": 0.092071533203125,
      "learning_rate": 4.999851702886719e-07,
      "loss": 0.0001,
      "reward": 1.741071492433548,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7455357555299997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4504
    },
    {
      "completion_length": 225.27233219146729,
      "epoch": 0.755438199421602,
      "grad_norm": 0.2513086311013245,
      "kl": 0.0858306884765625,
      "learning_rate": 4.999850393439348e-07,
      "loss": 0.0001,
      "reward": 1.7035714983940125,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7035714853554964,
      "rewards/format_reward_func": 1.0,
      "step": 4506
    },
    {
      "completion_length": 226.72768688201904,
      "epoch": 0.7557735026614695,
      "grad_norm": 0.3411021833760753,
      "kl": 0.0855865478515625,
      "learning_rate": 4.999849078236395e-07,
      "loss": 0.0001,
      "reward": 1.7035715132951736,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7035714630037546,
      "rewards/format_reward_func": 1.0,
      "step": 4508
    },
    {
      "completion_length": 223.00001049041748,
      "epoch": 0.756108805901337,
      "grad_norm": 0.2224263673764874,
      "kl": 0.09161376953125,
      "learning_rate": 4.999847757277862e-07,
      "loss": 0.0001,
      "reward": 1.8500000089406967,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8500000238418579,
      "rewards/format_reward_func": 1.0,
      "step": 4510
    },
    {
      "completion_length": 233.04911708831787,
      "epoch": 0.7564441091412045,
      "grad_norm": 0.003793286746061747,
      "kl": 0.0831298828125,
      "learning_rate": 4.999846430563753e-07,
      "loss": 0.0001,
      "reward": 1.7571429088711739,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 4512
    },
    {
      "completion_length": 226.94197368621826,
      "epoch": 0.7567794123810722,
      "grad_norm": 0.27291864041588626,
      "kl": 0.0889892578125,
      "learning_rate": 4.999845098094071e-07,
      "loss": 0.0001,
      "reward": 1.7250000834465027,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000387430191,
      "rewards/format_reward_func": 1.0,
      "step": 4514
    },
    {
      "completion_length": 224.39733219146729,
      "epoch": 0.7571147156209397,
      "grad_norm": 0.22265584848231698,
      "kl": 0.087921142578125,
      "learning_rate": 4.999843759868818e-07,
      "loss": 0.0001,
      "reward": 1.7589286118745804,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7633928917348385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4516
    },
    {
      "completion_length": 228.15179538726807,
      "epoch": 0.7574500188608072,
      "grad_norm": 0.20859738186600213,
      "kl": 0.08905029296875,
      "learning_rate": 4.999842415887999e-07,
      "loss": 0.0001,
      "reward": 1.8214286118745804,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8214285969734192,
      "rewards/format_reward_func": 1.0,
      "step": 4518
    },
    {
      "completion_length": 235.78125858306885,
      "epoch": 0.7577853221006748,
      "grad_norm": 0.2778610372869086,
      "kl": 0.10809326171875,
      "learning_rate": 4.999841066151615e-07,
      "loss": 0.0001,
      "reward": 1.7464286610484123,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 4520
    },
    {
      "completion_length": 222.90625858306885,
      "epoch": 0.7581206253405424,
      "grad_norm": 0.47212251372452274,
      "kl": 0.087066650390625,
      "learning_rate": 4.99983971065967e-07,
      "loss": 0.0001,
      "reward": 1.758928619325161,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.763392886146903,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4522
    },
    {
      "completion_length": 226.86161613464355,
      "epoch": 0.7584559285804099,
      "grad_norm": 0.17318141957091432,
      "kl": 0.08984375,
      "learning_rate": 4.999838349412168e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143357396126,
      "rewards/format_reward_func": 1.0,
      "step": 4524
    },
    {
      "completion_length": 235.1741180419922,
      "epoch": 0.7587912318202774,
      "grad_norm": 0.2131832104092294,
      "kl": 0.0755767822265625,
      "learning_rate": 4.99983698240911e-07,
      "loss": 0.0001,
      "reward": 1.7290179133415222,
      "reward_std": 0.04987628059461713,
      "rewards/equation_reward_func": 0.7348214574158192,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 4526
    },
    {
      "completion_length": 236.27233123779297,
      "epoch": 0.759126535060145,
      "grad_norm": 0.13201408256364378,
      "kl": 0.06096649169921875,
      "learning_rate": 4.999835609650501e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 4528
    },
    {
      "completion_length": 240.81697273254395,
      "epoch": 0.7594618383000126,
      "grad_norm": 0.21052478496355254,
      "kl": 0.0613861083984375,
      "learning_rate": 4.999834231136344e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143074274063,
      "rewards/format_reward_func": 1.0,
      "step": 4530
    },
    {
      "completion_length": 241.55358219146729,
      "epoch": 0.7597971415398801,
      "grad_norm": 0.13556321579045813,
      "kl": 0.058197021484375,
      "learning_rate": 4.999832846866641e-07,
      "loss": 0.0001,
      "reward": 1.7843750566244125,
      "reward_std": 0.04230013629421592,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 4532
    },
    {
      "completion_length": 227.63840579986572,
      "epoch": 0.7601324447797477,
      "grad_norm": 0.23882174356311658,
      "kl": 0.069305419921875,
      "learning_rate": 4.999831456841395e-07,
      "loss": 0.0001,
      "reward": 1.8125000670552254,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.8169643171131611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4534
    },
    {
      "completion_length": 237.0134048461914,
      "epoch": 0.7604677480196153,
      "grad_norm": 0.18469807951411582,
      "kl": 0.059722900390625,
      "learning_rate": 4.999830061060613e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 4536
    },
    {
      "completion_length": 233.2991180419922,
      "epoch": 0.7608030512594828,
      "grad_norm": 0.23399028758696575,
      "kl": 0.05745697021484375,
      "learning_rate": 4.999828659524293e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714562982321,
      "rewards/format_reward_func": 1.0,
      "step": 4538
    },
    {
      "completion_length": 239.99108219146729,
      "epoch": 0.7611383544993503,
      "grad_norm": 0.18787150584745277,
      "kl": 0.0531463623046875,
      "learning_rate": 4.999827252232441e-07,
      "loss": 0.0001,
      "reward": 1.7571429461240768,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428883820772,
      "rewards/format_reward_func": 1.0,
      "step": 4540
    },
    {
      "completion_length": 231.78125953674316,
      "epoch": 0.7614736577392179,
      "grad_norm": 0.3122036679038995,
      "kl": 0.05462646484375,
      "learning_rate": 4.99982583918506e-07,
      "loss": 0.0001,
      "reward": 1.7232143506407738,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7276786062866449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4542
    },
    {
      "completion_length": 237.63393878936768,
      "epoch": 0.7618089609790855,
      "grad_norm": 0.3230807616701815,
      "kl": 0.0553741455078125,
      "learning_rate": 4.999824420382153e-07,
      "loss": 0.0001,
      "reward": 1.7285715267062187,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7285714484751225,
      "rewards/format_reward_func": 1.0,
      "step": 4544
    },
    {
      "completion_length": 236.14286994934082,
      "epoch": 0.762144264218953,
      "grad_norm": 0.27987783469978955,
      "kl": 0.0552215576171875,
      "learning_rate": 4.999822995823724e-07,
      "loss": 0.0001,
      "reward": 1.7839286550879478,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7883928865194321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4546
    },
    {
      "completion_length": 235.84822368621826,
      "epoch": 0.7624795674588206,
      "grad_norm": 0.297388880050552,
      "kl": 0.0589752197265625,
      "learning_rate": 4.999821565509774e-07,
      "loss": 0.0001,
      "reward": 1.7392857745289803,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7392857503145933,
      "rewards/format_reward_func": 1.0,
      "step": 4548
    },
    {
      "completion_length": 227.46429538726807,
      "epoch": 0.7628148706986881,
      "grad_norm": 0.18228938587993593,
      "kl": 0.0537109375,
      "learning_rate": 4.999820129440309e-07,
      "loss": 0.0001,
      "reward": 1.7593750804662704,
      "reward_std": 0.02714784862473607,
      "rewards/equation_reward_func": 0.7607143223285675,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 4550
    },
    {
      "completion_length": 238.2500114440918,
      "epoch": 0.7631501739385557,
      "grad_norm": 0.29408530140054356,
      "kl": 0.0583953857421875,
      "learning_rate": 4.999818687615332e-07,
      "loss": 0.0001,
      "reward": 1.7379465028643608,
      "reward_std": 0.037249373737722635,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 4552
    },
    {
      "completion_length": 237.05358505249023,
      "epoch": 0.7634854771784232,
      "grad_norm": 0.21134377845046912,
      "kl": 0.0565338134765625,
      "learning_rate": 4.999817240034843e-07,
      "loss": 0.0001,
      "reward": 1.7964286506175995,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.796428594738245,
      "rewards/format_reward_func": 1.0,
      "step": 4554
    },
    {
      "completion_length": 237.1785831451416,
      "epoch": 0.7638207804182908,
      "grad_norm": 0.3860091198419233,
      "kl": 0.0579986572265625,
      "learning_rate": 4.99981578669885e-07,
      "loss": 0.0001,
      "reward": 1.7107143625617027,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7107143253087997,
      "rewards/format_reward_func": 1.0,
      "step": 4556
    },
    {
      "completion_length": 233.02233219146729,
      "epoch": 0.7641560836581583,
      "grad_norm": 0.22527084686531942,
      "kl": 0.05401611328125,
      "learning_rate": 4.999814327607354e-07,
      "loss": 0.0001,
      "reward": 1.8196429163217545,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.8241071775555611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4558
    },
    {
      "completion_length": 235.0357255935669,
      "epoch": 0.7644913868980259,
      "grad_norm": 0.20327881479355508,
      "kl": 0.0784149169921875,
      "learning_rate": 4.999812862760358e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286059141159,
      "rewards/format_reward_func": 1.0,
      "step": 4560
    },
    {
      "completion_length": 246.8794765472412,
      "epoch": 0.7648266901378935,
      "grad_norm": 0.27998225865935156,
      "kl": 0.05908203125,
      "learning_rate": 4.999811392157866e-07,
      "loss": 0.0001,
      "reward": 1.7178572118282318,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7267857566475868,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4562
    },
    {
      "completion_length": 230.05804634094238,
      "epoch": 0.765161993377761,
      "grad_norm": 0.22216394342281265,
      "kl": 0.0649871826171875,
      "learning_rate": 4.999809915799882e-07,
      "loss": 0.0001,
      "reward": 1.776785783469677,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7812500335276127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4564
    },
    {
      "completion_length": 238.883939743042,
      "epoch": 0.7654972966176286,
      "grad_norm": 0.31322171637204343,
      "kl": 0.060089111328125,
      "learning_rate": 4.999808433686408e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 4566
    },
    {
      "completion_length": 233.1607265472412,
      "epoch": 0.7658325998574961,
      "grad_norm": 0.2450290221357226,
      "kl": 0.0533905029296875,
      "learning_rate": 4.999806945817447e-07,
      "loss": 0.0001,
      "reward": 1.7910714820027351,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7955357506871223,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4568
    },
    {
      "completion_length": 241.15626049041748,
      "epoch": 0.7661679030973637,
      "grad_norm": 0.189232939152715,
      "kl": 0.050811767578125,
      "learning_rate": 4.999805452193006e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 4570
    },
    {
      "completion_length": 234.21429634094238,
      "epoch": 0.7665032063372312,
      "grad_norm": 0.2246744520966242,
      "kl": 0.05870819091796875,
      "learning_rate": 4.999803952813084e-07,
      "loss": 0.0001,
      "reward": 1.6964286491274834,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6964286137372255,
      "rewards/format_reward_func": 1.0,
      "step": 4572
    },
    {
      "completion_length": 232.60268783569336,
      "epoch": 0.7668385095770988,
      "grad_norm": 0.3987041317522369,
      "kl": 0.05547332763671875,
      "learning_rate": 4.999802447677688e-07,
      "loss": 0.0001,
      "reward": 1.8250000476837158,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8250000216066837,
      "rewards/format_reward_func": 1.0,
      "step": 4574
    },
    {
      "completion_length": 233.89733123779297,
      "epoch": 0.7671738128169664,
      "grad_norm": 0.285725683188748,
      "kl": 0.0704803466796875,
      "learning_rate": 4.999800936786818e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 4576
    },
    {
      "completion_length": 226.71429538726807,
      "epoch": 0.7675091160568339,
      "grad_norm": 0.13415930213725472,
      "kl": 0.0756378173828125,
      "learning_rate": 4.999799420140481e-07,
      "loss": 0.0001,
      "reward": 1.7410714998841286,
      "reward_std": 0.0328299580141902,
      "rewards/equation_reward_func": 0.745535746216774,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4578
    },
    {
      "completion_length": 223.12054824829102,
      "epoch": 0.7678444192967014,
      "grad_norm": 0.26553482124859545,
      "kl": 0.0814056396484375,
      "learning_rate": 4.999797897738679e-07,
      "loss": 0.0001,
      "reward": 1.8071429058909416,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428835391998,
      "rewards/format_reward_func": 1.0,
      "step": 4580
    },
    {
      "completion_length": 227.18750858306885,
      "epoch": 0.768179722536569,
      "grad_norm": 0.25025503420441153,
      "kl": 0.0826568603515625,
      "learning_rate": 4.999796369581414e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000223517418,
      "rewards/format_reward_func": 1.0,
      "step": 4582
    },
    {
      "completion_length": 227.69643783569336,
      "epoch": 0.7685150257764366,
      "grad_norm": 0.24404022688359478,
      "kl": 0.0773773193359375,
      "learning_rate": 4.999794835668692e-07,
      "loss": 0.0001,
      "reward": 1.7571429163217545,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 4584
    },
    {
      "completion_length": 230.09375953674316,
      "epoch": 0.7688503290163041,
      "grad_norm": 0.29632477108847594,
      "kl": 0.0727691650390625,
      "learning_rate": 4.999793296000515e-07,
      "loss": 0.0001,
      "reward": 1.8071429058909416,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428947150707,
      "rewards/format_reward_func": 1.0,
      "step": 4586
    },
    {
      "completion_length": 227.6205472946167,
      "epoch": 0.7691856322561716,
      "grad_norm": 0.13950592949953058,
      "kl": 0.0792388916015625,
      "learning_rate": 4.999791750576887e-07,
      "loss": 0.0001,
      "reward": 1.7464286461472511,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286126196384,
      "rewards/format_reward_func": 1.0,
      "step": 4588
    },
    {
      "completion_length": 226.1071548461914,
      "epoch": 0.7695209354960393,
      "grad_norm": 0.16795992322820116,
      "kl": 0.0785980224609375,
      "learning_rate": 4.999790199397813e-07,
      "loss": 0.0001,
      "reward": 1.7892857566475868,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857436090708,
      "rewards/format_reward_func": 1.0,
      "step": 4590
    },
    {
      "completion_length": 233.9732255935669,
      "epoch": 0.7698562387359068,
      "grad_norm": 0.2591469522626041,
      "kl": 0.0863189697265625,
      "learning_rate": 4.999788642463293e-07,
      "loss": 0.0001,
      "reward": 1.7678571939468384,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 4592
    },
    {
      "completion_length": 230.99108219146729,
      "epoch": 0.7701915419757743,
      "grad_norm": 0.18624617499016574,
      "kl": 0.083465576171875,
      "learning_rate": 4.999787079773333e-07,
      "loss": 0.0001,
      "reward": 1.710714377462864,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7107143178582191,
      "rewards/format_reward_func": 1.0,
      "step": 4594
    },
    {
      "completion_length": 226.89733123779297,
      "epoch": 0.7705268452156419,
      "grad_norm": 0.32854213372203656,
      "kl": 0.06622314453125,
      "learning_rate": 4.999785511327936e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 4596
    },
    {
      "completion_length": 228.39733123779297,
      "epoch": 0.7708621484555095,
      "grad_norm": 0.27153695010869333,
      "kl": 0.080352783203125,
      "learning_rate": 4.999783937127107e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857544124126,
      "rewards/format_reward_func": 1.0,
      "step": 4598
    },
    {
      "completion_length": 229.54018878936768,
      "epoch": 0.771197451695377,
      "grad_norm": 0.20648485712931536,
      "kl": 0.068572998046875,
      "learning_rate": 4.999782357170849e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571734577417,
      "rewards/format_reward_func": 1.0,
      "step": 4600
    },
    {
      "completion_length": 221.12947463989258,
      "epoch": 0.7715327549352445,
      "grad_norm": 0.2005184739035029,
      "kl": 0.070556640625,
      "learning_rate": 4.999780771459164e-07,
      "loss": 0.0001,
      "reward": 1.814285770058632,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.814285758882761,
      "rewards/format_reward_func": 1.0,
      "step": 4602
    },
    {
      "completion_length": 238.4553689956665,
      "epoch": 0.7718680581751122,
      "grad_norm": 0.31052239666651466,
      "kl": 0.065032958984375,
      "learning_rate": 4.999779179992057e-07,
      "loss": 0.0001,
      "reward": 1.7808036282658577,
      "reward_std": 0.037249373737722635,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 4604
    },
    {
      "completion_length": 238.0089406967163,
      "epoch": 0.7722033614149797,
      "grad_norm": 0.1775028297389513,
      "kl": 0.071075439453125,
      "learning_rate": 4.999777582769532e-07,
      "loss": 0.0001,
      "reward": 1.7107143849134445,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7107143178582191,
      "rewards/format_reward_func": 1.0,
      "step": 4606
    },
    {
      "completion_length": 240.54465293884277,
      "epoch": 0.7725386646548472,
      "grad_norm": 0.4550589791379301,
      "kl": 0.0648345947265625,
      "learning_rate": 4.999775979791591e-07,
      "loss": 0.0001,
      "reward": 1.7071429565548897,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7071429006755352,
      "rewards/format_reward_func": 1.0,
      "step": 4608
    },
    {
      "completion_length": 235.96875953674316,
      "epoch": 0.7728739678947147,
      "grad_norm": 0.2854396965216469,
      "kl": 0.067779541015625,
      "learning_rate": 4.999774371058239e-07,
      "loss": 0.0001,
      "reward": 1.7392857819795609,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857596278191,
      "rewards/format_reward_func": 1.0,
      "step": 4610
    },
    {
      "completion_length": 233.13393878936768,
      "epoch": 0.7732092711345824,
      "grad_norm": 0.3207019572513585,
      "kl": 0.085296630859375,
      "learning_rate": 4.999772756569482e-07,
      "loss": 0.0001,
      "reward": 1.725000075995922,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7250000573694706,
      "rewards/format_reward_func": 1.0,
      "step": 4612
    },
    {
      "completion_length": 236.45983219146729,
      "epoch": 0.7735445743744499,
      "grad_norm": 0.3600593221437672,
      "kl": 0.0930633544921875,
      "learning_rate": 4.999771136325318e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 4614
    },
    {
      "completion_length": 242.47322463989258,
      "epoch": 0.7738798776143174,
      "grad_norm": 0.4511198912935247,
      "kl": 0.069305419921875,
      "learning_rate": 4.999769510325756e-07,
      "loss": 0.0001,
      "reward": 1.7446429282426834,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071838885546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4616
    },
    {
      "completion_length": 235.93304634094238,
      "epoch": 0.7742151808541851,
      "grad_norm": 0.26591008144932465,
      "kl": 0.0665130615234375,
      "learning_rate": 4.999767878570797e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571693599224,
      "rewards/format_reward_func": 1.0,
      "step": 4618
    },
    {
      "completion_length": 245.4062614440918,
      "epoch": 0.7745504840940526,
      "grad_norm": 0.2758105569697166,
      "kl": 0.062347412109375,
      "learning_rate": 4.999766241060446e-07,
      "loss": 0.0001,
      "reward": 1.7392858043313026,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857540398836,
      "rewards/format_reward_func": 1.0,
      "step": 4620
    },
    {
      "completion_length": 239.96429443359375,
      "epoch": 0.7748857873339201,
      "grad_norm": 0.1850762925480341,
      "kl": 0.0596771240234375,
      "learning_rate": 4.999764597794706e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 4622
    },
    {
      "completion_length": 233.30358219146729,
      "epoch": 0.7752210905737876,
      "grad_norm": 0.2923727962502333,
      "kl": 0.0526580810546875,
      "learning_rate": 4.999762948773581e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 4624
    },
    {
      "completion_length": 237.18750953674316,
      "epoch": 0.7755563938136553,
      "grad_norm": 0.29674990766493126,
      "kl": 0.0712127685546875,
      "learning_rate": 4.999761293997074e-07,
      "loss": 0.0001,
      "reward": 1.741071492433548,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357573926449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4626
    },
    {
      "completion_length": 243.34376335144043,
      "epoch": 0.7758916970535228,
      "grad_norm": 0.12933399523501038,
      "kl": 0.065521240234375,
      "learning_rate": 4.999759633465191e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857357859612,
      "rewards/format_reward_func": 1.0,
      "step": 4628
    },
    {
      "completion_length": 243.59375953674316,
      "epoch": 0.7762270002933903,
      "grad_norm": 0.3046586914770116,
      "kl": 0.0637664794921875,
      "learning_rate": 4.999757967177934e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 4630
    },
    {
      "completion_length": 242.4151906967163,
      "epoch": 0.7765623035332578,
      "grad_norm": 0.18912232830456813,
      "kl": 0.056549072265625,
      "learning_rate": 4.999756295135306e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 4632
    },
    {
      "completion_length": 251.13393783569336,
      "epoch": 0.7768976067731255,
      "grad_norm": 0.45607946377712877,
      "kl": 0.0615692138671875,
      "learning_rate": 4.999754617337315e-07,
      "loss": 0.0001,
      "reward": 1.7285715118050575,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7285714633762836,
      "rewards/format_reward_func": 1.0,
      "step": 4634
    },
    {
      "completion_length": 254.1919765472412,
      "epoch": 0.777232910012993,
      "grad_norm": 0.20415688926947373,
      "kl": 0.0582427978515625,
      "learning_rate": 4.99975293378396e-07,
      "loss": 0.0001,
      "reward": 1.7696429342031479,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7741071656346321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4636
    },
    {
      "completion_length": 258.8482265472412,
      "epoch": 0.7775682132528605,
      "grad_norm": 0.35244348616472587,
      "kl": 0.06280517578125,
      "learning_rate": 4.999751244475247e-07,
      "loss": 0.0001,
      "reward": 1.7375000715255737,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.741964302957058,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4638
    },
    {
      "completion_length": 242.8080472946167,
      "epoch": 0.7779035164927282,
      "grad_norm": 0.1982457603260857,
      "kl": 0.0550079345703125,
      "learning_rate": 4.99974954941118e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 4640
    },
    {
      "completion_length": 244.2142972946167,
      "epoch": 0.7782388197325957,
      "grad_norm": 0.31280826798701716,
      "kl": 0.0586395263671875,
      "learning_rate": 4.999747848591763e-07,
      "loss": 0.0001,
      "reward": 1.7750000804662704,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 4642
    },
    {
      "completion_length": 245.86608219146729,
      "epoch": 0.7785741229724632,
      "grad_norm": 0.19968901429705566,
      "kl": 0.08355712890625,
      "learning_rate": 4.999746142017e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571812808514,
      "rewards/format_reward_func": 1.0,
      "step": 4644
    },
    {
      "completion_length": 250.78126525878906,
      "epoch": 0.7789094262123307,
      "grad_norm": 0.305616407661564,
      "kl": 0.063568115234375,
      "learning_rate": 4.999744429686894e-07,
      "loss": 0.0001,
      "reward": 1.7875000685453415,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7919643148779869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4646
    },
    {
      "completion_length": 250.68304824829102,
      "epoch": 0.7792447294521984,
      "grad_norm": 0.20280940768286793,
      "kl": 0.0625,
      "learning_rate": 4.99974271160145e-07,
      "loss": 0.0001,
      "reward": 1.8071429058909416,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428872644901,
      "rewards/format_reward_func": 1.0,
      "step": 4648
    },
    {
      "completion_length": 249.08929920196533,
      "epoch": 0.7795800326920659,
      "grad_norm": 0.26340091091539314,
      "kl": 0.0640411376953125,
      "learning_rate": 4.999740987760671e-07,
      "loss": 0.0001,
      "reward": 1.7504464983940125,
      "reward_std": 0.05997780757024884,
      "rewards/equation_reward_func": 0.7651786133646965,
      "rewards/format_reward_func": 0.9852678664028645,
      "step": 4650
    },
    {
      "completion_length": 248.95090579986572,
      "epoch": 0.7799153359319334,
      "grad_norm": 0.3629256277036554,
      "kl": 0.0641937255859375,
      "learning_rate": 4.999739258164561e-07,
      "loss": 0.0001,
      "reward": 1.7892857640981674,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 4652
    },
    {
      "completion_length": 243.58929538726807,
      "epoch": 0.780250639171801,
      "grad_norm": 0.1749645065771544,
      "kl": 0.0629425048828125,
      "learning_rate": 4.999737522813124e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571589291096,
      "rewards/format_reward_func": 1.0,
      "step": 4654
    },
    {
      "completion_length": 250.03125953674316,
      "epoch": 0.7805859424116686,
      "grad_norm": 0.5312740210004152,
      "kl": 0.0813446044921875,
      "learning_rate": 4.999735781706365e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7910714615136385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4656
    },
    {
      "completion_length": 248.3616189956665,
      "epoch": 0.7809212456515361,
      "grad_norm": 0.1763252818665863,
      "kl": 0.0592498779296875,
      "learning_rate": 4.999734034844289e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 4658
    },
    {
      "completion_length": 248.52233028411865,
      "epoch": 0.7812565488914036,
      "grad_norm": 0.27396564397788564,
      "kl": 0.080841064453125,
      "learning_rate": 4.999732282226896e-07,
      "loss": 0.0001,
      "reward": 1.7160715237259865,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7205357477068901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4660
    },
    {
      "completion_length": 249.57590579986572,
      "epoch": 0.7815918521312712,
      "grad_norm": 0.29187870403616006,
      "kl": 0.067626953125,
      "learning_rate": 4.999730523854194e-07,
      "loss": 0.0001,
      "reward": 1.7500000596046448,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7589286006987095,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4662
    },
    {
      "completion_length": 244.07143878936768,
      "epoch": 0.7819271553711388,
      "grad_norm": 0.2820194000485553,
      "kl": 0.0643157958984375,
      "learning_rate": 4.999728759726185e-07,
      "loss": 0.0001,
      "reward": 1.708928644657135,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7133928909897804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4664
    },
    {
      "completion_length": 246.00001049041748,
      "epoch": 0.7822624586110063,
      "grad_norm": 0.21389940795165543,
      "kl": 0.0571746826171875,
      "learning_rate": 4.999726989842874e-07,
      "loss": 0.0001,
      "reward": 1.8267857655882835,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.8312500230967999,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4666
    },
    {
      "completion_length": 250.70536994934082,
      "epoch": 0.7825977618508739,
      "grad_norm": 0.20641362672396715,
      "kl": 0.0699310302734375,
      "learning_rate": 4.999725214204263e-07,
      "loss": 0.0001,
      "reward": 1.7589286416769028,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7633928842842579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4668
    },
    {
      "completion_length": 249.31697368621826,
      "epoch": 0.7829330650907415,
      "grad_norm": 0.25385673762727295,
      "kl": 0.059173583984375,
      "learning_rate": 4.999723432810359e-07,
      "loss": 0.0001,
      "reward": 1.764285795390606,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 4670
    },
    {
      "completion_length": 246.61608505249023,
      "epoch": 0.783268368330609,
      "grad_norm": 0.2611840826725265,
      "kl": 0.0884552001953125,
      "learning_rate": 4.999721645661165e-07,
      "loss": 0.0001,
      "reward": 1.773214340209961,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7776785977184772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4672
    },
    {
      "completion_length": 250.42411994934082,
      "epoch": 0.7836036715704765,
      "grad_norm": 0.6915758002266459,
      "kl": 0.102630615234375,
      "learning_rate": 4.999719852756685e-07,
      "loss": 0.0001,
      "reward": 1.7857143729925156,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143022119999,
      "rewards/format_reward_func": 1.0,
      "step": 4674
    },
    {
      "completion_length": 244.9375114440918,
      "epoch": 0.7839389748103441,
      "grad_norm": 0.2730890094350734,
      "kl": 0.0650482177734375,
      "learning_rate": 4.999718054096922e-07,
      "loss": 0.0001,
      "reward": 1.7580358013510704,
      "reward_std": 0.059346460737288,
      "rewards/equation_reward_func": 0.7598214522004128,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 4676
    },
    {
      "completion_length": 252.31251430511475,
      "epoch": 0.7842742780502117,
      "grad_norm": 0.21485735246991344,
      "kl": 0.05242919921875,
      "learning_rate": 4.999716249681883e-07,
      "loss": 0.0001,
      "reward": 1.814285770058632,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8142857439815998,
      "rewards/format_reward_func": 1.0,
      "step": 4678
    },
    {
      "completion_length": 242.25447463989258,
      "epoch": 0.7846095812900792,
      "grad_norm": 0.20691538510174362,
      "kl": 0.0582122802734375,
      "learning_rate": 4.999714439511568e-07,
      "loss": 0.0001,
      "reward": 1.725000075995922,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7250000275671482,
      "rewards/format_reward_func": 1.0,
      "step": 4680
    },
    {
      "completion_length": 253.27679824829102,
      "epoch": 0.7849448845299468,
      "grad_norm": 0.21287063557356575,
      "kl": 0.05328369140625,
      "learning_rate": 4.999712623585985e-07,
      "loss": 0.0001,
      "reward": 1.7767857760190964,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7812500335276127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4682
    },
    {
      "completion_length": 261.8303689956665,
      "epoch": 0.7852801877698143,
      "grad_norm": 0.3163254069693305,
      "kl": 0.052947998046875,
      "learning_rate": 4.999710801905137e-07,
      "loss": 0.0001,
      "reward": 1.7553571835160255,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214522004128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4684
    },
    {
      "completion_length": 255.2634048461914,
      "epoch": 0.7856154910096819,
      "grad_norm": 0.13242133409732526,
      "kl": 0.04907989501953125,
      "learning_rate": 4.999708974469028e-07,
      "loss": 0.0,
      "reward": 1.7357143685221672,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7446428760886192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4686
    },
    {
      "completion_length": 261.0491189956665,
      "epoch": 0.7859507942495494,
      "grad_norm": 0.30653420990937486,
      "kl": 0.0510711669921875,
      "learning_rate": 4.999707141277662e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7321428842842579,
      "rewards/format_reward_func": 1.0,
      "step": 4688
    },
    {
      "completion_length": 248.20090007781982,
      "epoch": 0.786286097489417,
      "grad_norm": 0.1144669042372897,
      "kl": 0.04682159423828125,
      "learning_rate": 4.999705302331042e-07,
      "loss": 0.0,
      "reward": 1.7392857819795609,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 1.0,
      "step": 4690
    },
    {
      "completion_length": 249.48215579986572,
      "epoch": 0.7866214007292845,
      "grad_norm": 0.10345151057283293,
      "kl": 0.0513916015625,
      "learning_rate": 4.999703457629175e-07,
      "loss": 0.0001,
      "reward": 1.832142911851406,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8321428708732128,
      "rewards/format_reward_func": 1.0,
      "step": 4692
    },
    {
      "completion_length": 250.82590293884277,
      "epoch": 0.7869567039691521,
      "grad_norm": 0.3815717918348069,
      "kl": 0.056793212890625,
      "learning_rate": 4.999701607172063e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143126428127,
      "rewards/format_reward_func": 1.0,
      "step": 4694
    },
    {
      "completion_length": 244.66072750091553,
      "epoch": 0.7872920072090197,
      "grad_norm": 0.21475385548445075,
      "kl": 0.06305694580078125,
      "learning_rate": 4.999699750959712e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 4696
    },
    {
      "completion_length": 244.7321548461914,
      "epoch": 0.7876273104488872,
      "grad_norm": 0.30145265431663126,
      "kl": 0.0600738525390625,
      "learning_rate": 4.999697888992124e-07,
      "loss": 0.0001,
      "reward": 1.739732213318348,
      "reward_std": 0.05745242489501834,
      "rewards/equation_reward_func": 0.7446428835391998,
      "rewards/format_reward_func": 0.9950892888009548,
      "step": 4698
    },
    {
      "completion_length": 248.17411708831787,
      "epoch": 0.7879626136887548,
      "grad_norm": 0.167633043860595,
      "kl": 0.04900360107421875,
      "learning_rate": 4.999696021269305e-07,
      "loss": 0.0,
      "reward": 1.7285714969038963,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7285714615136385,
      "rewards/format_reward_func": 1.0,
      "step": 4700
    },
    {
      "completion_length": 244.21429538726807,
      "epoch": 0.7882979169286223,
      "grad_norm": 0.18431015600634554,
      "kl": 0.069671630859375,
      "learning_rate": 4.99969414779126e-07,
      "loss": 0.0001,
      "reward": 1.7928571850061417,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 4702
    },
    {
      "completion_length": 243.78126049041748,
      "epoch": 0.7886332201684899,
      "grad_norm": 0.15342521106660034,
      "kl": 0.0619964599609375,
      "learning_rate": 4.999692268557992e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 4704
    },
    {
      "completion_length": 244.00001335144043,
      "epoch": 0.7889685234083574,
      "grad_norm": 0.2383847607881836,
      "kl": 0.06097412109375,
      "learning_rate": 4.999690383569505e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 4706
    },
    {
      "completion_length": 243.5759038925171,
      "epoch": 0.789303826648225,
      "grad_norm": 0.3709918619220796,
      "kl": 0.05507659912109375,
      "learning_rate": 4.999688492825803e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.09596449043601751,
      "rewards/equation_reward_func": 0.7839285992085934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4708
    },
    {
      "completion_length": 237.4776906967163,
      "epoch": 0.7896391298880926,
      "grad_norm": 0.13338077156805117,
      "kl": 0.04788970947265625,
      "learning_rate": 4.999686596326894e-07,
      "loss": 0.0,
      "reward": 1.7803572118282318,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7848214358091354,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4710
    },
    {
      "completion_length": 250.5625123977661,
      "epoch": 0.7899744331279601,
      "grad_norm": 0.34551931468587194,
      "kl": 0.0578155517578125,
      "learning_rate": 4.999684694072776e-07,
      "loss": 0.0001,
      "reward": 1.7714286521077156,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7714285980910063,
      "rewards/format_reward_func": 1.0,
      "step": 4712
    },
    {
      "completion_length": 241.7053680419922,
      "epoch": 0.7903097363678276,
      "grad_norm": 0.24656622735147873,
      "kl": 0.0606536865234375,
      "learning_rate": 4.99968278606346e-07,
      "loss": 0.0001,
      "reward": 1.7750000804662704,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 4714
    },
    {
      "completion_length": 242.2321548461914,
      "epoch": 0.7906450396076952,
      "grad_norm": 0.2908306029397776,
      "kl": 0.06658172607421875,
      "learning_rate": 4.999680872298946e-07,
      "loss": 0.0001,
      "reward": 1.7267858013510704,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7312500365078449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4716
    },
    {
      "completion_length": 235.7767972946167,
      "epoch": 0.7909803428475628,
      "grad_norm": 0.1948066984295881,
      "kl": 0.0508880615234375,
      "learning_rate": 4.999678952779241e-07,
      "loss": 0.0001,
      "reward": 1.760714367032051,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 4718
    },
    {
      "completion_length": 238.86608123779297,
      "epoch": 0.7913156460874303,
      "grad_norm": 0.22120546745892436,
      "kl": 0.0527191162109375,
      "learning_rate": 4.999677027504347e-07,
      "loss": 0.0001,
      "reward": 1.8267857804894447,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8312500230967999,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4720
    },
    {
      "completion_length": 245.20983123779297,
      "epoch": 0.7916509493272978,
      "grad_norm": 0.1871303555582509,
      "kl": 0.05560302734375,
      "learning_rate": 4.99967509647427e-07,
      "loss": 0.0001,
      "reward": 1.7178572043776512,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7178571671247482,
      "rewards/format_reward_func": 1.0,
      "step": 4722
    },
    {
      "completion_length": 238.77233028411865,
      "epoch": 0.7919862525671655,
      "grad_norm": 0.4842632365108479,
      "kl": 0.0672454833984375,
      "learning_rate": 4.999673159689015e-07,
      "loss": 0.0001,
      "reward": 1.7803571745753288,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7848214618861675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4724
    },
    {
      "completion_length": 243.75000953674316,
      "epoch": 0.792321555807033,
      "grad_norm": 0.2471484051407418,
      "kl": 0.0733184814453125,
      "learning_rate": 4.999671217148585e-07,
      "loss": 0.0001,
      "reward": 1.7428572177886963,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571619093418,
      "rewards/format_reward_func": 1.0,
      "step": 4726
    },
    {
      "completion_length": 239.40180015563965,
      "epoch": 0.7926568590469005,
      "grad_norm": 0.21497333312573835,
      "kl": 0.0543670654296875,
      "learning_rate": 4.999669268852984e-07,
      "loss": 0.0001,
      "reward": 1.7375000715255737,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643066823483,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4728
    },
    {
      "completion_length": 247.93304634094238,
      "epoch": 0.792992162286768,
      "grad_norm": 0.1317931786223707,
      "kl": 0.05374908447265625,
      "learning_rate": 4.999667314802218e-07,
      "loss": 0.0001,
      "reward": 1.753571517765522,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 4730
    },
    {
      "completion_length": 240.290189743042,
      "epoch": 0.7933274655266357,
      "grad_norm": 0.2581755180585634,
      "kl": 0.05332183837890625,
      "learning_rate": 4.999665354996292e-07,
      "loss": 0.0001,
      "reward": 1.796428643167019,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 4732
    },
    {
      "completion_length": 239.62054824829102,
      "epoch": 0.7936627687665032,
      "grad_norm": 0.20370485342766165,
      "kl": 0.0496673583984375,
      "learning_rate": 4.99966338943521e-07,
      "loss": 0.0,
      "reward": 1.7785714864730835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 4734
    },
    {
      "completion_length": 242.45537090301514,
      "epoch": 0.7939980720063707,
      "grad_norm": 0.29268332541299974,
      "kl": 0.0518035888671875,
      "learning_rate": 4.999661418118975e-07,
      "loss": 0.0001,
      "reward": 1.7714286148548126,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 4736
    },
    {
      "completion_length": 244.91518783569336,
      "epoch": 0.7943333752462384,
      "grad_norm": 0.16324022350176343,
      "kl": 0.090301513671875,
      "learning_rate": 4.999659441047592e-07,
      "loss": 0.0001,
      "reward": 1.787500061094761,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7919643074274063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4738
    },
    {
      "completion_length": 246.04465675354004,
      "epoch": 0.7946686784861059,
      "grad_norm": 0.1255347734918534,
      "kl": 0.04901123046875,
      "learning_rate": 4.999657458221067e-07,
      "loss": 0.0,
      "reward": 1.7642857879400253,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857357859612,
      "rewards/format_reward_func": 1.0,
      "step": 4740
    },
    {
      "completion_length": 249.23215579986572,
      "epoch": 0.7950039817259734,
      "grad_norm": 0.16224730747256555,
      "kl": 0.0569610595703125,
      "learning_rate": 4.999655469639404e-07,
      "loss": 0.0001,
      "reward": 1.753571480512619,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714767873287,
      "rewards/format_reward_func": 1.0,
      "step": 4742
    },
    {
      "completion_length": 238.4955472946167,
      "epoch": 0.7953392849658409,
      "grad_norm": 0.18577863111781406,
      "kl": 0.05908203125,
      "learning_rate": 4.999653475302607e-07,
      "loss": 0.0001,
      "reward": 1.8232143223285675,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.8276786021888256,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4744
    },
    {
      "completion_length": 243.20090293884277,
      "epoch": 0.7956745882057086,
      "grad_norm": 0.21887712276758306,
      "kl": 0.047943115234375,
      "learning_rate": 4.99965147521068e-07,
      "loss": 0.0,
      "reward": 1.8214285895228386,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8214286155998707,
      "rewards/format_reward_func": 1.0,
      "step": 4746
    },
    {
      "completion_length": 247.2455472946167,
      "epoch": 0.7960098914455761,
      "grad_norm": 0.3119421742863889,
      "kl": 0.0528106689453125,
      "learning_rate": 4.999649469363629e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 4748
    },
    {
      "completion_length": 239.9419755935669,
      "epoch": 0.7963451946854436,
      "grad_norm": 0.28155752485961605,
      "kl": 0.05231475830078125,
      "learning_rate": 4.999647457761459e-07,
      "loss": 0.0001,
      "reward": 1.8178571984171867,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8178571611642838,
      "rewards/format_reward_func": 1.0,
      "step": 4750
    },
    {
      "completion_length": 252.48661708831787,
      "epoch": 0.7966804979253111,
      "grad_norm": 0.3151087780616481,
      "kl": 0.04834747314453125,
      "learning_rate": 4.999645440404173e-07,
      "loss": 0.0,
      "reward": 1.767857201397419,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.767857164144516,
      "rewards/format_reward_func": 1.0,
      "step": 4752
    },
    {
      "completion_length": 241.24554824829102,
      "epoch": 0.7970158011651788,
      "grad_norm": 0.2119678596279154,
      "kl": 0.057403564453125,
      "learning_rate": 4.999643417291776e-07,
      "loss": 0.0001,
      "reward": 1.8250000551342964,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8250000141561031,
      "rewards/format_reward_func": 1.0,
      "step": 4754
    },
    {
      "completion_length": 250.9196548461914,
      "epoch": 0.7973511044050463,
      "grad_norm": 0.19341988481809425,
      "kl": 0.0551300048828125,
      "learning_rate": 4.999641388424274e-07,
      "loss": 0.0001,
      "reward": 1.719642922282219,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7241071723401546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4756
    },
    {
      "completion_length": 244.50447368621826,
      "epoch": 0.7976864076449138,
      "grad_norm": 0.18595105296353032,
      "kl": 0.06219482421875,
      "learning_rate": 4.99963935380167e-07,
      "loss": 0.0001,
      "reward": 1.7482143491506577,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526785936206579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4758
    },
    {
      "completion_length": 248.3348331451416,
      "epoch": 0.7980217108847815,
      "grad_norm": 0.2563888180930464,
      "kl": 0.053497314453125,
      "learning_rate": 4.99963731342397e-07,
      "loss": 0.0001,
      "reward": 1.821428619325161,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8214285895228386,
      "rewards/format_reward_func": 1.0,
      "step": 4760
    },
    {
      "completion_length": 254.7500114440918,
      "epoch": 0.798357014124649,
      "grad_norm": 0.22967847403831104,
      "kl": 0.05162811279296875,
      "learning_rate": 4.999635267291178e-07,
      "loss": 0.0001,
      "reward": 1.7428571954369545,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7428571786731482,
      "rewards/format_reward_func": 1.0,
      "step": 4762
    },
    {
      "completion_length": 251.24554538726807,
      "epoch": 0.7986923173645165,
      "grad_norm": 0.18666718271717592,
      "kl": 0.0740814208984375,
      "learning_rate": 4.999633215403298e-07,
      "loss": 0.0001,
      "reward": 1.7500000521540642,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 4764
    },
    {
      "completion_length": 254.54911518096924,
      "epoch": 0.799027620604384,
      "grad_norm": 0.20649343618390373,
      "kl": 0.0608673095703125,
      "learning_rate": 4.999631157760337e-07,
      "loss": 0.0001,
      "reward": 1.717857226729393,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7178571857511997,
      "rewards/format_reward_func": 1.0,
      "step": 4766
    },
    {
      "completion_length": 258.2901916503906,
      "epoch": 0.7993629238442517,
      "grad_norm": 0.21921901486791798,
      "kl": 0.0643157958984375,
      "learning_rate": 4.999629094362298e-07,
      "loss": 0.0001,
      "reward": 1.735714390873909,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143089175224,
      "rewards/format_reward_func": 1.0,
      "step": 4768
    },
    {
      "completion_length": 249.4464406967163,
      "epoch": 0.7996982270841192,
      "grad_norm": 0.2943164089246413,
      "kl": 0.05615234375,
      "learning_rate": 4.999627025209186e-07,
      "loss": 0.0001,
      "reward": 1.7723214775323868,
      "reward_std": 0.06944798585027456,
      "rewards/equation_reward_func": 0.7741071917116642,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 4770
    },
    {
      "completion_length": 256.4866180419922,
      "epoch": 0.8000335303239867,
      "grad_norm": 0.17801495941107487,
      "kl": 0.057403564453125,
      "learning_rate": 4.999624950301005e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 4772
    },
    {
      "completion_length": 253.21876430511475,
      "epoch": 0.8003688335638544,
      "grad_norm": 0.22709532160911014,
      "kl": 0.0803985595703125,
      "learning_rate": 4.999622869637761e-07,
      "loss": 0.0001,
      "reward": 1.8285714909434319,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8285714574158192,
      "rewards/format_reward_func": 1.0,
      "step": 4774
    },
    {
      "completion_length": 255.7321548461914,
      "epoch": 0.8007041368037219,
      "grad_norm": 0.43890108257018107,
      "kl": 0.06640625,
      "learning_rate": 4.999620783219457e-07,
      "loss": 0.0001,
      "reward": 1.817857213318348,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.817857164889574,
      "rewards/format_reward_func": 1.0,
      "step": 4776
    },
    {
      "completion_length": 249.24108505249023,
      "epoch": 0.8010394400435894,
      "grad_norm": 0.21523694406650426,
      "kl": 0.060211181640625,
      "learning_rate": 4.999618691046101e-07,
      "loss": 0.0001,
      "reward": 1.8160715103149414,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8205357380211353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4778
    },
    {
      "completion_length": 259.96875953674316,
      "epoch": 0.8013747432834569,
      "grad_norm": 0.30152421281682196,
      "kl": 0.062408447265625,
      "learning_rate": 4.999616593117696e-07,
      "loss": 0.0001,
      "reward": 1.7446429282426834,
      "reward_std": 0.10859139636158943,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4780
    },
    {
      "completion_length": 248.758939743042,
      "epoch": 0.8017100465233246,
      "grad_norm": 0.18162921628223075,
      "kl": 0.05597686767578125,
      "learning_rate": 4.999614489434246e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 4782
    },
    {
      "completion_length": 249.38840198516846,
      "epoch": 0.8020453497631921,
      "grad_norm": 0.18843137481806166,
      "kl": 0.0561370849609375,
      "learning_rate": 4.999612379995757e-07,
      "loss": 0.0001,
      "reward": 1.7928571850061417,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 4784
    },
    {
      "completion_length": 253.75447750091553,
      "epoch": 0.8023806530030596,
      "grad_norm": 0.11907621852180425,
      "kl": 0.060516357421875,
      "learning_rate": 4.999610264802233e-07,
      "loss": 0.0001,
      "reward": 1.6660715192556381,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.6705357432365417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4786
    },
    {
      "completion_length": 261.1696586608887,
      "epoch": 0.8027159562429272,
      "grad_norm": 0.18805736926606073,
      "kl": 0.064910888671875,
      "learning_rate": 4.999608143853679e-07,
      "loss": 0.0001,
      "reward": 1.807142898440361,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428872644901,
      "rewards/format_reward_func": 1.0,
      "step": 4788
    },
    {
      "completion_length": 248.7276906967163,
      "epoch": 0.8030512594827948,
      "grad_norm": 0.21305119555940016,
      "kl": 0.05750274658203125,
      "learning_rate": 4.999606017150102e-07,
      "loss": 0.0001,
      "reward": 1.7750000730156898,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 4790
    },
    {
      "completion_length": 259.43305015563965,
      "epoch": 0.8033865627226623,
      "grad_norm": 0.21317110179286103,
      "kl": 0.05792236328125,
      "learning_rate": 4.999603884691504e-07,
      "loss": 0.0001,
      "reward": 1.753571517765522,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 4792
    },
    {
      "completion_length": 252.7232265472412,
      "epoch": 0.8037218659625298,
      "grad_norm": 0.15468881031021706,
      "kl": 0.0545654296875,
      "learning_rate": 4.999601746477891e-07,
      "loss": 0.0001,
      "reward": 1.74642863124609,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464285995811224,
      "rewards/format_reward_func": 1.0,
      "step": 4794
    },
    {
      "completion_length": 270.4776916503906,
      "epoch": 0.8040571692023974,
      "grad_norm": 0.18228137072443312,
      "kl": 0.0901336669921875,
      "learning_rate": 4.999599602509269e-07,
      "loss": 0.0001,
      "reward": 1.7321429252624512,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321429029107094,
      "rewards/format_reward_func": 1.0,
      "step": 4796
    },
    {
      "completion_length": 268.3571529388428,
      "epoch": 0.804392472442265,
      "grad_norm": 0.21559347957903086,
      "kl": 0.0575103759765625,
      "learning_rate": 4.999597452785641e-07,
      "loss": 0.0001,
      "reward": 1.7464286461472511,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7553571723401546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4798
    },
    {
      "completion_length": 265.1384048461914,
      "epoch": 0.8047277756821325,
      "grad_norm": 0.2657425103639842,
      "kl": 0.158416748046875,
      "learning_rate": 4.999595297307014e-07,
      "loss": 0.0002,
      "reward": 1.757142923772335,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 4800
    },
    {
      "completion_length": 274.2276916503906,
      "epoch": 0.8050630789220001,
      "grad_norm": 0.1548226820890832,
      "kl": 0.0614166259765625,
      "learning_rate": 4.99959313607339e-07,
      "loss": 0.0001,
      "reward": 1.7482143566012383,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526786103844643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4802
    },
    {
      "completion_length": 279.4017963409424,
      "epoch": 0.8053983821618677,
      "grad_norm": 0.2341381706367825,
      "kl": 0.0644378662109375,
      "learning_rate": 4.999590969084777e-07,
      "loss": 0.0001,
      "reward": 1.7625000551342964,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4804
    },
    {
      "completion_length": 270.4687604904175,
      "epoch": 0.8057336854017352,
      "grad_norm": 0.22770723007849342,
      "kl": 0.088226318359375,
      "learning_rate": 4.999588796341178e-07,
      "loss": 0.0001,
      "reward": 1.7625000551342964,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.766964316368103,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4806
    },
    {
      "completion_length": 277.86162090301514,
      "epoch": 0.8060689886416027,
      "grad_norm": 0.3219488111395403,
      "kl": 0.069610595703125,
      "learning_rate": 4.999586617842599e-07,
      "loss": 0.0001,
      "reward": 1.7607143446803093,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7696428894996643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4808
    },
    {
      "completion_length": 267.16965770721436,
      "epoch": 0.8064042918814703,
      "grad_norm": 0.21817359942228975,
      "kl": 0.0725860595703125,
      "learning_rate": 4.999584433589046e-07,
      "loss": 0.0001,
      "reward": 1.7196428924798965,
      "reward_std": 0.0833375845104456,
      "rewards/equation_reward_func": 0.7330357562750578,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 4810
    },
    {
      "completion_length": 275.1696548461914,
      "epoch": 0.8067395951213379,
      "grad_norm": 0.17800857048787785,
      "kl": 0.075469970703125,
      "learning_rate": 4.999582243580522e-07,
      "loss": 0.0001,
      "reward": 1.7446429431438446,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7491071671247482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4812
    },
    {
      "completion_length": 274.85268688201904,
      "epoch": 0.8070748983612054,
      "grad_norm": 0.24170594435818452,
      "kl": 0.0898284912109375,
      "learning_rate": 4.999580047817033e-07,
      "loss": 0.0001,
      "reward": 1.7696429342031479,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7830357402563095,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 4814
    },
    {
      "completion_length": 270.9776916503906,
      "epoch": 0.807410201601073,
      "grad_norm": 0.24363523164547604,
      "kl": 0.1192169189453125,
      "learning_rate": 4.999577846298584e-07,
      "loss": 0.0001,
      "reward": 1.7589286267757416,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7633928880095482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4816
    },
    {
      "completion_length": 281.633939743042,
      "epoch": 0.8077455048409405,
      "grad_norm": 0.2132089822493617,
      "kl": 0.0901947021484375,
      "learning_rate": 4.999575639025179e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 4818
    },
    {
      "completion_length": 272.4642972946167,
      "epoch": 0.8080808080808081,
      "grad_norm": 0.1495708958499177,
      "kl": 0.2392578125,
      "learning_rate": 4.999573425996826e-07,
      "loss": 0.0002,
      "reward": 1.7589286342263222,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7633928842842579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4820
    },
    {
      "completion_length": 281.1071548461914,
      "epoch": 0.8084161113206756,
      "grad_norm": 0.18808619582281788,
      "kl": 0.0755767822265625,
      "learning_rate": 4.999571207213527e-07,
      "loss": 0.0001,
      "reward": 1.819642886519432,
      "reward_std": 0.0530330091714859,
      "rewards/equation_reward_func": 0.8330357410013676,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 4822
    },
    {
      "completion_length": 276.33930110931396,
      "epoch": 0.8087514145605432,
      "grad_norm": 0.25074459221996265,
      "kl": 0.1247100830078125,
      "learning_rate": 4.999568982675288e-07,
      "loss": 0.0001,
      "reward": 1.7839286178350449,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7883928939700127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4824
    },
    {
      "completion_length": 268.6116180419922,
      "epoch": 0.8090867178004107,
      "grad_norm": 0.2757940434856731,
      "kl": 0.079986572265625,
      "learning_rate": 4.999566752382115e-07,
      "loss": 0.0001,
      "reward": 1.7714286670088768,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 4826
    },
    {
      "completion_length": 271.4196538925171,
      "epoch": 0.8094220210402783,
      "grad_norm": 0.37402314358434524,
      "kl": 0.088958740234375,
      "learning_rate": 4.999564516334014e-07,
      "loss": 0.0001,
      "reward": 1.7375000640749931,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7419643178582191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4828
    },
    {
      "completion_length": 282.1160821914673,
      "epoch": 0.8097573242801459,
      "grad_norm": 0.2814170427276152,
      "kl": 0.0948486328125,
      "learning_rate": 4.999562274530986e-07,
      "loss": 0.0001,
      "reward": 1.7714286521077156,
      "reward_std": 0.09091372694820166,
      "rewards/equation_reward_func": 0.7803571671247482,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4830
    },
    {
      "completion_length": 274.2634038925171,
      "epoch": 0.8100926275200134,
      "grad_norm": 0.45118148911042716,
      "kl": 0.0821533203125,
      "learning_rate": 4.999560026973041e-07,
      "loss": 0.0001,
      "reward": 1.7000000700354576,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7178571671247482,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 4832
    },
    {
      "completion_length": 275.20537281036377,
      "epoch": 0.810427930759881,
      "grad_norm": 0.18307441024953733,
      "kl": 0.08013916015625,
      "learning_rate": 4.999557773660181e-07,
      "loss": 0.0001,
      "reward": 1.7339286133646965,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7383928932249546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4834
    },
    {
      "completion_length": 276.81250953674316,
      "epoch": 0.8107632339997485,
      "grad_norm": 0.2023363492109368,
      "kl": 0.0673065185546875,
      "learning_rate": 4.999555514592412e-07,
      "loss": 0.0001,
      "reward": 1.716071479022503,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7205357365310192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4836
    },
    {
      "completion_length": 270.35715770721436,
      "epoch": 0.8110985372396161,
      "grad_norm": 0.21590173948082855,
      "kl": 0.079620361328125,
      "learning_rate": 4.999553249769741e-07,
      "loss": 0.0001,
      "reward": 1.807142898440361,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8071428816765547,
      "rewards/format_reward_func": 1.0,
      "step": 4838
    },
    {
      "completion_length": 276.65179920196533,
      "epoch": 0.8114338404794836,
      "grad_norm": 0.18878714466772603,
      "kl": 0.0695648193359375,
      "learning_rate": 4.999550979192169e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 4840
    },
    {
      "completion_length": 285.02680015563965,
      "epoch": 0.8117691437193512,
      "grad_norm": 0.1801698236755662,
      "kl": 0.0682830810546875,
      "learning_rate": 4.999548702859706e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7625000178813934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4842
    },
    {
      "completion_length": 266.7946529388428,
      "epoch": 0.8121044469592188,
      "grad_norm": 0.29530471967628835,
      "kl": 0.06537628173828125,
      "learning_rate": 4.999546420772355e-07,
      "loss": 0.0001,
      "reward": 1.7910714745521545,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7955357357859612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4844
    },
    {
      "completion_length": 277.7946548461914,
      "epoch": 0.8124397501990863,
      "grad_norm": 0.2510459183216338,
      "kl": 0.0666961669921875,
      "learning_rate": 4.99954413293012e-07,
      "loss": 0.0001,
      "reward": 1.7553572058677673,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7598214745521545,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4846
    },
    {
      "completion_length": 283.46876335144043,
      "epoch": 0.8127750534389538,
      "grad_norm": 0.1536660170745747,
      "kl": 0.0618133544921875,
      "learning_rate": 4.999541839333009e-07,
      "loss": 0.0001,
      "reward": 1.7178572192788124,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7267857454717159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4848
    },
    {
      "completion_length": 272.60715675354004,
      "epoch": 0.8131103566788214,
      "grad_norm": 0.2226667586865844,
      "kl": 0.070068359375,
      "learning_rate": 4.999539539981026e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571604192257,
      "rewards/format_reward_func": 1.0,
      "step": 4850
    },
    {
      "completion_length": 261.90179443359375,
      "epoch": 0.813445659918689,
      "grad_norm": 0.3051079389978345,
      "kl": 0.0652008056640625,
      "learning_rate": 4.999537234874175e-07,
      "loss": 0.0001,
      "reward": 1.8178572058677673,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8178571574389935,
      "rewards/format_reward_func": 1.0,
      "step": 4852
    },
    {
      "completion_length": 270.1562662124634,
      "epoch": 0.8137809631585565,
      "grad_norm": 0.1828016501904028,
      "kl": 0.062744140625,
      "learning_rate": 4.999534924012463e-07,
      "loss": 0.0001,
      "reward": 1.7589286118745804,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7633928805589676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4854
    },
    {
      "completion_length": 269.808048248291,
      "epoch": 0.814116266398424,
      "grad_norm": 0.320133616246475,
      "kl": 0.075286865234375,
      "learning_rate": 4.999532607395895e-07,
      "loss": 0.0001,
      "reward": 1.742857240140438,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571693599224,
      "rewards/format_reward_func": 1.0,
      "step": 4856
    },
    {
      "completion_length": 275.1205472946167,
      "epoch": 0.8144515696382917,
      "grad_norm": 0.16699386169014177,
      "kl": 0.0585174560546875,
      "learning_rate": 4.999530285024477e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 4858
    },
    {
      "completion_length": 272.2544765472412,
      "epoch": 0.8147868728781592,
      "grad_norm": 0.40172270093922924,
      "kl": 0.0598907470703125,
      "learning_rate": 4.999527956898213e-07,
      "loss": 0.0001,
      "reward": 1.7196429446339607,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.7330357432365417,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 4860
    },
    {
      "completion_length": 266.4866189956665,
      "epoch": 0.8151221761180267,
      "grad_norm": 0.37233589266703304,
      "kl": 0.065093994140625,
      "learning_rate": 4.999525623017109e-07,
      "loss": 0.0001,
      "reward": 1.8000000640749931,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.8089285809546709,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4862
    },
    {
      "completion_length": 267.90179347991943,
      "epoch": 0.8154574793578943,
      "grad_norm": 0.30506025414719584,
      "kl": 0.062347412109375,
      "learning_rate": 4.99952328338117e-07,
      "loss": 0.0001,
      "reward": 1.776785783469677,
      "reward_std": 0.09343910776078701,
      "rewards/equation_reward_func": 0.7812500298023224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4864
    },
    {
      "completion_length": 259.2142963409424,
      "epoch": 0.8157927825977619,
      "grad_norm": 0.12008669658374085,
      "kl": 0.062469482421875,
      "learning_rate": 4.999520937990401e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.792857151478529,
      "rewards/format_reward_func": 1.0,
      "step": 4866
    },
    {
      "completion_length": 255.54465198516846,
      "epoch": 0.8161280858376294,
      "grad_norm": 0.19604347972133174,
      "kl": 0.0647430419921875,
      "learning_rate": 4.999518586844809e-07,
      "loss": 0.0001,
      "reward": 1.7696429416537285,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4868
    },
    {
      "completion_length": 258.7634048461914,
      "epoch": 0.8164633890774969,
      "grad_norm": 0.31025431871878384,
      "kl": 0.06878662109375,
      "learning_rate": 4.999516229944397e-07,
      "loss": 0.0001,
      "reward": 1.7142857983708382,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7142857387661934,
      "rewards/format_reward_func": 1.0,
      "step": 4870
    },
    {
      "completion_length": 258.1473340988159,
      "epoch": 0.8167986923173646,
      "grad_norm": 0.059445911244861434,
      "kl": 0.0688018798828125,
      "learning_rate": 4.999513867289173e-07,
      "loss": 0.0001,
      "reward": 1.8107143342494965,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8107143212109804,
      "rewards/format_reward_func": 1.0,
      "step": 4872
    },
    {
      "completion_length": 252.93304920196533,
      "epoch": 0.8171339955572321,
      "grad_norm": 0.0881423719531909,
      "kl": 0.0708465576171875,
      "learning_rate": 4.999511498879142e-07,
      "loss": 0.0001,
      "reward": 1.7736607939004898,
      "reward_std": 0.013258251827210188,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 4874
    },
    {
      "completion_length": 256.0223321914673,
      "epoch": 0.8174692987970996,
      "grad_norm": 0.2124047604525655,
      "kl": 0.071258544921875,
      "learning_rate": 4.999509124714308e-07,
      "loss": 0.0001,
      "reward": 1.753571480512619,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.753571480512619,
      "rewards/format_reward_func": 1.0,
      "step": 4876
    },
    {
      "completion_length": 256.4375114440918,
      "epoch": 0.8178046020369671,
      "grad_norm": 0.18623706897044853,
      "kl": 0.0633697509765625,
      "learning_rate": 4.999506744794677e-07,
      "loss": 0.0001,
      "reward": 1.7678571864962578,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.776785746216774,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4878
    },
    {
      "completion_length": 255.67411994934082,
      "epoch": 0.8181399052768348,
      "grad_norm": 0.15097665520310913,
      "kl": 0.0752410888671875,
      "learning_rate": 4.999504359120255e-07,
      "loss": 0.0001,
      "reward": 1.7415179386734962,
      "reward_std": 0.032198611879721284,
      "rewards/equation_reward_func": 0.7428571730852127,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 4880
    },
    {
      "completion_length": 273.29019260406494,
      "epoch": 0.8184752085167023,
      "grad_norm": 0.26742352819239673,
      "kl": 0.0709228515625,
      "learning_rate": 4.999501967691048e-07,
      "loss": 0.0001,
      "reward": 1.6964286714792252,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7053571790456772,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4882
    },
    {
      "completion_length": 261.84376430511475,
      "epoch": 0.8188105117565698,
      "grad_norm": 0.25184865004850154,
      "kl": 0.0738372802734375,
      "learning_rate": 4.99949957050706e-07,
      "loss": 0.0001,
      "reward": 1.7290179282426834,
      "reward_std": 0.07007933035492897,
      "rewards/equation_reward_func": 0.7348214648663998,
      "rewards/format_reward_func": 0.994196429848671,
      "step": 4884
    },
    {
      "completion_length": 253.09375953674316,
      "epoch": 0.8191458149964373,
      "grad_norm": 0.21782260571475182,
      "kl": 0.079376220703125,
      "learning_rate": 4.999497167568297e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857372760773,
      "rewards/format_reward_func": 1.0,
      "step": 4886
    },
    {
      "completion_length": 253.2812614440918,
      "epoch": 0.819481118236305,
      "grad_norm": 0.16804666737366383,
      "kl": 0.0702362060546875,
      "learning_rate": 4.999494758874765e-07,
      "loss": 0.0001,
      "reward": 1.7642857655882835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857413738966,
      "rewards/format_reward_func": 1.0,
      "step": 4888
    },
    {
      "completion_length": 265.95983600616455,
      "epoch": 0.8198164214761725,
      "grad_norm": 0.09307030417656371,
      "kl": 0.0602874755859375,
      "learning_rate": 4.999492344426469e-07,
      "loss": 0.0001,
      "reward": 1.733928620815277,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7383928932249546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4890
    },
    {
      "completion_length": 259.1428699493408,
      "epoch": 0.82015172471604,
      "grad_norm": 0.25695249055536407,
      "kl": 0.0602569580078125,
      "learning_rate": 4.999489924223416e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143055647612,
      "rewards/format_reward_func": 1.0,
      "step": 4892
    },
    {
      "completion_length": 269.8214387893677,
      "epoch": 0.8204870279559077,
      "grad_norm": 0.19842259243217908,
      "kl": 0.0663299560546875,
      "learning_rate": 4.999487498265609e-07,
      "loss": 0.0001,
      "reward": 1.7410714849829674,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.745535746216774,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4894
    },
    {
      "completion_length": 257.30804443359375,
      "epoch": 0.8208223311957752,
      "grad_norm": 0.2123599581687482,
      "kl": 0.071533203125,
      "learning_rate": 4.999485066553056e-07,
      "loss": 0.0001,
      "reward": 1.764285795390606,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857283353806,
      "rewards/format_reward_func": 1.0,
      "step": 4896
    },
    {
      "completion_length": 270.36608028411865,
      "epoch": 0.8211576344356427,
      "grad_norm": 0.1736771649562477,
      "kl": 0.060089111328125,
      "learning_rate": 4.999482629085762e-07,
      "loss": 0.0001,
      "reward": 1.7892858013510704,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7982143089175224,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4898
    },
    {
      "completion_length": 257.4955463409424,
      "epoch": 0.8214929376755102,
      "grad_norm": 0.20787241332400688,
      "kl": 0.0594024658203125,
      "learning_rate": 4.99948018586373e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7803571820259094,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4900
    },
    {
      "completion_length": 259.64287090301514,
      "epoch": 0.8218282409153779,
      "grad_norm": 0.23193187383227143,
      "kl": 0.0570831298828125,
      "learning_rate": 4.99947773688697e-07,
      "loss": 0.0001,
      "reward": 1.7660714834928513,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4902
    },
    {
      "completion_length": 260.96429920196533,
      "epoch": 0.8221635441552454,
      "grad_norm": 0.17487735272584332,
      "kl": 0.05246734619140625,
      "learning_rate": 4.999475282155485e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428921073675,
      "rewards/format_reward_func": 1.0,
      "step": 4904
    },
    {
      "completion_length": 261.5446557998657,
      "epoch": 0.8224988473951129,
      "grad_norm": 0.27319235143934706,
      "kl": 0.0552215576171875,
      "learning_rate": 4.999472821669281e-07,
      "loss": 0.0001,
      "reward": 1.821428619325161,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8214285969734192,
      "rewards/format_reward_func": 1.0,
      "step": 4906
    },
    {
      "completion_length": 271.3348379135132,
      "epoch": 0.8228341506349806,
      "grad_norm": 0.2630989065803111,
      "kl": 0.0597381591796875,
      "learning_rate": 4.999470355428364e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7678571939468384,
      "rewards/format_reward_func": 1.0,
      "step": 4908
    },
    {
      "completion_length": 273.5089406967163,
      "epoch": 0.8231694538748481,
      "grad_norm": 0.17175984680368336,
      "kl": 0.0602264404296875,
      "learning_rate": 4.999467883432739e-07,
      "loss": 0.0001,
      "reward": 1.6875000819563866,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.6919643208384514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4910
    },
    {
      "completion_length": 261.35715675354004,
      "epoch": 0.8235047571147156,
      "grad_norm": 0.2206713915711781,
      "kl": 0.0681610107421875,
      "learning_rate": 4.999465405682414e-07,
      "loss": 0.0001,
      "reward": 1.733928620815277,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.7473214603960514,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 4912
    },
    {
      "completion_length": 277.81697368621826,
      "epoch": 0.8238400603545831,
      "grad_norm": 0.2808354461978824,
      "kl": 0.0636749267578125,
      "learning_rate": 4.999462922177391e-07,
      "loss": 0.0001,
      "reward": 1.7464286759495735,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 4914
    },
    {
      "completion_length": 270.495548248291,
      "epoch": 0.8241753635944508,
      "grad_norm": 0.37773820919936024,
      "kl": 0.0789031982421875,
      "learning_rate": 4.999460432917678e-07,
      "loss": 0.0001,
      "reward": 1.7357143685221672,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7446428909897804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4916
    },
    {
      "completion_length": 269.8794746398926,
      "epoch": 0.8245106668343183,
      "grad_norm": 0.26859032997452337,
      "kl": 0.0596160888671875,
      "learning_rate": 4.999457937903281e-07,
      "loss": 0.0001,
      "reward": 1.6767858043313026,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.6901786029338837,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 4918
    },
    {
      "completion_length": 268.6384048461914,
      "epoch": 0.8248459700741858,
      "grad_norm": 0.22619369001240797,
      "kl": 0.0674896240234375,
      "learning_rate": 4.999455437134205e-07,
      "loss": 0.0001,
      "reward": 1.7750000432133675,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 4920
    },
    {
      "completion_length": 273.4375162124634,
      "epoch": 0.8251812733140534,
      "grad_norm": 0.14572012351895092,
      "kl": 0.064910888671875,
      "learning_rate": 4.999452930610455e-07,
      "loss": 0.0001,
      "reward": 1.7107143625617027,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7107143178582191,
      "rewards/format_reward_func": 1.0,
      "step": 4922
    },
    {
      "completion_length": 273.55804347991943,
      "epoch": 0.825516576553921,
      "grad_norm": 0.19885069207259493,
      "kl": 0.0724029541015625,
      "learning_rate": 4.999450418332038e-07,
      "loss": 0.0001,
      "reward": 1.737500086426735,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419642992317677,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4924
    },
    {
      "completion_length": 267.4598321914673,
      "epoch": 0.8258518797937885,
      "grad_norm": 0.20614664708216593,
      "kl": 0.1008758544921875,
      "learning_rate": 4.99944790029896e-07,
      "loss": 0.0001,
      "reward": 1.7696429193019867,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7741071730852127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4926
    },
    {
      "completion_length": 270.4241199493408,
      "epoch": 0.826187183033656,
      "grad_norm": 0.39868854843655405,
      "kl": 0.0659942626953125,
      "learning_rate": 4.999445376511225e-07,
      "loss": 0.0001,
      "reward": 1.8125000596046448,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.8169643133878708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4928
    },
    {
      "completion_length": 273.05804920196533,
      "epoch": 0.8265224862735236,
      "grad_norm": 0.44631029880094397,
      "kl": 0.047637939453125,
      "learning_rate": 4.999442846968841e-07,
      "loss": 0.0,
      "reward": 1.7982143238186836,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8026786036789417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4930
    },
    {
      "completion_length": 264.0848321914673,
      "epoch": 0.8268577895133912,
      "grad_norm": 0.16727886038983347,
      "kl": 0.0599212646484375,
      "learning_rate": 4.999440311671812e-07,
      "loss": 0.0001,
      "reward": 1.8464286103844643,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8464285917580128,
      "rewards/format_reward_func": 1.0,
      "step": 4932
    },
    {
      "completion_length": 264.13840675354004,
      "epoch": 0.8271930927532587,
      "grad_norm": 0.29495113787224825,
      "kl": 0.0543212890625,
      "learning_rate": 4.999437770620146e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7535714469850063,
      "rewards/format_reward_func": 1.0,
      "step": 4934
    },
    {
      "completion_length": 258.9285831451416,
      "epoch": 0.8275283959931263,
      "grad_norm": 0.22670554901695683,
      "kl": 0.0487060546875,
      "learning_rate": 4.999435223813847e-07,
      "loss": 0.0,
      "reward": 1.8250000476837158,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.8339285813271999,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4936
    },
    {
      "completion_length": 267.41965770721436,
      "epoch": 0.8278636992329939,
      "grad_norm": 0.26918688987822686,
      "kl": 0.05322265625,
      "learning_rate": 4.999432671252921e-07,
      "loss": 0.0001,
      "reward": 1.7375000715255737,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7419643104076385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4938
    },
    {
      "completion_length": 263.3214416503906,
      "epoch": 0.8281990024728614,
      "grad_norm": 0.21402795920358994,
      "kl": 0.059417724609375,
      "learning_rate": 4.999430112937374e-07,
      "loss": 0.0001,
      "reward": 1.8071429133415222,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8071428835391998,
      "rewards/format_reward_func": 1.0,
      "step": 4940
    },
    {
      "completion_length": 271.71430110931396,
      "epoch": 0.8285343057127289,
      "grad_norm": 0.7274596587412323,
      "kl": 0.0613861083984375,
      "learning_rate": 4.999427548867214e-07,
      "loss": 0.0001,
      "reward": 1.689285784959793,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.689285758882761,
      "rewards/format_reward_func": 1.0,
      "step": 4942
    },
    {
      "completion_length": 242.38840103149414,
      "epoch": 0.8288696089525965,
      "grad_norm": 0.196112449776157,
      "kl": 0.0539093017578125,
      "learning_rate": 4.999424979042443e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428626775742,
      "rewards/format_reward_func": 1.0,
      "step": 4944
    },
    {
      "completion_length": 268.79465675354004,
      "epoch": 0.8292049121924641,
      "grad_norm": 0.14449989796392068,
      "kl": 0.06292724609375,
      "learning_rate": 4.99942240346307e-07,
      "loss": 0.0001,
      "reward": 1.7392857819795609,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7482143193483353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4946
    },
    {
      "completion_length": 259.68751430511475,
      "epoch": 0.8295402154323316,
      "grad_norm": 0.19207518609316115,
      "kl": 0.05278778076171875,
      "learning_rate": 4.999419822129099e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 4948
    },
    {
      "completion_length": 269.71429538726807,
      "epoch": 0.8298755186721992,
      "grad_norm": 0.19097486547502238,
      "kl": 0.15569305419921875,
      "learning_rate": 4.999417235040538e-07,
      "loss": 0.0002,
      "reward": 1.7214286401867867,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7392857428640127,
      "rewards/format_reward_func": 0.9821428619325161,
      "step": 4950
    },
    {
      "completion_length": 260.8214406967163,
      "epoch": 0.8302108219120667,
      "grad_norm": 0.15030844373911773,
      "kl": 0.088470458984375,
      "learning_rate": 4.99941464219739e-07,
      "loss": 0.0001,
      "reward": 1.7446429207921028,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7491071838885546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4952
    },
    {
      "completion_length": 257.60715770721436,
      "epoch": 0.8305461251519343,
      "grad_norm": 0.2758773487892338,
      "kl": 0.069061279296875,
      "learning_rate": 4.999412043599665e-07,
      "loss": 0.0001,
      "reward": 1.68571437895298,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.6937500294297934,
      "rewards/format_reward_func": 0.9919642880558968,
      "step": 4954
    },
    {
      "completion_length": 262.9910821914673,
      "epoch": 0.8308814283918018,
      "grad_norm": 0.21574039383575827,
      "kl": 0.05950927734375,
      "learning_rate": 4.999409439247366e-07,
      "loss": 0.0001,
      "reward": 1.7915179207921028,
      "reward_std": 0.04230013699270785,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 4956
    },
    {
      "completion_length": 257.2946548461914,
      "epoch": 0.8312167316316694,
      "grad_norm": 0.19333692041023248,
      "kl": 0.0700531005859375,
      "learning_rate": 4.999406829140499e-07,
      "loss": 0.0001,
      "reward": 1.7428571954369545,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7517857514321804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4958
    },
    {
      "completion_length": 258.95983505249023,
      "epoch": 0.831552034871537,
      "grad_norm": 0.28450642542702165,
      "kl": 0.1026763916015625,
      "learning_rate": 4.999404213279072e-07,
      "loss": 0.0001,
      "reward": 1.7107143551111221,
      "reward_std": 0.09091372601687908,
      "rewards/equation_reward_func": 0.7196428775787354,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4960
    },
    {
      "completion_length": 258.48662185668945,
      "epoch": 0.8318873381114045,
      "grad_norm": 0.11666481478020153,
      "kl": 0.0623321533203125,
      "learning_rate": 4.999401591663088e-07,
      "loss": 0.0001,
      "reward": 1.7964286357164383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285910129547,
      "rewards/format_reward_func": 1.0,
      "step": 4962
    },
    {
      "completion_length": 261.14287090301514,
      "epoch": 0.8322226413512721,
      "grad_norm": 0.2595529695459257,
      "kl": 0.0776519775390625,
      "learning_rate": 4.999398964292556e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7464286033064127,
      "rewards/format_reward_func": 1.0,
      "step": 4964
    },
    {
      "completion_length": 255.86161422729492,
      "epoch": 0.8325579445911396,
      "grad_norm": 0.11477647603067251,
      "kl": 0.05771636962890625,
      "learning_rate": 4.999396331167481e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 4966
    },
    {
      "completion_length": 251.63840866088867,
      "epoch": 0.8328932478310072,
      "grad_norm": 0.3354314841872632,
      "kl": 0.0924072265625,
      "learning_rate": 4.999393692287868e-07,
      "loss": 0.0001,
      "reward": 1.7589286267757416,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7633928880095482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4968
    },
    {
      "completion_length": 259.92858505249023,
      "epoch": 0.8332285510708747,
      "grad_norm": 0.22319074901580088,
      "kl": 0.05889892578125,
      "learning_rate": 4.999391047653726e-07,
      "loss": 0.0001,
      "reward": 1.73214291036129,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428898721933,
      "rewards/format_reward_func": 1.0,
      "step": 4970
    },
    {
      "completion_length": 253.16965579986572,
      "epoch": 0.8335638543107423,
      "grad_norm": 0.20776365236196978,
      "kl": 0.0605621337890625,
      "learning_rate": 4.999388397265057e-07,
      "loss": 0.0001,
      "reward": 1.7750000730156898,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 4972
    },
    {
      "completion_length": 258.81250953674316,
      "epoch": 0.8338991575506098,
      "grad_norm": 0.3116271721626666,
      "kl": 0.0717926025390625,
      "learning_rate": 4.999385741121871e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 4974
    },
    {
      "completion_length": 257.9732246398926,
      "epoch": 0.8342344607904774,
      "grad_norm": 0.16783852777353048,
      "kl": 0.0492401123046875,
      "learning_rate": 4.999383079224171e-07,
      "loss": 0.0,
      "reward": 1.760714367032051,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 4976
    },
    {
      "completion_length": 267.3125123977661,
      "epoch": 0.834569764030345,
      "grad_norm": 0.2939545154505324,
      "kl": 0.1884918212890625,
      "learning_rate": 4.999380411571965e-07,
      "loss": 0.0002,
      "reward": 1.7214286178350449,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7303571756929159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4978
    },
    {
      "completion_length": 271.2857255935669,
      "epoch": 0.8349050672702125,
      "grad_norm": 0.13913934639957815,
      "kl": 0.26397705078125,
      "learning_rate": 4.999377738165259e-07,
      "loss": 0.0003,
      "reward": 1.74642863124609,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7553571723401546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4980
    },
    {
      "completion_length": 258.9553680419922,
      "epoch": 0.83524037051008,
      "grad_norm": 0.2516402147403652,
      "kl": 0.0686798095703125,
      "learning_rate": 4.999375059004057e-07,
      "loss": 0.0001,
      "reward": 1.7446429207921028,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7491071820259094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4982
    },
    {
      "completion_length": 268.2321557998657,
      "epoch": 0.8355756737499476,
      "grad_norm": 0.2667209451787418,
      "kl": 0.1626129150390625,
      "learning_rate": 4.999372374088369e-07,
      "loss": 0.0002,
      "reward": 1.7803571969270706,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214488476515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4984
    },
    {
      "completion_length": 260.77233695983887,
      "epoch": 0.8359109769898152,
      "grad_norm": 0.26300885880997005,
      "kl": 0.1019744873046875,
      "learning_rate": 4.999369683418199e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 4986
    },
    {
      "completion_length": 259.6116199493408,
      "epoch": 0.8362462802296827,
      "grad_norm": 0.3083753439451682,
      "kl": 0.171356201171875,
      "learning_rate": 4.999366986993552e-07,
      "loss": 0.0002,
      "reward": 1.8321429267525673,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8321428745985031,
      "rewards/format_reward_func": 1.0,
      "step": 4988
    },
    {
      "completion_length": 278.30804920196533,
      "epoch": 0.8365815834695502,
      "grad_norm": 0.4166251158747486,
      "kl": 0.43697357177734375,
      "learning_rate": 4.999364284814435e-07,
      "loss": 0.0004,
      "reward": 1.753571480512619,
      "reward_std": 0.08586296625435352,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 4990
    },
    {
      "completion_length": 273.8884048461914,
      "epoch": 0.8369168867094179,
      "grad_norm": 0.22651730612716978,
      "kl": 0.0768585205078125,
      "learning_rate": 4.999361576880856e-07,
      "loss": 0.0001,
      "reward": 1.7178572416305542,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7267857380211353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 4992
    },
    {
      "completion_length": 265.2321548461914,
      "epoch": 0.8372521899492854,
      "grad_norm": 0.19169125415619304,
      "kl": 0.3283843994140625,
      "learning_rate": 4.99935886319282e-07,
      "loss": 0.0003,
      "reward": 1.8142857551574707,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8142857626080513,
      "rewards/format_reward_func": 1.0,
      "step": 4994
    },
    {
      "completion_length": 264.4375123977661,
      "epoch": 0.8375874931891529,
      "grad_norm": 0.14860434722498145,
      "kl": 0.04929351806640625,
      "learning_rate": 4.999356143750332e-07,
      "loss": 0.0,
      "reward": 1.7089286297559738,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.713392898440361,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4996
    },
    {
      "completion_length": 262.3259057998657,
      "epoch": 0.8379227964290205,
      "grad_norm": 0.3040842393844373,
      "kl": 0.06220245361328125,
      "learning_rate": 4.999353418553402e-07,
      "loss": 0.0001,
      "reward": 1.7946429178118706,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7991071734577417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 4998
    },
    {
      "completion_length": 256.0982255935669,
      "epoch": 0.8382580996688881,
      "grad_norm": 0.9815141329046756,
      "kl": 0.08681488037109375,
      "learning_rate": 4.999350687602031e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7928571850061417,
      "rewards/format_reward_func": 1.0,
      "step": 5000
    },
    {
      "completion_length": 257.6517963409424,
      "epoch": 0.8385934029087556,
      "grad_norm": 0.28596476806596116,
      "kl": 0.1295928955078125,
      "learning_rate": 4.99934795089623e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571566939354,
      "rewards/format_reward_func": 1.0,
      "step": 5002
    },
    {
      "completion_length": 254.4642972946167,
      "epoch": 0.8389287061486231,
      "grad_norm": 0.2751226439728887,
      "kl": 0.3716888427734375,
      "learning_rate": 4.999345208436002e-07,
      "loss": 0.0004,
      "reward": 1.7410714998841286,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7455357387661934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5004
    },
    {
      "completion_length": 264.1964406967163,
      "epoch": 0.8392640093884907,
      "grad_norm": 0.17527055408946596,
      "kl": 0.0655517578125,
      "learning_rate": 4.999342460221355e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 5006
    },
    {
      "completion_length": 268.5848340988159,
      "epoch": 0.8395993126283583,
      "grad_norm": 0.30189495346003126,
      "kl": 0.08837127685546875,
      "learning_rate": 4.999339706252295e-07,
      "loss": 0.0001,
      "reward": 1.741517961025238,
      "reward_std": 0.0688166399486363,
      "rewards/equation_reward_func": 0.7428571674972773,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 5008
    },
    {
      "completion_length": 271.37947940826416,
      "epoch": 0.8399346158682258,
      "grad_norm": 0.2656036792725558,
      "kl": 0.06268310546875,
      "learning_rate": 4.999336946528828e-07,
      "loss": 0.0001,
      "reward": 1.7696429342031479,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7741071805357933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5010
    },
    {
      "completion_length": 262.29019260406494,
      "epoch": 0.8402699191080933,
      "grad_norm": 0.2827278859733368,
      "kl": 0.06427001953125,
      "learning_rate": 4.999334181050961e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 5012
    },
    {
      "completion_length": 267.7767972946167,
      "epoch": 0.840605222347961,
      "grad_norm": 0.1459763848123801,
      "kl": 0.2068939208984375,
      "learning_rate": 4.9993314098187e-07,
      "loss": 0.0002,
      "reward": 1.7464286237955093,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 5014
    },
    {
      "completion_length": 260.0580472946167,
      "epoch": 0.8409405255878285,
      "grad_norm": 0.11177792675682865,
      "kl": 0.141387939453125,
      "learning_rate": 4.999328632832052e-07,
      "loss": 0.0001,
      "reward": 1.8321429044008255,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8321428783237934,
      "rewards/format_reward_func": 1.0,
      "step": 5016
    },
    {
      "completion_length": 269.8392963409424,
      "epoch": 0.841275828827696,
      "grad_norm": 0.2513538542042273,
      "kl": 0.0636138916015625,
      "learning_rate": 4.999325850091022e-07,
      "loss": 0.0001,
      "reward": 1.7767857685685158,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.7901785969734192,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5018
    },
    {
      "completion_length": 263.446439743042,
      "epoch": 0.8416111320675635,
      "grad_norm": 0.1607530545343153,
      "kl": 0.3533172607421875,
      "learning_rate": 4.999323061595617e-07,
      "loss": 0.0004,
      "reward": 1.7129465118050575,
      "reward_std": 0.0625031883828342,
      "rewards/equation_reward_func": 0.7232143208384514,
      "rewards/format_reward_func": 0.9897321499884129,
      "step": 5020
    },
    {
      "completion_length": 278.18304538726807,
      "epoch": 0.8419464353074312,
      "grad_norm": 0.7651100930666153,
      "kl": 2.2594146728515625,
      "learning_rate": 4.999320267345844e-07,
      "loss": 0.0023,
      "reward": 1.707589365541935,
      "reward_std": 0.09028238267637789,
      "rewards/equation_reward_func": 0.7223214656114578,
      "rewards/format_reward_func": 0.9852678664028645,
      "step": 5022
    },
    {
      "completion_length": 265.76786708831787,
      "epoch": 0.8422817385472987,
      "grad_norm": 0.5229851016677849,
      "kl": 0.22228240966796875,
      "learning_rate": 4.999317467341709e-07,
      "loss": 0.0002,
      "reward": 1.7272322103381157,
      "reward_std": 0.07765547605231404,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 0.980803582817316,
      "step": 5024
    },
    {
      "completion_length": 265.40626335144043,
      "epoch": 0.8426170417871662,
      "grad_norm": 0.4540552845227721,
      "kl": 0.666046142578125,
      "learning_rate": 4.99931466158322e-07,
      "loss": 0.0007,
      "reward": 1.7071429193019867,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7250000275671482,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5026
    },
    {
      "completion_length": 256.67411708831787,
      "epoch": 0.8429523450270339,
      "grad_norm": 0.2477658548989634,
      "kl": 0.38092041015625,
      "learning_rate": 4.99931185007038e-07,
      "loss": 0.0004,
      "reward": 1.7392857819795609,
      "reward_std": 0.07576144114136696,
      "rewards/equation_reward_func": 0.7482143230736256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5028
    },
    {
      "completion_length": 261.7321557998657,
      "epoch": 0.8432876482669014,
      "grad_norm": 294.4796959677068,
      "kl": 292.18275451660156,
      "learning_rate": 4.999309032803199e-07,
      "loss": 0.2905,
      "reward": 1.7553571984171867,
      "reward_std": 0.0833375845104456,
      "rewards/equation_reward_func": 0.7687500268220901,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5030
    },
    {
      "completion_length": 254.8660831451416,
      "epoch": 0.8436229515067689,
      "grad_norm": 0.3487734623831873,
      "kl": 0.2953338623046875,
      "learning_rate": 4.99930620978168e-07,
      "loss": 0.0003,
      "reward": 1.7625000849366188,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5032
    },
    {
      "completion_length": 250.8214406967163,
      "epoch": 0.8439582547466364,
      "grad_norm": 0.845100666891351,
      "kl": 1.3850021362304688,
      "learning_rate": 4.999303381005833e-07,
      "loss": 0.0014,
      "reward": 1.7571429312229156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.757142897695303,
      "rewards/format_reward_func": 1.0,
      "step": 5034
    },
    {
      "completion_length": 260.23662090301514,
      "epoch": 0.8442935579865041,
      "grad_norm": 0.29851671677387165,
      "kl": 0.46588134765625,
      "learning_rate": 4.999300546475663e-07,
      "loss": 0.0005,
      "reward": 1.7897321954369545,
      "reward_std": 0.09533314313739538,
      "rewards/equation_reward_func": 0.8044643066823483,
      "rewards/format_reward_func": 0.9852678626775742,
      "step": 5036
    },
    {
      "completion_length": 259.9821557998657,
      "epoch": 0.8446288612263716,
      "grad_norm": 0.19617775336306847,
      "kl": 0.188018798828125,
      "learning_rate": 4.999297706191175e-07,
      "loss": 0.0002,
      "reward": 1.739285796880722,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857447266579,
      "rewards/format_reward_func": 1.0,
      "step": 5038
    },
    {
      "completion_length": 259.6562623977661,
      "epoch": 0.8449641644662391,
      "grad_norm": 0.36903900589044275,
      "kl": 0.31150054931640625,
      "learning_rate": 4.999294860152378e-07,
      "loss": 0.0003,
      "reward": 1.7035715207457542,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7125000394880772,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5040
    },
    {
      "completion_length": 251.31697845458984,
      "epoch": 0.8452994677061068,
      "grad_norm": 0.34778111463721867,
      "kl": 0.58111572265625,
      "learning_rate": 4.999292008359277e-07,
      "loss": 0.0006,
      "reward": 1.7892857864499092,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.789285734295845,
      "rewards/format_reward_func": 1.0,
      "step": 5042
    },
    {
      "completion_length": 259.6517972946167,
      "epoch": 0.8456347709459743,
      "grad_norm": 0.22109059202099945,
      "kl": 0.05115509033203125,
      "learning_rate": 4.99928915081188e-07,
      "loss": 0.0001,
      "reward": 1.701785758137703,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.706250037997961,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5044
    },
    {
      "completion_length": 255.47322463989258,
      "epoch": 0.8459700741858418,
      "grad_norm": 0.767684363928975,
      "kl": 1.7428512573242188,
      "learning_rate": 4.999286287510192e-07,
      "loss": 0.0017,
      "reward": 1.7303572073578835,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7348214611411095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5046
    },
    {
      "completion_length": 247.4866180419922,
      "epoch": 0.8463053774257093,
      "grad_norm": 0.2620184516888097,
      "kl": 0.0875244140625,
      "learning_rate": 4.999283418454221e-07,
      "loss": 0.0001,
      "reward": 1.7696429044008255,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7741071842610836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5048
    },
    {
      "completion_length": 257.3928699493408,
      "epoch": 0.846640680665577,
      "grad_norm": 0.2051462165393263,
      "kl": 0.0811309814453125,
      "learning_rate": 4.999280543643973e-07,
      "loss": 0.0001,
      "reward": 1.7428572252392769,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571693599224,
      "rewards/format_reward_func": 1.0,
      "step": 5050
    },
    {
      "completion_length": 257.3348340988159,
      "epoch": 0.8469759839054445,
      "grad_norm": 0.23980940731317518,
      "kl": 0.0730438232421875,
      "learning_rate": 4.999277663079453e-07,
      "loss": 0.0001,
      "reward": 1.753571517765522,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 5052
    },
    {
      "completion_length": 256.02233505249023,
      "epoch": 0.847311287145312,
      "grad_norm": 0.2466820519649015,
      "kl": 0.0859832763671875,
      "learning_rate": 4.99927477676067e-07,
      "loss": 0.0001,
      "reward": 1.7446429282426834,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071708500385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5054
    },
    {
      "completion_length": 242.81697368621826,
      "epoch": 0.8476465903851796,
      "grad_norm": 0.19081753990918476,
      "kl": 0.04564666748046875,
      "learning_rate": 4.99927188468763e-07,
      "loss": 0.0,
      "reward": 1.7910714894533157,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7955357450991869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5056
    },
    {
      "completion_length": 250.4375114440918,
      "epoch": 0.8479818936250472,
      "grad_norm": 0.27938780172578787,
      "kl": 0.1655731201171875,
      "learning_rate": 4.99926898686034e-07,
      "loss": 0.0002,
      "reward": 1.7464286163449287,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.746428593993187,
      "rewards/format_reward_func": 1.0,
      "step": 5058
    },
    {
      "completion_length": 253.05804443359375,
      "epoch": 0.8483171968649147,
      "grad_norm": 0.5356787768614212,
      "kl": 0.06327056884765625,
      "learning_rate": 4.999266083278806e-07,
      "loss": 0.0001,
      "reward": 1.7446429058909416,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7491071820259094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5060
    },
    {
      "completion_length": 237.91965198516846,
      "epoch": 0.8486525001047822,
      "grad_norm": 0.24795477280872208,
      "kl": 0.11028289794921875,
      "learning_rate": 4.999263173943034e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285925030708,
      "rewards/format_reward_func": 1.0,
      "step": 5062
    },
    {
      "completion_length": 246.70983219146729,
      "epoch": 0.8489878033446498,
      "grad_norm": 0.2763103500220934,
      "kl": 0.081085205078125,
      "learning_rate": 4.999260258853032e-07,
      "loss": 0.0001,
      "reward": 1.8089286386966705,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.8133928813040257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5064
    },
    {
      "completion_length": 252.1830472946167,
      "epoch": 0.8493231065845174,
      "grad_norm": 0.10641035138334937,
      "kl": 0.04872894287109375,
      "learning_rate": 4.999257338008806e-07,
      "loss": 0.0,
      "reward": 1.7517857626080513,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7562500387430191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5066
    },
    {
      "completion_length": 255.8259038925171,
      "epoch": 0.8496584098243849,
      "grad_norm": 0.23307262313618435,
      "kl": 0.04785919189453125,
      "learning_rate": 4.999254411410363e-07,
      "loss": 0.0,
      "reward": 1.730357214808464,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7348214536905289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5068
    },
    {
      "completion_length": 244.45983219146729,
      "epoch": 0.8499937130642525,
      "grad_norm": 0.23193165760967088,
      "kl": 0.2115325927734375,
      "learning_rate": 4.99925147905771e-07,
      "loss": 0.0002,
      "reward": 1.785714328289032,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7946428805589676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5070
    },
    {
      "completion_length": 244.90626049041748,
      "epoch": 0.8503290163041201,
      "grad_norm": 0.24690095810292512,
      "kl": 0.053955078125,
      "learning_rate": 4.999248540950853e-07,
      "loss": 0.0001,
      "reward": 1.7785715162754059,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 5072
    },
    {
      "completion_length": 253.98661613464355,
      "epoch": 0.8506643195439876,
      "grad_norm": 0.23389187074428028,
      "kl": 0.1283416748046875,
      "learning_rate": 4.999245597089799e-07,
      "loss": 0.0001,
      "reward": 1.787500061094761,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7919643074274063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5074
    },
    {
      "completion_length": 247.71875762939453,
      "epoch": 0.8509996227838551,
      "grad_norm": 0.8323127056861116,
      "kl": 0.1199188232421875,
      "learning_rate": 4.999242647474555e-07,
      "loss": 0.0001,
      "reward": 1.7821428999304771,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7910714596509933,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5076
    },
    {
      "completion_length": 236.50000953674316,
      "epoch": 0.8513349260237227,
      "grad_norm": 0.18800976984029746,
      "kl": 0.1182861328125,
      "learning_rate": 4.99923969210513e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.753571467474103,
      "rewards/format_reward_func": 1.0,
      "step": 5078
    },
    {
      "completion_length": 239.78126049041748,
      "epoch": 0.8516702292635903,
      "grad_norm": 0.2201142403006137,
      "kl": 0.0549774169921875,
      "learning_rate": 4.999236730981526e-07,
      "loss": 0.0001,
      "reward": 1.714285783469677,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7142857536673546,
      "rewards/format_reward_func": 1.0,
      "step": 5080
    },
    {
      "completion_length": 247.9062623977661,
      "epoch": 0.8520055325034578,
      "grad_norm": 4.761260457058568,
      "kl": 0.383270263671875,
      "learning_rate": 4.999233764103753e-07,
      "loss": 0.0004,
      "reward": 1.753571480512619,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7625000402331352,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5082
    },
    {
      "completion_length": 240.20983409881592,
      "epoch": 0.8523408357433254,
      "grad_norm": 0.126562800778106,
      "kl": 0.554901123046875,
      "learning_rate": 4.999230791471818e-07,
      "loss": 0.0006,
      "reward": 1.7714286521077156,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 5084
    },
    {
      "completion_length": 230.77233409881592,
      "epoch": 0.8526761389831929,
      "grad_norm": 0.29796437906708106,
      "kl": 0.05060577392578125,
      "learning_rate": 4.999227813085725e-07,
      "loss": 0.0001,
      "reward": 1.8285714834928513,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8285714499652386,
      "rewards/format_reward_func": 1.0,
      "step": 5086
    },
    {
      "completion_length": 239.77679538726807,
      "epoch": 0.8530114422230605,
      "grad_norm": 0.30966121200378954,
      "kl": 0.2183380126953125,
      "learning_rate": 4.999224828945485e-07,
      "loss": 0.0002,
      "reward": 1.7946429029107094,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.799107164144516,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5088
    },
    {
      "completion_length": 244.41519165039062,
      "epoch": 0.853346745462928,
      "grad_norm": 0.14491462362379912,
      "kl": 0.11983489990234375,
      "learning_rate": 4.999221839051102e-07,
      "loss": 0.0001,
      "reward": 1.7375000789761543,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7419643215835094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5090
    },
    {
      "completion_length": 234.41965293884277,
      "epoch": 0.8536820487027956,
      "grad_norm": 0.24775391265290617,
      "kl": 0.1589813232421875,
      "learning_rate": 4.999218843402584e-07,
      "loss": 0.0002,
      "reward": 1.718303643167019,
      "reward_std": 0.0650285689625889,
      "rewards/equation_reward_func": 0.724107164889574,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 5092
    },
    {
      "completion_length": 236.27233123779297,
      "epoch": 0.8540173519426631,
      "grad_norm": 0.4428321684880137,
      "kl": 0.21246337890625,
      "learning_rate": 4.999215841999937e-07,
      "loss": 0.0002,
      "reward": 1.6910714730620384,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.7223214618861675,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 5094
    },
    {
      "completion_length": 236.79465293884277,
      "epoch": 0.8543526551825307,
      "grad_norm": 0.3418168828278188,
      "kl": 0.14962005615234375,
      "learning_rate": 4.999212834843169e-07,
      "loss": 0.0001,
      "reward": 1.798214353621006,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.8116071708500385,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5096
    },
    {
      "completion_length": 230.79911708831787,
      "epoch": 0.8546879584223983,
      "grad_norm": 0.20071501266811895,
      "kl": 0.1268768310546875,
      "learning_rate": 4.999209821932287e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 5098
    },
    {
      "completion_length": 226.96875858306885,
      "epoch": 0.8550232616622658,
      "grad_norm": 0.17248835077943814,
      "kl": 0.098907470703125,
      "learning_rate": 4.999206803267296e-07,
      "loss": 0.0001,
      "reward": 1.689285822212696,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.6982143241912127,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5100
    },
    {
      "completion_length": 228.93751049041748,
      "epoch": 0.8553585649021334,
      "grad_norm": 0.30204625045537503,
      "kl": 0.187408447265625,
      "learning_rate": 4.999203778848206e-07,
      "loss": 0.0002,
      "reward": 1.7410714775323868,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357480794191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5102
    },
    {
      "completion_length": 223.1071538925171,
      "epoch": 0.8556938681420009,
      "grad_norm": 0.783522810256943,
      "kl": 2.531463623046875,
      "learning_rate": 4.999200748675021e-07,
      "loss": 0.0025,
      "reward": 1.7250000908970833,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7339285984635353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5104
    },
    {
      "completion_length": 234.47768878936768,
      "epoch": 0.8560291713818685,
      "grad_norm": 0.24328087656232975,
      "kl": 0.229888916015625,
      "learning_rate": 4.99919771274775e-07,
      "loss": 0.0002,
      "reward": 1.7200893685221672,
      "reward_std": 0.06250318791717291,
      "rewards/equation_reward_func": 0.721428606659174,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 5106
    },
    {
      "completion_length": 244.42858219146729,
      "epoch": 0.856364474621736,
      "grad_norm": 0.29232745791069964,
      "kl": 1.5421829223632812,
      "learning_rate": 4.999194671066398e-07,
      "loss": 0.0015,
      "reward": 1.721428632736206,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7303571812808514,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5108
    },
    {
      "completion_length": 252.26340675354004,
      "epoch": 0.8566997778616036,
      "grad_norm": 0.4569750985390259,
      "kl": 1.1767578125,
      "learning_rate": 4.999191623630974e-07,
      "loss": 0.0012,
      "reward": 1.6625000834465027,
      "reward_std": 0.11364215891808271,
      "rewards/equation_reward_func": 0.6848214641213417,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 5110
    },
    {
      "completion_length": 247.93751430511475,
      "epoch": 0.8570350811014712,
      "grad_norm": 0.34339815751092734,
      "kl": 0.3735198974609375,
      "learning_rate": 4.999188570441485e-07,
      "loss": 0.0004,
      "reward": 1.7035714611411095,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.7214286047965288,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5112
    },
    {
      "completion_length": 228.1562614440918,
      "epoch": 0.8573703843413387,
      "grad_norm": 0.24380516996038454,
      "kl": 0.1014862060546875,
      "learning_rate": 4.999185511497937e-07,
      "loss": 0.0001,
      "reward": 1.8160714954137802,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.8205357380211353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5114
    },
    {
      "completion_length": 238.89286613464355,
      "epoch": 0.8577056875812062,
      "grad_norm": 0.2579426290533824,
      "kl": 0.203277587890625,
      "learning_rate": 4.999182446800336e-07,
      "loss": 0.0002,
      "reward": 1.732142947614193,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5116
    },
    {
      "completion_length": 229.06251049041748,
      "epoch": 0.8580409908210738,
      "grad_norm": 0.39017097066306555,
      "kl": 1.1303558349609375,
      "learning_rate": 4.999179376348691e-07,
      "loss": 0.0011,
      "reward": 1.7343750596046448,
      "reward_std": 0.08270624093711376,
      "rewards/equation_reward_func": 0.7535714525729418,
      "rewards/format_reward_func": 0.980803582817316,
      "step": 5118
    },
    {
      "completion_length": 226.82143878936768,
      "epoch": 0.8583762940609414,
      "grad_norm": 0.8587286627098762,
      "kl": 0.360626220703125,
      "learning_rate": 4.999176300143009e-07,
      "loss": 0.0004,
      "reward": 1.751785784959793,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7562500275671482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5120
    },
    {
      "completion_length": 229.28126049041748,
      "epoch": 0.8587115973008089,
      "grad_norm": 0.19873741773246498,
      "kl": 0.1075286865234375,
      "learning_rate": 4.999173218183296e-07,
      "loss": 0.0001,
      "reward": 1.776785783469677,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7812500186264515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5122
    },
    {
      "completion_length": 227.12054538726807,
      "epoch": 0.8590469005406765,
      "grad_norm": 0.2766724367610363,
      "kl": 0.2878875732421875,
      "learning_rate": 4.99917013046956e-07,
      "loss": 0.0003,
      "reward": 1.698214367032051,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7026786021888256,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5124
    },
    {
      "completion_length": 225.43750858306885,
      "epoch": 0.8593822037805441,
      "grad_norm": 0.45199430854316097,
      "kl": 0.27734375,
      "learning_rate": 4.999167037001807e-07,
      "loss": 0.0003,
      "reward": 1.7625000849366188,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.766964316368103,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5126
    },
    {
      "completion_length": 220.16965198516846,
      "epoch": 0.8597175070204116,
      "grad_norm": 0.1847114816856559,
      "kl": 0.318359375,
      "learning_rate": 4.999163937780046e-07,
      "loss": 0.0003,
      "reward": 1.8035714775323868,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 5128
    },
    {
      "completion_length": 227.0312623977661,
      "epoch": 0.8600528102602791,
      "grad_norm": 0.34900171200182317,
      "kl": 0.3245086669921875,
      "learning_rate": 4.999160832804282e-07,
      "loss": 0.0003,
      "reward": 1.7464286461472511,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 5130
    },
    {
      "completion_length": 218.93304634094238,
      "epoch": 0.8603881135001467,
      "grad_norm": 0.2367511358892103,
      "kl": 0.117279052734375,
      "learning_rate": 4.999157722074524e-07,
      "loss": 0.0001,
      "reward": 1.8022321909666061,
      "reward_std": 0.027147848159074783,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 5132
    },
    {
      "completion_length": 219.32590198516846,
      "epoch": 0.8607234167400143,
      "grad_norm": 0.1688927659968083,
      "kl": 0.1425018310546875,
      "learning_rate": 4.999154605590778e-07,
      "loss": 0.0001,
      "reward": 1.8178571909666061,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.817857164889574,
      "rewards/format_reward_func": 1.0,
      "step": 5134
    },
    {
      "completion_length": 227.3794755935669,
      "epoch": 0.8610587199798818,
      "grad_norm": 0.19784225720299922,
      "kl": 0.2835845947265625,
      "learning_rate": 4.999151483353052e-07,
      "loss": 0.0003,
      "reward": 1.7660714834928513,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5136
    },
    {
      "completion_length": 242.0803680419922,
      "epoch": 0.8613940232197493,
      "grad_norm": 0.26270531928407537,
      "kl": 0.5782623291015625,
      "learning_rate": 4.999148355361351e-07,
      "loss": 0.0006,
      "reward": 1.775000050663948,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000394880772,
      "rewards/format_reward_func": 1.0,
      "step": 5138
    },
    {
      "completion_length": 238.0357265472412,
      "epoch": 0.8617293264596169,
      "grad_norm": 0.22026295954848013,
      "kl": 0.259002685546875,
      "learning_rate": 4.999145221615685e-07,
      "loss": 0.0003,
      "reward": 1.7678572088479996,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7767857350409031,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5140
    },
    {
      "completion_length": 228.8259048461914,
      "epoch": 0.8620646296994845,
      "grad_norm": 0.14104751882657987,
      "kl": 0.0952606201171875,
      "learning_rate": 4.99914208211606e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 5142
    },
    {
      "completion_length": 236.7812614440918,
      "epoch": 0.862399932939352,
      "grad_norm": 0.2925028960594958,
      "kl": 0.1213836669921875,
      "learning_rate": 4.999138936862484e-07,
      "loss": 0.0001,
      "reward": 1.7714286148548126,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7803571708500385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5144
    },
    {
      "completion_length": 242.68751049041748,
      "epoch": 0.8627352361792195,
      "grad_norm": 0.4819099304816873,
      "kl": 0.167633056640625,
      "learning_rate": 4.999135785854962e-07,
      "loss": 0.0002,
      "reward": 1.7214286401867867,
      "reward_std": 0.12121830228716135,
      "rewards/equation_reward_func": 0.7392857372760773,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5146
    },
    {
      "completion_length": 249.82143878936768,
      "epoch": 0.8630705394190872,
      "grad_norm": 0.25499206507829786,
      "kl": 0.2075958251953125,
      "learning_rate": 4.999132629093503e-07,
      "loss": 0.0002,
      "reward": 1.7035714983940125,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7125000450760126,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5148
    },
    {
      "completion_length": 238.89733409881592,
      "epoch": 0.8634058426589547,
      "grad_norm": 0.2327296549492035,
      "kl": 0.1017913818359375,
      "learning_rate": 4.999129466578116e-07,
      "loss": 0.0001,
      "reward": 1.6696429401636124,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.6741071790456772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5150
    },
    {
      "completion_length": 247.5759038925171,
      "epoch": 0.8637411458988222,
      "grad_norm": 0.20944422452788633,
      "kl": 0.100006103515625,
      "learning_rate": 4.999126298308805e-07,
      "loss": 0.0001,
      "reward": 1.7375000715255737,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7419643104076385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5152
    },
    {
      "completion_length": 244.74108409881592,
      "epoch": 0.8640764491386898,
      "grad_norm": 0.3227991323663998,
      "kl": 0.1121673583984375,
      "learning_rate": 4.999123124285578e-07,
      "loss": 0.0001,
      "reward": 1.7714286521077156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 5154
    },
    {
      "completion_length": 246.30358219146729,
      "epoch": 0.8644117523785574,
      "grad_norm": 0.6971793031048401,
      "kl": 0.1251983642578125,
      "learning_rate": 4.999119944508445e-07,
      "loss": 0.0001,
      "reward": 1.7071429416537285,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7160714715719223,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5156
    },
    {
      "completion_length": 241.41518688201904,
      "epoch": 0.8647470556184249,
      "grad_norm": 0.30123485204649847,
      "kl": 0.0756683349609375,
      "learning_rate": 4.999116758977409e-07,
      "loss": 0.0001,
      "reward": 1.7232143878936768,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7276785969734192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5158
    },
    {
      "completion_length": 242.10715293884277,
      "epoch": 0.8650823588582924,
      "grad_norm": 0.19740171135735748,
      "kl": 0.0821380615234375,
      "learning_rate": 4.999113567692481e-07,
      "loss": 0.0001,
      "reward": 1.7571429461240768,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428753435612,
      "rewards/format_reward_func": 1.0,
      "step": 5160
    },
    {
      "completion_length": 241.02679634094238,
      "epoch": 0.8654176620981601,
      "grad_norm": 0.34486952501648477,
      "kl": 0.049346923828125,
      "learning_rate": 4.999110370653667e-07,
      "loss": 0.0,
      "reward": 1.7839286476373672,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7883928846567869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5162
    },
    {
      "completion_length": 235.915189743042,
      "epoch": 0.8657529653380276,
      "grad_norm": 0.23825360495441034,
      "kl": 0.05047607421875,
      "learning_rate": 4.999107167860973e-07,
      "loss": 0.0001,
      "reward": 1.7946429327130318,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7991071604192257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5164
    },
    {
      "completion_length": 237.2946538925171,
      "epoch": 0.8660882685778951,
      "grad_norm": 0.23002121289385286,
      "kl": 0.0514068603515625,
      "learning_rate": 4.999103959314409e-07,
      "loss": 0.0001,
      "reward": 1.7642858028411865,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857357859612,
      "rewards/format_reward_func": 1.0,
      "step": 5166
    },
    {
      "completion_length": 238.7589406967163,
      "epoch": 0.8664235718177626,
      "grad_norm": 0.7932820511428698,
      "kl": 0.05147552490234375,
      "learning_rate": 4.999100745013981e-07,
      "loss": 0.0001,
      "reward": 1.7375000715255737,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7419643104076385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5168
    },
    {
      "completion_length": 236.88840675354004,
      "epoch": 0.8667588750576303,
      "grad_norm": 0.27955379277920045,
      "kl": 0.0646514892578125,
      "learning_rate": 4.999097524959695e-07,
      "loss": 0.0001,
      "reward": 1.691071517765522,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.6955357603728771,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5170
    },
    {
      "completion_length": 256.50893783569336,
      "epoch": 0.8670941782974978,
      "grad_norm": 0.29478152619315,
      "kl": 0.074310302734375,
      "learning_rate": 4.999094299151562e-07,
      "loss": 0.0001,
      "reward": 1.7214286625385284,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7214285898953676,
      "rewards/format_reward_func": 1.0,
      "step": 5172
    },
    {
      "completion_length": 235.45983123779297,
      "epoch": 0.8674294815373653,
      "grad_norm": 0.32795899856400945,
      "kl": 0.0535736083984375,
      "learning_rate": 4.999091067589585e-07,
      "loss": 0.0001,
      "reward": 1.725000061094761,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7250000387430191,
      "rewards/format_reward_func": 1.0,
      "step": 5174
    },
    {
      "completion_length": 237.6116180419922,
      "epoch": 0.867764784777233,
      "grad_norm": 0.19602738115025517,
      "kl": 0.0596466064453125,
      "learning_rate": 4.999087830273777e-07,
      "loss": 0.0001,
      "reward": 1.7821428999304771,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428999304771,
      "rewards/format_reward_func": 1.0,
      "step": 5176
    },
    {
      "completion_length": 242.0625123977661,
      "epoch": 0.8681000880171005,
      "grad_norm": 0.5267813801735062,
      "kl": 0.0671844482421875,
      "learning_rate": 4.99908458720414e-07,
      "loss": 0.0001,
      "reward": 1.7285714894533157,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7464285995811224,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5178
    },
    {
      "completion_length": 230.95536994934082,
      "epoch": 0.868435391256968,
      "grad_norm": 0.24433297853268698,
      "kl": 0.064056396484375,
      "learning_rate": 4.999081338380684e-07,
      "loss": 0.0001,
      "reward": 1.7500000968575478,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7580357417464256,
      "rewards/format_reward_func": 0.9919642880558968,
      "step": 5180
    },
    {
      "completion_length": 246.0982255935669,
      "epoch": 0.8687706944968355,
      "grad_norm": 0.8587767902492196,
      "kl": 0.1986541748046875,
      "learning_rate": 4.999078083803416e-07,
      "loss": 0.0002,
      "reward": 1.7321429252624512,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7410714663565159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5182
    },
    {
      "completion_length": 231.10715293884277,
      "epoch": 0.8691059977367032,
      "grad_norm": 0.22419756108609906,
      "kl": 0.0513763427734375,
      "learning_rate": 4.999074823472344e-07,
      "loss": 0.0001,
      "reward": 1.7285715192556381,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.728571455925703,
      "rewards/format_reward_func": 1.0,
      "step": 5184
    },
    {
      "completion_length": 234.24108219146729,
      "epoch": 0.8694413009765707,
      "grad_norm": 0.2572663801825886,
      "kl": 0.12969970703125,
      "learning_rate": 4.999071557387475e-07,
      "loss": 0.0001,
      "reward": 1.7267857939004898,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7312500402331352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5186
    },
    {
      "completion_length": 237.3482265472412,
      "epoch": 0.8697766042164382,
      "grad_norm": 0.2707232720959836,
      "kl": 0.117584228515625,
      "learning_rate": 4.999068285548816e-07,
      "loss": 0.0001,
      "reward": 1.7732143476605415,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.777678593993187,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5188
    },
    {
      "completion_length": 229.05804538726807,
      "epoch": 0.8701119074563058,
      "grad_norm": 0.1857235486891341,
      "kl": 0.05426025390625,
      "learning_rate": 4.999065007956377e-07,
      "loss": 0.0001,
      "reward": 1.7714286148548126,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714286111295223,
      "rewards/format_reward_func": 1.0,
      "step": 5190
    },
    {
      "completion_length": 238.2723331451416,
      "epoch": 0.8704472106961734,
      "grad_norm": 0.4293622096350722,
      "kl": 0.111297607421875,
      "learning_rate": 4.999061724610163e-07,
      "loss": 0.0001,
      "reward": 1.7553572207689285,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7598214671015739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5192
    },
    {
      "completion_length": 236.36161994934082,
      "epoch": 0.8707825139360409,
      "grad_norm": 0.2534201839770564,
      "kl": 0.2327423095703125,
      "learning_rate": 4.999058435510182e-07,
      "loss": 0.0002,
      "reward": 1.7642857730388641,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857655882835,
      "rewards/format_reward_func": 1.0,
      "step": 5194
    },
    {
      "completion_length": 242.7053689956665,
      "epoch": 0.8711178171759084,
      "grad_norm": 0.4089098177812318,
      "kl": 0.1260986328125,
      "learning_rate": 4.999055140656442e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 5196
    },
    {
      "completion_length": 242.01786708831787,
      "epoch": 0.871453120415776,
      "grad_norm": 0.2138423740410924,
      "kl": 0.1017303466796875,
      "learning_rate": 4.99905184004895e-07,
      "loss": 0.0001,
      "reward": 1.7857143580913544,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143245637417,
      "rewards/format_reward_func": 1.0,
      "step": 5198
    },
    {
      "completion_length": 246.15179824829102,
      "epoch": 0.8717884236556436,
      "grad_norm": 0.25927983778414143,
      "kl": 0.0787200927734375,
      "learning_rate": 4.999048533687715e-07,
      "loss": 0.0001,
      "reward": 1.6803572252392769,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.6937500238418579,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5200
    },
    {
      "completion_length": 241.31250953674316,
      "epoch": 0.8721237268955111,
      "grad_norm": 0.25433932775478646,
      "kl": 0.0538330078125,
      "learning_rate": 4.999045221572743e-07,
      "loss": 0.0001,
      "reward": 1.705357238650322,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7098214626312256,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5202
    },
    {
      "completion_length": 240.00447368621826,
      "epoch": 0.8724590301353787,
      "grad_norm": 0.1208999441426067,
      "kl": 0.084716796875,
      "learning_rate": 4.999041903704043e-07,
      "loss": 0.0001,
      "reward": 1.7250000685453415,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7250000331550837,
      "rewards/format_reward_func": 1.0,
      "step": 5204
    },
    {
      "completion_length": 248.1384048461914,
      "epoch": 0.8727943333752463,
      "grad_norm": 0.347887490986824,
      "kl": 0.2652587890625,
      "learning_rate": 4.999038580081621e-07,
      "loss": 0.0003,
      "reward": 1.721428669989109,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7214286029338837,
      "rewards/format_reward_func": 1.0,
      "step": 5206
    },
    {
      "completion_length": 241.5580472946167,
      "epoch": 0.8731296366151138,
      "grad_norm": 0.11248959369685928,
      "kl": 0.09877777099609375,
      "learning_rate": 4.999035250705486e-07,
      "loss": 0.0001,
      "reward": 1.7428572326898575,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571730852127,
      "rewards/format_reward_func": 1.0,
      "step": 5208
    },
    {
      "completion_length": 246.4553680419922,
      "epoch": 0.8734649398549813,
      "grad_norm": 0.20355932834966817,
      "kl": 0.119964599609375,
      "learning_rate": 4.999031915575645e-07,
      "loss": 0.0001,
      "reward": 1.678571492433548,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.6785714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 5210
    },
    {
      "completion_length": 240.62501335144043,
      "epoch": 0.8738002430948489,
      "grad_norm": 0.3614959079050482,
      "kl": 0.0868377685546875,
      "learning_rate": 4.999028574692107e-07,
      "loss": 0.0001,
      "reward": 1.7232143729925156,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7276785932481289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5212
    },
    {
      "completion_length": 232.2321548461914,
      "epoch": 0.8741355463347165,
      "grad_norm": 0.27345804390142686,
      "kl": 0.267913818359375,
      "learning_rate": 4.999025228054878e-07,
      "loss": 0.0003,
      "reward": 1.7857143431901932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 5214
    },
    {
      "completion_length": 243.81250953674316,
      "epoch": 0.874470849574584,
      "grad_norm": 0.25686369513436424,
      "kl": 0.05169677734375,
      "learning_rate": 4.999021875663967e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321429010480642,
      "rewards/format_reward_func": 1.0,
      "step": 5216
    },
    {
      "completion_length": 236.5892972946167,
      "epoch": 0.8748061528144516,
      "grad_norm": 0.2938128211300729,
      "kl": 0.0554046630859375,
      "learning_rate": 4.99901851751938e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 5218
    },
    {
      "completion_length": 238.8482255935669,
      "epoch": 0.8751414560543191,
      "grad_norm": 0.2940552545555363,
      "kl": 0.202545166015625,
      "learning_rate": 4.999015153621126e-07,
      "loss": 0.0002,
      "reward": 1.7428572252392769,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 5220
    },
    {
      "completion_length": 235.82143783569336,
      "epoch": 0.8754767592941867,
      "grad_norm": 0.18830219720647964,
      "kl": 0.124786376953125,
      "learning_rate": 4.999011783969213e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 5222
    },
    {
      "completion_length": 245.26786994934082,
      "epoch": 0.8758120625340542,
      "grad_norm": 0.38066338937574323,
      "kl": 0.1001739501953125,
      "learning_rate": 4.999008408563649e-07,
      "loss": 0.0001,
      "reward": 1.7321429178118706,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7410714514553547,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5224
    },
    {
      "completion_length": 233.4732255935669,
      "epoch": 0.8761473657739218,
      "grad_norm": 0.4616713763436838,
      "kl": 0.1026153564453125,
      "learning_rate": 4.999005027404439e-07,
      "loss": 0.0001,
      "reward": 1.7821428999304771,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 5226
    },
    {
      "completion_length": 239.80358409881592,
      "epoch": 0.8764826690137894,
      "grad_norm": 0.2662263067010209,
      "kl": 0.0876007080078125,
      "learning_rate": 4.999001640491595e-07,
      "loss": 0.0001,
      "reward": 1.751785784959793,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7562500387430191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5228
    },
    {
      "completion_length": 238.1294755935669,
      "epoch": 0.8768179722536569,
      "grad_norm": 0.2532932673862093,
      "kl": 0.064971923828125,
      "learning_rate": 4.998998247825121e-07,
      "loss": 0.0001,
      "reward": 1.7428572177886963,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 5230
    },
    {
      "completion_length": 239.14733219146729,
      "epoch": 0.8771532754935245,
      "grad_norm": 0.22009531892447384,
      "kl": 0.06134033203125,
      "learning_rate": 4.998994849405027e-07,
      "loss": 0.0001,
      "reward": 1.780357226729393,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5232
    },
    {
      "completion_length": 242.4062614440918,
      "epoch": 0.877488578733392,
      "grad_norm": 0.1460440667442655,
      "kl": 0.0941925048828125,
      "learning_rate": 4.99899144523132e-07,
      "loss": 0.0001,
      "reward": 1.705357201397419,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7187500353902578,
      "rewards/format_reward_func": 0.9866071455180645,
      "step": 5234
    },
    {
      "completion_length": 239.28126049041748,
      "epoch": 0.8778238819732596,
      "grad_norm": 0.25190830542060527,
      "kl": 0.0667724609375,
      "learning_rate": 4.998988035304009e-07,
      "loss": 0.0001,
      "reward": 1.7232143729925156,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7276786044239998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5236
    },
    {
      "completion_length": 244.16518783569336,
      "epoch": 0.8781591852131271,
      "grad_norm": 0.17071947606065108,
      "kl": 0.089874267578125,
      "learning_rate": 4.998984619623101e-07,
      "loss": 0.0001,
      "reward": 1.7339286655187607,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7383928932249546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5238
    },
    {
      "completion_length": 230.227689743042,
      "epoch": 0.8784944884529947,
      "grad_norm": 0.3664008933351735,
      "kl": 0.0554656982421875,
      "learning_rate": 4.998981198188603e-07,
      "loss": 0.0001,
      "reward": 1.814285784959793,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.8142857439815998,
      "rewards/format_reward_func": 1.0,
      "step": 5240
    },
    {
      "completion_length": 229.21876049041748,
      "epoch": 0.8788297916928622,
      "grad_norm": 0.21221439980005558,
      "kl": 0.06195068359375,
      "learning_rate": 4.998977771000525e-07,
      "loss": 0.0001,
      "reward": 1.7642857655882835,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.764285733923316,
      "rewards/format_reward_func": 1.0,
      "step": 5242
    },
    {
      "completion_length": 234.59376049041748,
      "epoch": 0.8791650949327298,
      "grad_norm": 0.3746951037284132,
      "kl": 0.07000732421875,
      "learning_rate": 4.998974338058872e-07,
      "loss": 0.0001,
      "reward": 1.6857143566012383,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.6857143249362707,
      "rewards/format_reward_func": 1.0,
      "step": 5244
    },
    {
      "completion_length": 242.1696538925171,
      "epoch": 0.8795003981725974,
      "grad_norm": 0.18881294484194702,
      "kl": 0.0675506591796875,
      "learning_rate": 4.998970899363655e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 5246
    },
    {
      "completion_length": 226.1384038925171,
      "epoch": 0.8798357014124649,
      "grad_norm": 0.22507211877639,
      "kl": 0.07757568359375,
      "learning_rate": 4.998967454914879e-07,
      "loss": 0.0001,
      "reward": 1.830357201397419,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.8348214440047741,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5248
    },
    {
      "completion_length": 245.1026906967163,
      "epoch": 0.8801710046523324,
      "grad_norm": 0.22393823771667665,
      "kl": 0.06451416015625,
      "learning_rate": 4.998964004712555e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 5250
    },
    {
      "completion_length": 232.92411708831787,
      "epoch": 0.8805063078922,
      "grad_norm": 0.2144873023449417,
      "kl": 0.0620269775390625,
      "learning_rate": 4.998960548756689e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7732143141329288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5252
    },
    {
      "completion_length": 232.87054347991943,
      "epoch": 0.8808416111320676,
      "grad_norm": 0.1989102781559067,
      "kl": 0.05689239501953125,
      "learning_rate": 4.998957087047288e-07,
      "loss": 0.0001,
      "reward": 1.7464286237955093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 5254
    },
    {
      "completion_length": 237.9062614440918,
      "epoch": 0.8811769143719351,
      "grad_norm": 0.2512641577169618,
      "kl": 0.058868408203125,
      "learning_rate": 4.998953619584363e-07,
      "loss": 0.0001,
      "reward": 1.7803572118282318,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7848214693367481,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5256
    },
    {
      "completion_length": 244.99108219146729,
      "epoch": 0.8815122176118027,
      "grad_norm": 0.25304485784456665,
      "kl": 0.0676422119140625,
      "learning_rate": 4.998950146367918e-07,
      "loss": 0.0001,
      "reward": 1.7053572162985802,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7098214570432901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5258
    },
    {
      "completion_length": 236.65179538726807,
      "epoch": 0.8818475208516702,
      "grad_norm": 0.2993844993610931,
      "kl": 0.0611572265625,
      "learning_rate": 4.998946667397966e-07,
      "loss": 0.0001,
      "reward": 1.7357143834233284,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 5260
    },
    {
      "completion_length": 235.5044755935669,
      "epoch": 0.8821828240915378,
      "grad_norm": 0.30016998238421877,
      "kl": 0.0649871826171875,
      "learning_rate": 4.99894318267451e-07,
      "loss": 0.0001,
      "reward": 1.785714365541935,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 5262
    },
    {
      "completion_length": 239.3973331451416,
      "epoch": 0.8825181273314053,
      "grad_norm": 0.3114488879046236,
      "kl": 0.08050537109375,
      "learning_rate": 4.99893969219756e-07,
      "loss": 0.0001,
      "reward": 1.7892857939004898,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7982143089175224,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5264
    },
    {
      "completion_length": 230.040189743042,
      "epoch": 0.8828534305712729,
      "grad_norm": 0.3517487633510106,
      "kl": 0.0715179443359375,
      "learning_rate": 4.998936195967126e-07,
      "loss": 0.0001,
      "reward": 1.7236607894301414,
      "reward_std": 0.06755394977517426,
      "rewards/equation_reward_func": 0.7250000424683094,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 5266
    },
    {
      "completion_length": 235.5178680419922,
      "epoch": 0.8831887338111405,
      "grad_norm": 0.41806628054869266,
      "kl": 0.07794189453125,
      "learning_rate": 4.998932693983213e-07,
      "loss": 0.0001,
      "reward": 1.7053572461009026,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7098214700818062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5268
    },
    {
      "completion_length": 232.10268783569336,
      "epoch": 0.883524037051008,
      "grad_norm": 0.2858866580292781,
      "kl": 0.08392333984375,
      "learning_rate": 4.998929186245832e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 5270
    },
    {
      "completion_length": 233.68751049041748,
      "epoch": 0.8838593402908755,
      "grad_norm": 0.25485525873994247,
      "kl": 0.0716400146484375,
      "learning_rate": 4.998925672754987e-07,
      "loss": 0.0001,
      "reward": 1.7464286610484123,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 5272
    },
    {
      "completion_length": 226.62947463989258,
      "epoch": 0.8841946435307431,
      "grad_norm": 0.17777709268158565,
      "kl": 0.0927581787109375,
      "learning_rate": 4.998922153510691e-07,
      "loss": 0.0001,
      "reward": 1.807142935693264,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428872644901,
      "rewards/format_reward_func": 1.0,
      "step": 5274
    },
    {
      "completion_length": 240.352689743042,
      "epoch": 0.8845299467706107,
      "grad_norm": 0.18581282838696533,
      "kl": 0.2295989990234375,
      "learning_rate": 4.998918628512949e-07,
      "loss": 0.0002,
      "reward": 1.7089286372065544,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7133928909897804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5276
    },
    {
      "completion_length": 233.8303680419922,
      "epoch": 0.8848652500104782,
      "grad_norm": 0.3049793148610497,
      "kl": 0.0879058837890625,
      "learning_rate": 4.998915097761769e-07,
      "loss": 0.0001,
      "reward": 1.803571492433548,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 5278
    },
    {
      "completion_length": 232.40179443359375,
      "epoch": 0.8852005532503457,
      "grad_norm": 0.09532660700253795,
      "kl": 0.06298828125,
      "learning_rate": 4.998911561257161e-07,
      "loss": 0.0001,
      "reward": 1.8071429058909416,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.807142898440361,
      "rewards/format_reward_func": 1.0,
      "step": 5280
    },
    {
      "completion_length": 226.49554634094238,
      "epoch": 0.8855358564902134,
      "grad_norm": 0.2932981971207065,
      "kl": 0.06439208984375,
      "learning_rate": 4.998908018999131e-07,
      "loss": 0.0001,
      "reward": 1.732142947614193,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321428786963224,
      "rewards/format_reward_func": 1.0,
      "step": 5282
    },
    {
      "completion_length": 238.65179634094238,
      "epoch": 0.8858711597300809,
      "grad_norm": 0.32438201855121496,
      "kl": 0.0819091796875,
      "learning_rate": 4.998904470987689e-07,
      "loss": 0.0001,
      "reward": 1.7678571939468384,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.767857177183032,
      "rewards/format_reward_func": 1.0,
      "step": 5284
    },
    {
      "completion_length": 238.85268878936768,
      "epoch": 0.8862064629699484,
      "grad_norm": 0.3012234729171621,
      "kl": 0.08880615234375,
      "learning_rate": 4.998900917222842e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 5286
    },
    {
      "completion_length": 236.5580472946167,
      "epoch": 0.886541766209816,
      "grad_norm": 0.2963282908676942,
      "kl": 0.07366943359375,
      "learning_rate": 4.998897357704598e-07,
      "loss": 0.0001,
      "reward": 1.8017857819795609,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.8062500208616257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5288
    },
    {
      "completion_length": 233.37054538726807,
      "epoch": 0.8868770694496836,
      "grad_norm": 0.3279870054415193,
      "kl": 0.06890869140625,
      "learning_rate": 4.998893792432966e-07,
      "loss": 0.0001,
      "reward": 1.7642857506871223,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857413738966,
      "rewards/format_reward_func": 1.0,
      "step": 5290
    },
    {
      "completion_length": 252.91072463989258,
      "epoch": 0.8872123726895511,
      "grad_norm": 0.27756577372857344,
      "kl": 0.121978759765625,
      "learning_rate": 4.998890221407956e-07,
      "loss": 0.0001,
      "reward": 1.748214341700077,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7526785973459482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5292
    },
    {
      "completion_length": 236.54018688201904,
      "epoch": 0.8875476759294186,
      "grad_norm": 0.2331425793873391,
      "kl": 0.1655120849609375,
      "learning_rate": 4.998886644629572e-07,
      "loss": 0.0002,
      "reward": 1.7857143580913544,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 5294
    },
    {
      "completion_length": 243.5491189956665,
      "epoch": 0.8878829791692863,
      "grad_norm": 0.16776429680449167,
      "kl": 0.079864501953125,
      "learning_rate": 4.998883062097824e-07,
      "loss": 0.0001,
      "reward": 1.7705357745289803,
      "reward_std": 0.0416687922552228,
      "rewards/equation_reward_func": 0.7767857387661934,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 5296
    },
    {
      "completion_length": 240.41965293884277,
      "epoch": 0.8882182824091538,
      "grad_norm": 0.8081502933749403,
      "kl": 0.257476806640625,
      "learning_rate": 4.998879473812722e-07,
      "loss": 0.0003,
      "reward": 1.8214286267757416,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8214286006987095,
      "rewards/format_reward_func": 1.0,
      "step": 5298
    },
    {
      "completion_length": 244.4687614440918,
      "epoch": 0.8885535856490213,
      "grad_norm": 0.14530388506531866,
      "kl": 0.0879058837890625,
      "learning_rate": 4.998875879774273e-07,
      "loss": 0.0001,
      "reward": 1.7500000521540642,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 5300
    },
    {
      "completion_length": 229.28126335144043,
      "epoch": 0.8888888888888888,
      "grad_norm": 0.2260688976115879,
      "kl": 0.09368896484375,
      "learning_rate": 4.998872279982485e-07,
      "loss": 0.0001,
      "reward": 1.792857214808464,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 5302
    },
    {
      "completion_length": 234.7812614440918,
      "epoch": 0.8892241921287565,
      "grad_norm": 0.1600646981940207,
      "kl": 0.0696563720703125,
      "learning_rate": 4.998868674437365e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571622818708,
      "rewards/format_reward_func": 1.0,
      "step": 5304
    },
    {
      "completion_length": 245.08929824829102,
      "epoch": 0.889559495368624,
      "grad_norm": 0.20507006335606773,
      "kl": 0.068084716796875,
      "learning_rate": 4.998865063138926e-07,
      "loss": 0.0001,
      "reward": 1.7482143715023994,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526785954833031,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5306
    },
    {
      "completion_length": 236.44197750091553,
      "epoch": 0.8898947986084915,
      "grad_norm": 0.24173203157702566,
      "kl": 0.0780029296875,
      "learning_rate": 4.99886144608717e-07,
      "loss": 0.0001,
      "reward": 1.7821429371833801,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7901786025613546,
      "rewards/format_reward_func": 0.9919642880558968,
      "step": 5308
    },
    {
      "completion_length": 243.21876335144043,
      "epoch": 0.8902301018483592,
      "grad_norm": 0.2508191564510342,
      "kl": 0.076171875,
      "learning_rate": 4.99885782328211e-07,
      "loss": 0.0001,
      "reward": 1.7392857745289803,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857596278191,
      "rewards/format_reward_func": 1.0,
      "step": 5310
    },
    {
      "completion_length": 246.66518688201904,
      "epoch": 0.8905654050882267,
      "grad_norm": 0.19502896555021526,
      "kl": 0.07415771484375,
      "learning_rate": 4.998854194723752e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143059372902,
      "rewards/format_reward_func": 1.0,
      "step": 5312
    },
    {
      "completion_length": 239.36608028411865,
      "epoch": 0.8909007083280942,
      "grad_norm": 0.2431954955354549,
      "kl": 0.072662353515625,
      "learning_rate": 4.998850560412106e-07,
      "loss": 0.0001,
      "reward": 1.764285795390606,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 5314
    },
    {
      "completion_length": 261.59376335144043,
      "epoch": 0.8912360115679617,
      "grad_norm": 0.23926305068072765,
      "kl": 0.07647705078125,
      "learning_rate": 4.998846920347178e-07,
      "loss": 0.0001,
      "reward": 1.7732143849134445,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7776785865426064,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5316
    },
    {
      "completion_length": 253.9419755935669,
      "epoch": 0.8915713148078294,
      "grad_norm": 0.2160253207774375,
      "kl": 0.0731964111328125,
      "learning_rate": 4.99884327452898e-07,
      "loss": 0.0001,
      "reward": 1.7428571954369545,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571730852127,
      "rewards/format_reward_func": 1.0,
      "step": 5318
    },
    {
      "completion_length": 250.19197750091553,
      "epoch": 0.8919066180476969,
      "grad_norm": 0.20380295428031342,
      "kl": 0.072662353515625,
      "learning_rate": 4.998839622957517e-07,
      "loss": 0.0001,
      "reward": 1.7285715267062187,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7285714633762836,
      "rewards/format_reward_func": 1.0,
      "step": 5320
    },
    {
      "completion_length": 243.31697463989258,
      "epoch": 0.8922419212875644,
      "grad_norm": 0.22780821035437138,
      "kl": 0.084014892578125,
      "learning_rate": 4.998835965632798e-07,
      "loss": 0.0001,
      "reward": 1.7125000804662704,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7169643081724644,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5322
    },
    {
      "completion_length": 252.35715293884277,
      "epoch": 0.892577224527432,
      "grad_norm": 0.22647910989616787,
      "kl": 0.07513427734375,
      "learning_rate": 4.998832302554834e-07,
      "loss": 0.0001,
      "reward": 1.739285796880722,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857521772385,
      "rewards/format_reward_func": 1.0,
      "step": 5324
    },
    {
      "completion_length": 255.96876430511475,
      "epoch": 0.8929125277672996,
      "grad_norm": 0.27754938674440754,
      "kl": 0.08740234375,
      "learning_rate": 4.99882863372363e-07,
      "loss": 0.0001,
      "reward": 1.7375000789761543,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643141329288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5326
    },
    {
      "completion_length": 259.9509048461914,
      "epoch": 0.8932478310071671,
      "grad_norm": 0.1811040255508447,
      "kl": 0.084320068359375,
      "learning_rate": 4.998824959139196e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143152505159,
      "rewards/format_reward_func": 1.0,
      "step": 5328
    },
    {
      "completion_length": 248.4196548461914,
      "epoch": 0.8935831342470346,
      "grad_norm": 0.5155124794678484,
      "kl": 0.07611083984375,
      "learning_rate": 4.998821278801542e-07,
      "loss": 0.0001,
      "reward": 1.719642959535122,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.724107176065445,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5330
    },
    {
      "completion_length": 250.12947940826416,
      "epoch": 0.8939184374869023,
      "grad_norm": 0.24714452364489936,
      "kl": 0.078399658203125,
      "learning_rate": 4.998817592710674e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 5332
    },
    {
      "completion_length": 246.7901906967163,
      "epoch": 0.8942537407267698,
      "grad_norm": 0.26676101095918225,
      "kl": 0.0773773193359375,
      "learning_rate": 4.998813900866601e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857544124126,
      "rewards/format_reward_func": 1.0,
      "step": 5334
    },
    {
      "completion_length": 250.9821538925171,
      "epoch": 0.8945890439666373,
      "grad_norm": 0.839157664979466,
      "kl": 0.082855224609375,
      "learning_rate": 4.998810203269333e-07,
      "loss": 0.0001,
      "reward": 1.7392858117818832,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.748214315623045,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5336
    },
    {
      "completion_length": 252.33483123779297,
      "epoch": 0.8949243472065049,
      "grad_norm": 0.12935000666624785,
      "kl": 0.0791015625,
      "learning_rate": 4.998806499918876e-07,
      "loss": 0.0001,
      "reward": 1.775000050663948,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 5338
    },
    {
      "completion_length": 246.33036708831787,
      "epoch": 0.8952596504463725,
      "grad_norm": 0.10639964651343883,
      "kl": 0.09002685546875,
      "learning_rate": 4.998802790815241e-07,
      "loss": 0.0001,
      "reward": 1.8017857670783997,
      "reward_std": 0.04798224661499262,
      "rewards/equation_reward_func": 0.8151785917580128,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5340
    },
    {
      "completion_length": 248.7634048461914,
      "epoch": 0.89559495368624,
      "grad_norm": 0.33624572209025144,
      "kl": 0.0796356201171875,
      "learning_rate": 4.998799075958435e-07,
      "loss": 0.0001,
      "reward": 1.7464286610484123,
      "reward_std": 0.0858629634603858,
      "rewards/equation_reward_func": 0.7464285995811224,
      "rewards/format_reward_func": 1.0,
      "step": 5342
    },
    {
      "completion_length": 259.901798248291,
      "epoch": 0.8959302569261075,
      "grad_norm": 0.1254125349254734,
      "kl": 0.082733154296875,
      "learning_rate": 4.998795355348467e-07,
      "loss": 0.0001,
      "reward": 1.7375000566244125,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643215835094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5344
    },
    {
      "completion_length": 263.9285840988159,
      "epoch": 0.8962655601659751,
      "grad_norm": 0.31235861346966837,
      "kl": 0.1109619140625,
      "learning_rate": 4.998791628985346e-07,
      "loss": 0.0001,
      "reward": 1.733928643167019,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7383928820490837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5346
    },
    {
      "completion_length": 242.00447750091553,
      "epoch": 0.8966008634058427,
      "grad_norm": 0.1923375555144056,
      "kl": 0.08599853515625,
      "learning_rate": 4.99878789686908e-07,
      "loss": 0.0001,
      "reward": 1.7517857626080513,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7562500312924385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5348
    },
    {
      "completion_length": 253.73215293884277,
      "epoch": 0.8969361666457102,
      "grad_norm": 0.3144491273579619,
      "kl": 0.157440185546875,
      "learning_rate": 4.998784158999677e-07,
      "loss": 0.0002,
      "reward": 1.7482143566012383,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7526785880327225,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5350
    },
    {
      "completion_length": 258.86608505249023,
      "epoch": 0.8972714698855778,
      "grad_norm": 0.3800286104807005,
      "kl": 0.0876312255859375,
      "learning_rate": 4.998780415377148e-07,
      "loss": 0.0001,
      "reward": 1.7946428880095482,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7991071678698063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5352
    },
    {
      "completion_length": 255.0178680419922,
      "epoch": 0.8976067731254453,
      "grad_norm": 0.5843896226137154,
      "kl": 0.1268310546875,
      "learning_rate": 4.998776666001499e-07,
      "loss": 0.0001,
      "reward": 1.6750000715255737,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.6839286163449287,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5354
    },
    {
      "completion_length": 252.58929634094238,
      "epoch": 0.8979420763653129,
      "grad_norm": 0.3539484283441876,
      "kl": 0.2587127685546875,
      "learning_rate": 4.998772910872739e-07,
      "loss": 0.0003,
      "reward": 1.7035714909434319,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7125000301748514,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5356
    },
    {
      "completion_length": 254.33929538726807,
      "epoch": 0.8982773796051804,
      "grad_norm": 0.46943661426880745,
      "kl": 0.193206787109375,
      "learning_rate": 4.998769149990878e-07,
      "loss": 0.0002,
      "reward": 1.7928571999073029,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 5358
    },
    {
      "completion_length": 244.76786994934082,
      "epoch": 0.898612682845048,
      "grad_norm": 0.12745735511328252,
      "kl": 0.22625732421875,
      "learning_rate": 4.998765383355924e-07,
      "loss": 0.0002,
      "reward": 1.753571517765522,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 5360
    },
    {
      "completion_length": 254.78572750091553,
      "epoch": 0.8989479860849156,
      "grad_norm": 0.20917256499861125,
      "kl": 0.3976593017578125,
      "learning_rate": 4.998761610967885e-07,
      "loss": 0.0004,
      "reward": 1.7982143610715866,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.8026786036789417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5362
    },
    {
      "completion_length": 242.13393688201904,
      "epoch": 0.8992832893247831,
      "grad_norm": 0.4313738380742602,
      "kl": 0.1184539794921875,
      "learning_rate": 4.998757832826772e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 5364
    },
    {
      "completion_length": 255.21875953674316,
      "epoch": 0.8996185925646507,
      "grad_norm": 0.5463979836566241,
      "kl": 0.6548309326171875,
      "learning_rate": 4.99875404893259e-07,
      "loss": 0.0007,
      "reward": 1.7017857879400253,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.706250037997961,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5366
    },
    {
      "completion_length": 249.6517972946167,
      "epoch": 0.8999538958045182,
      "grad_norm": 0.2889413935758667,
      "kl": 0.3985443115234375,
      "learning_rate": 4.998750259285351e-07,
      "loss": 0.0004,
      "reward": 1.7428572177886963,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7428571656346321,
      "rewards/format_reward_func": 1.0,
      "step": 5368
    },
    {
      "completion_length": 246.79912090301514,
      "epoch": 0.9002891990443858,
      "grad_norm": 0.31076407243427856,
      "kl": 0.1228179931640625,
      "learning_rate": 4.998746463885062e-07,
      "loss": 0.0001,
      "reward": 1.7982143387198448,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8026785962283611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5370
    },
    {
      "completion_length": 247.64733123779297,
      "epoch": 0.9006245022842533,
      "grad_norm": 0.4925333749051105,
      "kl": 1.430389404296875,
      "learning_rate": 4.998742662731732e-07,
      "loss": 0.0014,
      "reward": 1.79464291036129,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7991071566939354,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5372
    },
    {
      "completion_length": 244.7232265472412,
      "epoch": 0.9009598055241209,
      "grad_norm": 0.20763244478129034,
      "kl": 1.0489349365234375,
      "learning_rate": 4.998738855825371e-07,
      "loss": 0.001,
      "reward": 1.76071435213089,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143260538578,
      "rewards/format_reward_func": 1.0,
      "step": 5374
    },
    {
      "completion_length": 245.29911518096924,
      "epoch": 0.9012951087639884,
      "grad_norm": 0.7021375598228516,
      "kl": 0.698516845703125,
      "learning_rate": 4.998735043165986e-07,
      "loss": 0.0007,
      "reward": 1.7821429073810577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428962051868,
      "rewards/format_reward_func": 1.0,
      "step": 5376
    },
    {
      "completion_length": 257.50894260406494,
      "epoch": 0.901630412003856,
      "grad_norm": 0.7357447369391495,
      "kl": 0.44970703125,
      "learning_rate": 4.998731224753586e-07,
      "loss": 0.0004,
      "reward": 1.7321429327130318,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7410714663565159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5378
    },
    {
      "completion_length": 247.5491189956665,
      "epoch": 0.9019657152437235,
      "grad_norm": 0.47121200401910107,
      "kl": 1.12872314453125,
      "learning_rate": 4.99872740058818e-07,
      "loss": 0.0011,
      "reward": 1.7125000804662704,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7169643118977547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5380
    },
    {
      "completion_length": 247.50893783569336,
      "epoch": 0.9023010184835911,
      "grad_norm": 0.3110160210673541,
      "kl": 0.1286468505859375,
      "learning_rate": 4.998723570669778e-07,
      "loss": 0.0001,
      "reward": 1.7071429416537285,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7071428969502449,
      "rewards/format_reward_func": 1.0,
      "step": 5382
    },
    {
      "completion_length": 255.0625123977661,
      "epoch": 0.9026363217234586,
      "grad_norm": 0.33694169060014484,
      "kl": 0.8502655029296875,
      "learning_rate": 4.998719734998387e-07,
      "loss": 0.0008,
      "reward": 1.6785715445876122,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.6875000335276127,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5384
    },
    {
      "completion_length": 247.7053680419922,
      "epoch": 0.9029716249633262,
      "grad_norm": 0.17383252788350392,
      "kl": 0.1181640625,
      "learning_rate": 4.998715893574018e-07,
      "loss": 0.0001,
      "reward": 1.721428669989109,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7214285954833031,
      "rewards/format_reward_func": 1.0,
      "step": 5386
    },
    {
      "completion_length": 249.31697463989258,
      "epoch": 0.9033069282031938,
      "grad_norm": 0.2062389071756386,
      "kl": 0.0739898681640625,
      "learning_rate": 4.998712046396677e-07,
      "loss": 0.0001,
      "reward": 1.7607143223285675,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 5388
    },
    {
      "completion_length": 253.4375114440918,
      "epoch": 0.9036422314430613,
      "grad_norm": 0.1814458101267558,
      "kl": 0.0583648681640625,
      "learning_rate": 4.998708193466375e-07,
      "loss": 0.0001,
      "reward": 1.7267857864499092,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7312500439584255,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5390
    },
    {
      "completion_length": 239.55804824829102,
      "epoch": 0.9039775346829289,
      "grad_norm": 0.29584868151822824,
      "kl": 0.0521392822265625,
      "learning_rate": 4.998704334783121e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857361584902,
      "rewards/format_reward_func": 1.0,
      "step": 5392
    },
    {
      "completion_length": 250.7455472946167,
      "epoch": 0.9043128379227964,
      "grad_norm": 0.23987515671916398,
      "kl": 0.0930328369140625,
      "learning_rate": 4.998700470346923e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000264495611,
      "rewards/format_reward_func": 1.0,
      "step": 5394
    },
    {
      "completion_length": 236.6741189956665,
      "epoch": 0.904648141162664,
      "grad_norm": 0.27100398738808573,
      "kl": 0.0775299072265625,
      "learning_rate": 4.998696600157789e-07,
      "loss": 0.0001,
      "reward": 1.7142858132719994,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7142857387661934,
      "rewards/format_reward_func": 1.0,
      "step": 5396
    },
    {
      "completion_length": 249.26340007781982,
      "epoch": 0.9049834444025315,
      "grad_norm": 0.19531226401891985,
      "kl": 0.1512298583984375,
      "learning_rate": 4.998692724215731e-07,
      "loss": 0.0002,
      "reward": 1.7571429535746574,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428753435612,
      "rewards/format_reward_func": 1.0,
      "step": 5398
    },
    {
      "completion_length": 251.31250953674316,
      "epoch": 0.9053187476423991,
      "grad_norm": 0.18254214716236256,
      "kl": 0.05153656005859375,
      "learning_rate": 4.998688842520754e-07,
      "loss": 0.0001,
      "reward": 1.746428668498993,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 5400
    },
    {
      "completion_length": 244.5714406967163,
      "epoch": 0.9056540508822667,
      "grad_norm": 0.19710436019518554,
      "kl": 0.0549468994140625,
      "learning_rate": 4.998684955072869e-07,
      "loss": 0.0001,
      "reward": 1.7678572237491608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 5402
    },
    {
      "completion_length": 248.0982255935669,
      "epoch": 0.9059893541221342,
      "grad_norm": 0.2126268791462505,
      "kl": 0.0541839599609375,
      "learning_rate": 4.998681061872086e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 5404
    },
    {
      "completion_length": 245.34822463989258,
      "epoch": 0.9063246573620017,
      "grad_norm": 0.28491657265214865,
      "kl": 0.1408843994140625,
      "learning_rate": 4.998677162918411e-07,
      "loss": 0.0001,
      "reward": 1.751785784959793,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500461935997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5406
    },
    {
      "completion_length": 250.87947750091553,
      "epoch": 0.9066599606018693,
      "grad_norm": 0.28512756855223254,
      "kl": 0.187103271484375,
      "learning_rate": 4.998673258211857e-07,
      "loss": 0.0002,
      "reward": 1.7464286386966705,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.755357176065445,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5408
    },
    {
      "completion_length": 250.35269165039062,
      "epoch": 0.9069952638417369,
      "grad_norm": 0.1916385409286891,
      "kl": 0.05348968505859375,
      "learning_rate": 4.998669347752429e-07,
      "loss": 0.0001,
      "reward": 1.8142857551574707,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8142857402563095,
      "rewards/format_reward_func": 1.0,
      "step": 5410
    },
    {
      "completion_length": 258.12500953674316,
      "epoch": 0.9073305670816044,
      "grad_norm": 3.139980507250418,
      "kl": 0.21661376953125,
      "learning_rate": 4.998665431540138e-07,
      "loss": 0.0002,
      "reward": 1.7142857909202576,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7232143245637417,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5412
    },
    {
      "completion_length": 258.0803689956665,
      "epoch": 0.907665870321472,
      "grad_norm": 0.358562731594068,
      "kl": 0.0805206298828125,
      "learning_rate": 4.998661509574993e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143223285675,
      "rewards/format_reward_func": 1.0,
      "step": 5414
    },
    {
      "completion_length": 254.50893783569336,
      "epoch": 0.9080011735613396,
      "grad_norm": 0.27085475348892585,
      "kl": 0.0615692138671875,
      "learning_rate": 4.998657581857002e-07,
      "loss": 0.0001,
      "reward": 1.8107143267989159,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.810714315623045,
      "rewards/format_reward_func": 1.0,
      "step": 5416
    },
    {
      "completion_length": 239.04911994934082,
      "epoch": 0.9083364768012071,
      "grad_norm": 0.278262296727244,
      "kl": 0.0838623046875,
      "learning_rate": 4.998653648386175e-07,
      "loss": 0.0001,
      "reward": 1.703571505844593,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7035714704543352,
      "rewards/format_reward_func": 1.0,
      "step": 5418
    },
    {
      "completion_length": 240.16965293884277,
      "epoch": 0.9086717800410746,
      "grad_norm": 0.2400452957091018,
      "kl": 0.051971435546875,
      "learning_rate": 4.998649709162522e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7660714536905289,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5420
    },
    {
      "completion_length": 236.52233123779297,
      "epoch": 0.9090070832809422,
      "grad_norm": 0.00552479084519166,
      "kl": 0.05666351318359375,
      "learning_rate": 4.998645764186051e-07,
      "loss": 0.0001,
      "reward": 1.7357143461704254,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7357143331319094,
      "rewards/format_reward_func": 1.0,
      "step": 5422
    },
    {
      "completion_length": 242.33036994934082,
      "epoch": 0.9093423865208098,
      "grad_norm": 0.29723655679749067,
      "kl": 0.0811309814453125,
      "learning_rate": 4.99864181345677e-07,
      "loss": 0.0001,
      "reward": 1.7285714745521545,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7285714633762836,
      "rewards/format_reward_func": 1.0,
      "step": 5424
    },
    {
      "completion_length": 246.1607255935669,
      "epoch": 0.9096776897606773,
      "grad_norm": 0.13343396730306012,
      "kl": 0.0643768310546875,
      "learning_rate": 4.99863785697469e-07,
      "loss": 0.0001,
      "reward": 1.7785715237259865,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714454948902,
      "rewards/format_reward_func": 1.0,
      "step": 5426
    },
    {
      "completion_length": 238.2500114440918,
      "epoch": 0.9100129930005448,
      "grad_norm": 0.10802697735688958,
      "kl": 0.0711669921875,
      "learning_rate": 4.998633894739818e-07,
      "loss": 0.0001,
      "reward": 1.7767857536673546,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7812500447034836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5428
    },
    {
      "completion_length": 255.61161518096924,
      "epoch": 0.9103482962404125,
      "grad_norm": 0.15751124690918902,
      "kl": 0.06842041015625,
      "learning_rate": 4.998629926752165e-07,
      "loss": 0.0001,
      "reward": 1.725000075995922,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000294297934,
      "rewards/format_reward_func": 1.0,
      "step": 5430
    },
    {
      "completion_length": 253.02233409881592,
      "epoch": 0.91068359948028,
      "grad_norm": 0.07871958210206086,
      "kl": 0.106658935546875,
      "learning_rate": 4.998625953011739e-07,
      "loss": 0.0001,
      "reward": 1.7428571954369545,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7517857551574707,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5432
    },
    {
      "completion_length": 239.16965579986572,
      "epoch": 0.9110189027201475,
      "grad_norm": 0.17591567198150765,
      "kl": 0.04869842529296875,
      "learning_rate": 4.99862197351855e-07,
      "loss": 0.0,
      "reward": 1.7821429371833801,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 5434
    },
    {
      "completion_length": 241.68304634094238,
      "epoch": 0.911354205960015,
      "grad_norm": 0.24591806078737033,
      "kl": 0.0575103759765625,
      "learning_rate": 4.998617988272608e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000394880772,
      "rewards/format_reward_func": 1.0,
      "step": 5436
    },
    {
      "completion_length": 243.97768878936768,
      "epoch": 0.9116895091998827,
      "grad_norm": 0.2791332987456658,
      "kl": 0.064788818359375,
      "learning_rate": 4.998613997273919e-07,
      "loss": 0.0001,
      "reward": 1.742857202887535,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571842610836,
      "rewards/format_reward_func": 1.0,
      "step": 5438
    },
    {
      "completion_length": 248.4241189956665,
      "epoch": 0.9120248124397502,
      "grad_norm": 0.157219240713236,
      "kl": 0.0563507080078125,
      "learning_rate": 4.998610000522495e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 5440
    },
    {
      "completion_length": 249.42858695983887,
      "epoch": 0.9123601156796177,
      "grad_norm": 0.22393359111992003,
      "kl": 0.17645263671875,
      "learning_rate": 4.998605998018344e-07,
      "loss": 0.0002,
      "reward": 1.7160715237259865,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7205357439815998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5442
    },
    {
      "completion_length": 246.8526906967163,
      "epoch": 0.9126954189194854,
      "grad_norm": 0.22293059464193454,
      "kl": 0.0731201171875,
      "learning_rate": 4.998601989761477e-07,
      "loss": 0.0001,
      "reward": 1.7410715073347092,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7455357387661934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5444
    },
    {
      "completion_length": 232.83036708831787,
      "epoch": 0.9130307221593529,
      "grad_norm": 0.004513281336610052,
      "kl": 0.05118560791015625,
      "learning_rate": 4.9985979757519e-07,
      "loss": 0.0001,
      "reward": 1.8250000327825546,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8250000365078449,
      "rewards/format_reward_func": 1.0,
      "step": 5446
    },
    {
      "completion_length": 246.4642972946167,
      "epoch": 0.9133660253992204,
      "grad_norm": 0.28111220060756315,
      "kl": 0.11065673828125,
      "learning_rate": 4.998593955989625e-07,
      "loss": 0.0001,
      "reward": 1.7803571671247482,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.7937500290572643,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5448
    },
    {
      "completion_length": 248.61161708831787,
      "epoch": 0.9137013286390879,
      "grad_norm": 0.26719790440911256,
      "kl": 0.0706329345703125,
      "learning_rate": 4.99858993047466e-07,
      "loss": 0.0001,
      "reward": 1.8214286118745804,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8214286006987095,
      "rewards/format_reward_func": 1.0,
      "step": 5450
    },
    {
      "completion_length": 246.4598331451416,
      "epoch": 0.9140366318789556,
      "grad_norm": 0.27475256082811733,
      "kl": 0.2954254150390625,
      "learning_rate": 4.998585899207015e-07,
      "loss": 0.0003,
      "reward": 1.7500000968575478,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 5452
    },
    {
      "completion_length": 258.51787185668945,
      "epoch": 0.9143719351188231,
      "grad_norm": 0.2363268462501324,
      "kl": 0.075775146484375,
      "learning_rate": 4.998581862186698e-07,
      "loss": 0.0001,
      "reward": 1.753571480512619,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.753571467474103,
      "rewards/format_reward_func": 1.0,
      "step": 5454
    },
    {
      "completion_length": 248.1875123977661,
      "epoch": 0.9147072383586906,
      "grad_norm": 0.35017889785356593,
      "kl": 0.1490478515625,
      "learning_rate": 4.99857781941372e-07,
      "loss": 0.0001,
      "reward": 1.7357143685221672,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7357143145054579,
      "rewards/format_reward_func": 1.0,
      "step": 5456
    },
    {
      "completion_length": 247.55804538726807,
      "epoch": 0.9150425415985582,
      "grad_norm": 0.2851086680739538,
      "kl": 0.08409881591796875,
      "learning_rate": 4.998573770888089e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 5458
    },
    {
      "completion_length": 252.24108123779297,
      "epoch": 0.9153778448384258,
      "grad_norm": 0.12642756304879468,
      "kl": 0.055938720703125,
      "learning_rate": 4.998569716609815e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 5460
    },
    {
      "completion_length": 248.8571548461914,
      "epoch": 0.9157131480782933,
      "grad_norm": 0.11845124265324933,
      "kl": 0.1025543212890625,
      "learning_rate": 4.998565656578907e-07,
      "loss": 0.0001,
      "reward": 1.7517857626080513,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500350177288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5462
    },
    {
      "completion_length": 245.77679824829102,
      "epoch": 0.9160484513181608,
      "grad_norm": 0.2892291151263095,
      "kl": 0.2335357666015625,
      "learning_rate": 4.998561590795375e-07,
      "loss": 0.0002,
      "reward": 1.8107143640518188,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 5464
    },
    {
      "completion_length": 247.86608028411865,
      "epoch": 0.9163837545580285,
      "grad_norm": 0.14281412571223723,
      "kl": 0.1459503173828125,
      "learning_rate": 4.998557519259227e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 5466
    },
    {
      "completion_length": 254.5312623977661,
      "epoch": 0.916719057797896,
      "grad_norm": 0.2005936378687353,
      "kl": 0.145904541015625,
      "learning_rate": 4.998553441970474e-07,
      "loss": 0.0001,
      "reward": 1.7589286342263222,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928917348385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5468
    },
    {
      "completion_length": 239.5625114440918,
      "epoch": 0.9170543610377635,
      "grad_norm": 0.0922149462674823,
      "kl": 0.0469512939453125,
      "learning_rate": 4.998549358929124e-07,
      "loss": 0.0,
      "reward": 1.7178572341799736,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7178571783006191,
      "rewards/format_reward_func": 1.0,
      "step": 5470
    },
    {
      "completion_length": 242.3080472946167,
      "epoch": 0.9173896642776311,
      "grad_norm": 0.28370406486244265,
      "kl": 0.07550048828125,
      "learning_rate": 4.998545270135187e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 5472
    },
    {
      "completion_length": 255.2678680419922,
      "epoch": 0.9177249675174987,
      "grad_norm": 0.3061601835724448,
      "kl": 0.0880889892578125,
      "learning_rate": 4.998541175588672e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.09091372601687908,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 5474
    },
    {
      "completion_length": 258.0134038925171,
      "epoch": 0.9180602707573662,
      "grad_norm": 0.2438152833657782,
      "kl": 0.05542755126953125,
      "learning_rate": 4.998537075289589e-07,
      "loss": 0.0001,
      "reward": 1.8196429088711739,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8241071589291096,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5476
    },
    {
      "completion_length": 251.0580472946167,
      "epoch": 0.9183955739972337,
      "grad_norm": 0.3451355955311913,
      "kl": 0.1912841796875,
      "learning_rate": 4.998532969237948e-07,
      "loss": 0.0002,
      "reward": 1.7803572118282318,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5478
    },
    {
      "completion_length": 246.39287090301514,
      "epoch": 0.9187308772371013,
      "grad_norm": 0.12898136469831076,
      "kl": 0.3510284423828125,
      "learning_rate": 4.998528857433758e-07,
      "loss": 0.0004,
      "reward": 1.7732143476605415,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7776786033064127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5480
    },
    {
      "completion_length": 252.08036994934082,
      "epoch": 0.9190661804769689,
      "grad_norm": 0.198850014831597,
      "kl": 0.5845413208007812,
      "learning_rate": 4.998524739877027e-07,
      "loss": 0.0006,
      "reward": 1.7000000551342964,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7089286018162966,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5482
    },
    {
      "completion_length": 246.92411994934082,
      "epoch": 0.9194014837168364,
      "grad_norm": 0.20629485243538223,
      "kl": 0.1067047119140625,
      "learning_rate": 4.998520616567767e-07,
      "loss": 0.0001,
      "reward": 1.8142857775092125,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857365310192,
      "rewards/format_reward_func": 1.0,
      "step": 5484
    },
    {
      "completion_length": 257.71875858306885,
      "epoch": 0.919736786956704,
      "grad_norm": 0.27743791130919937,
      "kl": 0.06072998046875,
      "learning_rate": 4.998516487505985e-07,
      "loss": 0.0001,
      "reward": 1.7053572162985802,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.709821455180645,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5486
    },
    {
      "completion_length": 251.83929634094238,
      "epoch": 0.9200720901965715,
      "grad_norm": 0.26182652616513147,
      "kl": 0.3108367919921875,
      "learning_rate": 4.998512352691692e-07,
      "loss": 0.0003,
      "reward": 1.762500062584877,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5488
    },
    {
      "completion_length": 256.5044755935669,
      "epoch": 0.9204073934364391,
      "grad_norm": 0.228511123466136,
      "kl": 0.05645751953125,
      "learning_rate": 4.998508212124896e-07,
      "loss": 0.0001,
      "reward": 1.7660715132951736,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7705357223749161,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5490
    },
    {
      "completion_length": 249.4866189956665,
      "epoch": 0.9207426966763066,
      "grad_norm": 0.328227301695069,
      "kl": 0.3878936767578125,
      "learning_rate": 4.99850406580561e-07,
      "loss": 0.0004,
      "reward": 1.7857143580913544,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7857143059372902,
      "rewards/format_reward_func": 1.0,
      "step": 5492
    },
    {
      "completion_length": 251.33483600616455,
      "epoch": 0.9210779999161742,
      "grad_norm": 0.22654887015954236,
      "kl": 0.05487060546875,
      "learning_rate": 4.99849991373384e-07,
      "loss": 0.0001,
      "reward": 1.8392857685685158,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8392857350409031,
      "rewards/format_reward_func": 1.0,
      "step": 5494
    },
    {
      "completion_length": 259.1205463409424,
      "epoch": 0.9214133031560418,
      "grad_norm": 0.22116648315395465,
      "kl": 0.05206298828125,
      "learning_rate": 4.998495755909597e-07,
      "loss": 0.0001,
      "reward": 1.7732143253087997,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.777678593993187,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5496
    },
    {
      "completion_length": 249.49108123779297,
      "epoch": 0.9217486063959093,
      "grad_norm": 0.2728720425990515,
      "kl": 0.3609466552734375,
      "learning_rate": 4.998491592332891e-07,
      "loss": 0.0004,
      "reward": 1.7607143372297287,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 5498
    },
    {
      "completion_length": 253.9375114440918,
      "epoch": 0.9220839096357769,
      "grad_norm": 0.22240895206575162,
      "kl": 0.1055145263671875,
      "learning_rate": 4.99848742300373e-07,
      "loss": 0.0001,
      "reward": 1.7803572043776512,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7848214507102966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5500
    },
    {
      "completion_length": 255.65626049041748,
      "epoch": 0.9224192128756444,
      "grad_norm": 0.24399409381101786,
      "kl": 0.30101776123046875,
      "learning_rate": 4.998483247922125e-07,
      "loss": 0.0003,
      "reward": 1.7625000700354576,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643051922321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5502
    },
    {
      "completion_length": 251.9598331451416,
      "epoch": 0.922754516115512,
      "grad_norm": 0.22450025929828568,
      "kl": 0.062225341796875,
      "learning_rate": 4.998479067088085e-07,
      "loss": 0.0001,
      "reward": 1.8321429193019867,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8321428820490837,
      "rewards/format_reward_func": 1.0,
      "step": 5504
    },
    {
      "completion_length": 248.4509038925171,
      "epoch": 0.9230898193553795,
      "grad_norm": 0.3051567211790164,
      "kl": 0.3009185791015625,
      "learning_rate": 4.99847488050162e-07,
      "loss": 0.0003,
      "reward": 1.771428644657135,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 5506
    },
    {
      "completion_length": 258.60269260406494,
      "epoch": 0.9234251225952471,
      "grad_norm": 0.14467945045617023,
      "kl": 0.0584869384765625,
      "learning_rate": 4.998470688162739e-07,
      "loss": 0.0001,
      "reward": 1.7500000596046448,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000447034836,
      "rewards/format_reward_func": 1.0,
      "step": 5508
    },
    {
      "completion_length": 244.5848331451416,
      "epoch": 0.9237604258351146,
      "grad_norm": 0.2562804961215929,
      "kl": 0.14105224609375,
      "learning_rate": 4.998466490071452e-07,
      "loss": 0.0001,
      "reward": 1.7875000312924385,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7919643111526966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5510
    },
    {
      "completion_length": 263.23662281036377,
      "epoch": 0.9240957290749822,
      "grad_norm": 0.14269246424130785,
      "kl": 0.2320556640625,
      "learning_rate": 4.99846228622777e-07,
      "loss": 0.0002,
      "reward": 1.7821429148316383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 5512
    },
    {
      "completion_length": 253.27233219146729,
      "epoch": 0.9244310323148497,
      "grad_norm": 0.19061742235708712,
      "kl": 0.2888031005859375,
      "learning_rate": 4.9984580766317e-07,
      "loss": 0.0003,
      "reward": 1.764285795390606,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857581377029,
      "rewards/format_reward_func": 1.0,
      "step": 5514
    },
    {
      "completion_length": 258.8705463409424,
      "epoch": 0.9247663355547173,
      "grad_norm": 0.19888817017665691,
      "kl": 0.2613983154296875,
      "learning_rate": 4.998453861283254e-07,
      "loss": 0.0003,
      "reward": 1.723214365541935,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.736607164144516,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5516
    },
    {
      "completion_length": 261.30804538726807,
      "epoch": 0.9251016387945848,
      "grad_norm": 0.31133274406414835,
      "kl": 0.060394287109375,
      "learning_rate": 4.998449640182442e-07,
      "loss": 0.0001,
      "reward": 1.7696429416537285,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7741071730852127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5518
    },
    {
      "completion_length": 263.91519355773926,
      "epoch": 0.9254369420344524,
      "grad_norm": 0.21421569618201042,
      "kl": 0.256683349609375,
      "learning_rate": 4.998445413329271e-07,
      "loss": 0.0003,
      "reward": 1.6821429505944252,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.6821428909897804,
      "rewards/format_reward_func": 1.0,
      "step": 5520
    },
    {
      "completion_length": 255.39733219146729,
      "epoch": 0.92577224527432,
      "grad_norm": 0.2180324591587215,
      "kl": 0.2082366943359375,
      "learning_rate": 4.998441180723753e-07,
      "loss": 0.0002,
      "reward": 1.782142922282219,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7821428664028645,
      "rewards/format_reward_func": 1.0,
      "step": 5522
    },
    {
      "completion_length": 248.8437623977661,
      "epoch": 0.9261075485141875,
      "grad_norm": 0.12138142361987297,
      "kl": 0.113922119140625,
      "learning_rate": 4.998436942365896e-07,
      "loss": 0.0001,
      "reward": 1.8142857626080513,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8142857328057289,
      "rewards/format_reward_func": 1.0,
      "step": 5524
    },
    {
      "completion_length": 258.85715675354004,
      "epoch": 0.926442851754055,
      "grad_norm": 0.28283183000256845,
      "kl": 0.074371337890625,
      "learning_rate": 4.998432698255712e-07,
      "loss": 0.0001,
      "reward": 1.7446429133415222,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071783006191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5526
    },
    {
      "completion_length": 255.95090293884277,
      "epoch": 0.9267781549939226,
      "grad_norm": 0.24545344039802608,
      "kl": 0.1540679931640625,
      "learning_rate": 4.998428448393209e-07,
      "loss": 0.0002,
      "reward": 1.7553572207689285,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7598214522004128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5528
    },
    {
      "completion_length": 255.63840293884277,
      "epoch": 0.9271134582337902,
      "grad_norm": 0.2534769583940854,
      "kl": 0.1455841064453125,
      "learning_rate": 4.998424192778396e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 5530
    },
    {
      "completion_length": 250.9553680419922,
      "epoch": 0.9274487614736577,
      "grad_norm": 0.24871677747576773,
      "kl": 0.070220947265625,
      "learning_rate": 4.998419931411286e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143074274063,
      "rewards/format_reward_func": 1.0,
      "step": 5532
    },
    {
      "completion_length": 259.0491180419922,
      "epoch": 0.9277840647135253,
      "grad_norm": 0.1870510929710293,
      "kl": 0.07379150390625,
      "learning_rate": 4.998415664291887e-07,
      "loss": 0.0001,
      "reward": 1.8000000417232513,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 5534
    },
    {
      "completion_length": 258.1071529388428,
      "epoch": 0.9281193679533929,
      "grad_norm": 0.25610770723960113,
      "kl": 0.12774658203125,
      "learning_rate": 4.998411391420209e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 5536
    },
    {
      "completion_length": 245.25893688201904,
      "epoch": 0.9284546711932604,
      "grad_norm": 0.30114561591144784,
      "kl": 0.083526611328125,
      "learning_rate": 4.998407112796261e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 5538
    },
    {
      "completion_length": 241.30804920196533,
      "epoch": 0.9287899744331279,
      "grad_norm": 0.2860699239934635,
      "kl": 0.094268798828125,
      "learning_rate": 4.998402828420052e-07,
      "loss": 0.0001,
      "reward": 1.7553572058677673,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7598214671015739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5540
    },
    {
      "completion_length": 253.16072750091553,
      "epoch": 0.9291252776729955,
      "grad_norm": 0.2555147566584149,
      "kl": 0.0799713134765625,
      "learning_rate": 4.998398538291596e-07,
      "loss": 0.0001,
      "reward": 1.7071429416537285,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7071428876370192,
      "rewards/format_reward_func": 1.0,
      "step": 5542
    },
    {
      "completion_length": 244.32590293884277,
      "epoch": 0.9294605809128631,
      "grad_norm": 0.1987351795108615,
      "kl": 0.0678253173828125,
      "learning_rate": 4.998394242410899e-07,
      "loss": 0.0001,
      "reward": 1.8142857626080513,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8142857477068901,
      "rewards/format_reward_func": 1.0,
      "step": 5544
    },
    {
      "completion_length": 255.3080472946167,
      "epoch": 0.9297958841527306,
      "grad_norm": 0.31618605204352274,
      "kl": 0.121063232421875,
      "learning_rate": 4.998389940777972e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7410714626312256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5546
    },
    {
      "completion_length": 268.62501335144043,
      "epoch": 0.9301311873925981,
      "grad_norm": 0.3458136837670348,
      "kl": 0.0794830322265625,
      "learning_rate": 4.998385633392825e-07,
      "loss": 0.0001,
      "reward": 1.7285714894533157,
      "reward_std": 0.09091372694820166,
      "rewards/equation_reward_func": 0.7375000230967999,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5548
    },
    {
      "completion_length": 265.83930015563965,
      "epoch": 0.9304664906324658,
      "grad_norm": 0.2956670786981416,
      "kl": 0.2076568603515625,
      "learning_rate": 4.998381320255468e-07,
      "loss": 0.0002,
      "reward": 1.7000000551342964,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7089286111295223,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5550
    },
    {
      "completion_length": 263.7857265472412,
      "epoch": 0.9308017938723333,
      "grad_norm": 0.22854230090747885,
      "kl": 0.12713623046875,
      "learning_rate": 4.998377001365911e-07,
      "loss": 0.0001,
      "reward": 1.6910715252161026,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.6955357491970062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5552
    },
    {
      "completion_length": 265.04465675354004,
      "epoch": 0.9311370971122008,
      "grad_norm": 0.23869813596626355,
      "kl": 0.3279571533203125,
      "learning_rate": 4.998372676724164e-07,
      "loss": 0.0003,
      "reward": 1.7178572043776512,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7267857491970062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5554
    },
    {
      "completion_length": 264.2053699493408,
      "epoch": 0.9314724003520684,
      "grad_norm": 0.21032106039875323,
      "kl": 0.152618408203125,
      "learning_rate": 4.998368346330237e-07,
      "loss": 0.0002,
      "reward": 1.7339286729693413,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7473214641213417,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5556
    },
    {
      "completion_length": 266.86608600616455,
      "epoch": 0.931807703591936,
      "grad_norm": 0.2956750550588605,
      "kl": 0.080657958984375,
      "learning_rate": 4.998364010184139e-07,
      "loss": 0.0001,
      "reward": 1.7375000789761543,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7419643122702837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5558
    },
    {
      "completion_length": 258.6651887893677,
      "epoch": 0.9321430068318035,
      "grad_norm": 0.24094408593698557,
      "kl": 0.07884979248046875,
      "learning_rate": 4.99835966828588e-07,
      "loss": 0.0001,
      "reward": 1.7214286401867867,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7214285992085934,
      "rewards/format_reward_func": 1.0,
      "step": 5560
    },
    {
      "completion_length": 256.8660821914673,
      "epoch": 0.932478310071671,
      "grad_norm": 0.19758210940866042,
      "kl": 0.1866455078125,
      "learning_rate": 4.998355320635473e-07,
      "loss": 0.0002,
      "reward": 1.7857143431901932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143245637417,
      "rewards/format_reward_func": 1.0,
      "step": 5562
    },
    {
      "completion_length": 251.83929920196533,
      "epoch": 0.9328136133115387,
      "grad_norm": 0.20512783737454493,
      "kl": 0.07763671875,
      "learning_rate": 4.998350967232925e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714730620384,
      "rewards/format_reward_func": 1.0,
      "step": 5564
    },
    {
      "completion_length": 269.28126430511475,
      "epoch": 0.9331489165514062,
      "grad_norm": 0.1919354926759305,
      "kl": 0.0897216796875,
      "learning_rate": 4.998346608078245e-07,
      "loss": 0.0001,
      "reward": 1.7000000774860382,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7089286148548126,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5566
    },
    {
      "completion_length": 250.1607265472412,
      "epoch": 0.9334842197912737,
      "grad_norm": 0.10457544698349448,
      "kl": 0.0626678466796875,
      "learning_rate": 4.998342243171447e-07,
      "loss": 0.0001,
      "reward": 1.7464286237955093,
      "reward_std": 0.017677669413387775,
      "rewards/equation_reward_func": 0.754464328289032,
      "rewards/format_reward_func": 0.9919642917811871,
      "step": 5568
    },
    {
      "completion_length": 264.3973331451416,
      "epoch": 0.9338195230311412,
      "grad_norm": 0.22229885242314276,
      "kl": 0.0751495361328125,
      "learning_rate": 4.998337872512538e-07,
      "loss": 0.0001,
      "reward": 1.839285783469677,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.8482143133878708,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5570
    },
    {
      "completion_length": 270.1384057998657,
      "epoch": 0.9341548262710089,
      "grad_norm": 0.18494009709638623,
      "kl": 0.115081787109375,
      "learning_rate": 4.998333496101529e-07,
      "loss": 0.0001,
      "reward": 1.6750000789761543,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6750000324100256,
      "rewards/format_reward_func": 1.0,
      "step": 5572
    },
    {
      "completion_length": 259.0535840988159,
      "epoch": 0.9344901295108764,
      "grad_norm": 0.09959741588528473,
      "kl": 0.0759735107421875,
      "learning_rate": 4.998329113938429e-07,
      "loss": 0.0001,
      "reward": 1.700000062584877,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7000000271946192,
      "rewards/format_reward_func": 1.0,
      "step": 5574
    },
    {
      "completion_length": 252.05358505249023,
      "epoch": 0.9348254327507439,
      "grad_norm": 0.2517501606709083,
      "kl": 0.0615692138671875,
      "learning_rate": 4.998324726023249e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 5576
    },
    {
      "completion_length": 252.37500953674316,
      "epoch": 0.9351607359906116,
      "grad_norm": 0.22949889960028877,
      "kl": 0.080596923828125,
      "learning_rate": 4.998320332356001e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428753435612,
      "rewards/format_reward_func": 1.0,
      "step": 5578
    },
    {
      "completion_length": 259.05805110931396,
      "epoch": 0.9354960392304791,
      "grad_norm": 0.3646697098932177,
      "kl": 0.4967041015625,
      "learning_rate": 4.998315932936693e-07,
      "loss": 0.0005,
      "reward": 1.735714353621006,
      "reward_std": 0.11111677903681993,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5580
    },
    {
      "completion_length": 245.58036708831787,
      "epoch": 0.9358313424703466,
      "grad_norm": 0.25969544709589587,
      "kl": 0.085906982421875,
      "learning_rate": 4.998311527765334e-07,
      "loss": 0.0001,
      "reward": 1.7875000536441803,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7919643111526966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5582
    },
    {
      "completion_length": 250.86161994934082,
      "epoch": 0.9361666457102141,
      "grad_norm": 0.17861812568478064,
      "kl": 0.08895111083984375,
      "learning_rate": 4.998307116841937e-07,
      "loss": 0.0001,
      "reward": 1.7714285999536514,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 5584
    },
    {
      "completion_length": 262.96876335144043,
      "epoch": 0.9365019489500818,
      "grad_norm": 0.25962272007727116,
      "kl": 0.0623779296875,
      "learning_rate": 4.998302700166509e-07,
      "loss": 0.0001,
      "reward": 1.7232143580913544,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7267857529222965,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 5586
    },
    {
      "completion_length": 267.6384029388428,
      "epoch": 0.9368372521899493,
      "grad_norm": 0.2566924669697665,
      "kl": 0.137725830078125,
      "learning_rate": 4.998298277739063e-07,
      "loss": 0.0001,
      "reward": 1.7821428999304771,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7910714447498322,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5588
    },
    {
      "completion_length": 270.07590198516846,
      "epoch": 0.9371725554298168,
      "grad_norm": 0.2571152286005929,
      "kl": 0.11667633056640625,
      "learning_rate": 4.998293849559608e-07,
      "loss": 0.0001,
      "reward": 1.7696429044008255,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7741071842610836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5590
    },
    {
      "completion_length": 265.2142972946167,
      "epoch": 0.9375078586696844,
      "grad_norm": 0.2686328008496726,
      "kl": 0.0756072998046875,
      "learning_rate": 4.998289415628154e-07,
      "loss": 0.0001,
      "reward": 1.7500000521540642,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7589286062866449,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5592
    },
    {
      "completion_length": 262.8526916503906,
      "epoch": 0.937843161909552,
      "grad_norm": 0.2147661255591824,
      "kl": 0.1193695068359375,
      "learning_rate": 4.998284975944712e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7589286044239998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5594
    },
    {
      "completion_length": 261.5044755935669,
      "epoch": 0.9381784651494195,
      "grad_norm": 0.31342327171802786,
      "kl": 0.064178466796875,
      "learning_rate": 4.998280530509291e-07,
      "loss": 0.0001,
      "reward": 1.7464286759495735,
      "reward_std": 0.09596448857337236,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 5596
    },
    {
      "completion_length": 258.48215198516846,
      "epoch": 0.938513768389287,
      "grad_norm": 0.24752753734293131,
      "kl": 0.1938629150390625,
      "learning_rate": 4.998276079321903e-07,
      "loss": 0.0002,
      "reward": 1.7553572207689285,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7598214596509933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5598
    },
    {
      "completion_length": 251.5401906967163,
      "epoch": 0.9388490716291547,
      "grad_norm": 0.14067875762090168,
      "kl": 0.108612060546875,
      "learning_rate": 4.998271622382556e-07,
      "loss": 0.0001,
      "reward": 1.814285784959793,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8142857328057289,
      "rewards/format_reward_func": 1.0,
      "step": 5600
    },
    {
      "completion_length": 243.477689743042,
      "epoch": 0.9391843748690222,
      "grad_norm": 0.2324244712555495,
      "kl": 0.0554046630859375,
      "learning_rate": 4.998267159691262e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571429014205933,
      "rewards/format_reward_func": 1.0,
      "step": 5602
    },
    {
      "completion_length": 263.3616189956665,
      "epoch": 0.9395196781088897,
      "grad_norm": 0.19419565752990522,
      "kl": 0.233978271484375,
      "learning_rate": 4.998262691248031e-07,
      "loss": 0.0002,
      "reward": 1.7892857789993286,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857436090708,
      "rewards/format_reward_func": 1.0,
      "step": 5604
    },
    {
      "completion_length": 247.4642972946167,
      "epoch": 0.9398549813487573,
      "grad_norm": 0.1919762132952126,
      "kl": 0.1018218994140625,
      "learning_rate": 4.998258217052872e-07,
      "loss": 0.0001,
      "reward": 1.7196429148316383,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7241071797907352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5606
    },
    {
      "completion_length": 252.6919765472412,
      "epoch": 0.9401902845886249,
      "grad_norm": 0.1919943550574376,
      "kl": 0.2514801025390625,
      "learning_rate": 4.998253737105797e-07,
      "loss": 0.0003,
      "reward": 1.7714286223053932,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 5608
    },
    {
      "completion_length": 250.2366189956665,
      "epoch": 0.9405255878284924,
      "grad_norm": 0.29611682678625656,
      "kl": 0.1829986572265625,
      "learning_rate": 4.998249251406815e-07,
      "loss": 0.0002,
      "reward": 1.789285771548748,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 5610
    },
    {
      "completion_length": 263.4553680419922,
      "epoch": 0.9408608910683599,
      "grad_norm": 0.23130672291701776,
      "kl": 0.094757080078125,
      "learning_rate": 4.998244759955939e-07,
      "loss": 0.0001,
      "reward": 1.7357143461704254,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7446428909897804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5612
    },
    {
      "completion_length": 268.04019355773926,
      "epoch": 0.9411961943082275,
      "grad_norm": 0.30392357057779124,
      "kl": 0.2783203125,
      "learning_rate": 4.998240262753174e-07,
      "loss": 0.0003,
      "reward": 1.7178572490811348,
      "reward_std": 0.09596449043601751,
      "rewards/equation_reward_func": 0.7267857454717159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5614
    },
    {
      "completion_length": 257.6339387893677,
      "epoch": 0.9415314975480951,
      "grad_norm": 0.4255414977969514,
      "kl": 0.2381591796875,
      "learning_rate": 4.998235759798537e-07,
      "loss": 0.0002,
      "reward": 1.7250000536441803,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7339286021888256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5616
    },
    {
      "completion_length": 257.5401906967163,
      "epoch": 0.9418668007879626,
      "grad_norm": 0.292460391135546,
      "kl": 0.360076904296875,
      "learning_rate": 4.998231251092033e-07,
      "loss": 0.0004,
      "reward": 1.68392863124609,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.6883928906172514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5618
    },
    {
      "completion_length": 268.0848321914673,
      "epoch": 0.9422021040278302,
      "grad_norm": 0.2446423277783523,
      "kl": 0.21514892578125,
      "learning_rate": 4.998226736633675e-07,
      "loss": 0.0002,
      "reward": 1.726785808801651,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7312500327825546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5620
    },
    {
      "completion_length": 256.3392972946167,
      "epoch": 0.9425374072676977,
      "grad_norm": 0.18823448189637623,
      "kl": 0.12542724609375,
      "learning_rate": 4.998222216423472e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7616071663796902,
      "rewards/format_reward_func": 0.9919642880558968,
      "step": 5622
    },
    {
      "completion_length": 256.7053699493408,
      "epoch": 0.9428727105075653,
      "grad_norm": 0.2138723503939617,
      "kl": 0.1275787353515625,
      "learning_rate": 4.998217690461435e-07,
      "loss": 0.0001,
      "reward": 1.7410715147852898,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.7544643245637417,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5624
    },
    {
      "completion_length": 250.16072750091553,
      "epoch": 0.9432080137474328,
      "grad_norm": 0.16148957638978453,
      "kl": 0.0900421142578125,
      "learning_rate": 4.998213158747576e-07,
      "loss": 0.0001,
      "reward": 1.758928619325161,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7723214477300644,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5626
    },
    {
      "completion_length": 264.17411518096924,
      "epoch": 0.9435433169873004,
      "grad_norm": 0.30775280324654475,
      "kl": 0.090179443359375,
      "learning_rate": 4.998208621281903e-07,
      "loss": 0.0001,
      "reward": 1.7607143744826317,
      "reward_std": 0.08586296439170837,
      "rewards/equation_reward_func": 0.7696428783237934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5628
    },
    {
      "completion_length": 256.93304538726807,
      "epoch": 0.943878620227168,
      "grad_norm": 0.4164457824510602,
      "kl": 0.1088409423828125,
      "learning_rate": 4.998204078064429e-07,
      "loss": 0.0001,
      "reward": 1.737500086426735,
      "reward_std": 0.0883883461356163,
      "rewards/equation_reward_func": 0.7508928924798965,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5630
    },
    {
      "completion_length": 257.83483505249023,
      "epoch": 0.9442139234670355,
      "grad_norm": 0.18818507517460953,
      "kl": 0.1334228515625,
      "learning_rate": 4.998199529095162e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7910714596509933,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5632
    },
    {
      "completion_length": 270.9375114440918,
      "epoch": 0.944549226706903,
      "grad_norm": 0.2851531511191286,
      "kl": 0.18560791015625,
      "learning_rate": 4.998194974374113e-07,
      "loss": 0.0002,
      "reward": 1.7392857745289803,
      "reward_std": 0.0858629634603858,
      "rewards/equation_reward_func": 0.7482143193483353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5634
    },
    {
      "completion_length": 264.75001525878906,
      "epoch": 0.9448845299467706,
      "grad_norm": 0.253065033016023,
      "kl": 0.06976318359375,
      "learning_rate": 4.998190413901292e-07,
      "loss": 0.0001,
      "reward": 1.7696429267525673,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7910714596509933,
      "rewards/format_reward_func": 0.9785714372992516,
      "step": 5636
    },
    {
      "completion_length": 265.07144260406494,
      "epoch": 0.9452198331866382,
      "grad_norm": 0.23839217310733235,
      "kl": 0.061492919921875,
      "learning_rate": 4.998185847676712e-07,
      "loss": 0.0001,
      "reward": 1.7928571924567223,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571794182062,
      "rewards/format_reward_func": 1.0,
      "step": 5638
    },
    {
      "completion_length": 261.7678699493408,
      "epoch": 0.9455551364265057,
      "grad_norm": 0.19998156568588835,
      "kl": 0.1107177734375,
      "learning_rate": 4.998181275700382e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 5640
    },
    {
      "completion_length": 270.2634048461914,
      "epoch": 0.9458904396663733,
      "grad_norm": 0.25558352437163384,
      "kl": 0.1804656982421875,
      "learning_rate": 4.998176697972311e-07,
      "loss": 0.0002,
      "reward": 1.7785715013742447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 5642
    },
    {
      "completion_length": 254.98661994934082,
      "epoch": 0.9462257429062408,
      "grad_norm": 0.14052163514370772,
      "kl": 0.221527099609375,
      "learning_rate": 4.998172114492513e-07,
      "loss": 0.0002,
      "reward": 1.7535714954137802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 5644
    },
    {
      "completion_length": 257.1875123977661,
      "epoch": 0.9465610461461084,
      "grad_norm": 0.27107773602436486,
      "kl": 0.11501312255859375,
      "learning_rate": 4.998167525260994e-07,
      "loss": 0.0001,
      "reward": 1.7321429401636124,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428954601288,
      "rewards/format_reward_func": 1.0,
      "step": 5646
    },
    {
      "completion_length": 266.7857275009155,
      "epoch": 0.9468963493859759,
      "grad_norm": 0.34060283731390234,
      "kl": 0.3077545166015625,
      "learning_rate": 4.99816293027777e-07,
      "loss": 0.0003,
      "reward": 1.7089286595582962,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7133928872644901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5648
    },
    {
      "completion_length": 259.46429920196533,
      "epoch": 0.9472316526258435,
      "grad_norm": 0.32314258962676723,
      "kl": 0.0685882568359375,
      "learning_rate": 4.998158329542847e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714715719223,
      "rewards/format_reward_func": 1.0,
      "step": 5650
    },
    {
      "completion_length": 256.6651945114136,
      "epoch": 0.947566955865711,
      "grad_norm": 0.23348653009831258,
      "kl": 0.140716552734375,
      "learning_rate": 4.998153723056237e-07,
      "loss": 0.0001,
      "reward": 1.7339286506175995,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7383928932249546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5652
    },
    {
      "completion_length": 261.8705463409424,
      "epoch": 0.9479022591055786,
      "grad_norm": 0.5032103429997516,
      "kl": 0.4618377685546875,
      "learning_rate": 4.998149110817952e-07,
      "loss": 0.0005,
      "reward": 1.7785714641213417,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7875000201165676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5654
    },
    {
      "completion_length": 248.64287090301514,
      "epoch": 0.9482375623454462,
      "grad_norm": 0.26493899916194824,
      "kl": 0.0601654052734375,
      "learning_rate": 4.998144492828e-07,
      "loss": 0.0001,
      "reward": 1.7357143834233284,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7357143051922321,
      "rewards/format_reward_func": 1.0,
      "step": 5656
    },
    {
      "completion_length": 277.3750114440918,
      "epoch": 0.9485728655853137,
      "grad_norm": 0.1620975210345513,
      "kl": 0.1508026123046875,
      "learning_rate": 4.998139869086394e-07,
      "loss": 0.0002,
      "reward": 1.7053571864962578,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7276785969734192,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 5658
    },
    {
      "completion_length": 265.5803699493408,
      "epoch": 0.9489081688251813,
      "grad_norm": 0.14163608078600906,
      "kl": 0.1465911865234375,
      "learning_rate": 4.998135239593145e-07,
      "loss": 0.0001,
      "reward": 1.7517857775092125,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500461935997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5660
    },
    {
      "completion_length": 263.0134086608887,
      "epoch": 0.9492434720650488,
      "grad_norm": 0.19857088284391342,
      "kl": 0.0649871826171875,
      "learning_rate": 4.998130604348261e-07,
      "loss": 0.0001,
      "reward": 1.8357143327593803,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8357143104076385,
      "rewards/format_reward_func": 1.0,
      "step": 5662
    },
    {
      "completion_length": 264.9285840988159,
      "epoch": 0.9495787753049164,
      "grad_norm": 0.29489920317894763,
      "kl": 0.0663299560546875,
      "learning_rate": 4.998125963351754e-07,
      "loss": 0.0001,
      "reward": 1.744642935693264,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7580357417464256,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5664
    },
    {
      "completion_length": 271.7410840988159,
      "epoch": 0.9499140785447839,
      "grad_norm": 0.2585833121153054,
      "kl": 0.0933837890625,
      "learning_rate": 4.998121316603635e-07,
      "loss": 0.0001,
      "reward": 1.7107143700122833,
      "reward_std": 0.08586296439170837,
      "rewards/equation_reward_func": 0.7196428887546062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5666
    },
    {
      "completion_length": 256.977689743042,
      "epoch": 0.9502493817846515,
      "grad_norm": 0.004380553101243798,
      "kl": 0.069793701171875,
      "learning_rate": 4.998116664103914e-07,
      "loss": 0.0001,
      "reward": 1.8250000327825546,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8250000290572643,
      "rewards/format_reward_func": 1.0,
      "step": 5668
    },
    {
      "completion_length": 268.883939743042,
      "epoch": 0.9505846850245191,
      "grad_norm": 0.405321103068457,
      "kl": 0.142242431640625,
      "learning_rate": 4.998112005852603e-07,
      "loss": 0.0001,
      "reward": 1.74642863124609,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5670
    },
    {
      "completion_length": 263.4464416503906,
      "epoch": 0.9509199882643866,
      "grad_norm": 0.7514470251323826,
      "kl": 0.2141571044921875,
      "learning_rate": 4.998107341849712e-07,
      "loss": 0.0002,
      "reward": 1.789285771548748,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.789285734295845,
      "rewards/format_reward_func": 1.0,
      "step": 5672
    },
    {
      "completion_length": 262.5357255935669,
      "epoch": 0.9512552915042541,
      "grad_norm": 0.17197998736240475,
      "kl": 0.0726470947265625,
      "learning_rate": 4.998102672095251e-07,
      "loss": 0.0001,
      "reward": 1.7428571954369545,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571842610836,
      "rewards/format_reward_func": 1.0,
      "step": 5674
    },
    {
      "completion_length": 264.42411708831787,
      "epoch": 0.9515905947441217,
      "grad_norm": 0.24326647158367354,
      "kl": 0.080718994140625,
      "learning_rate": 4.998097996589233e-07,
      "loss": 0.0001,
      "reward": 1.7732143625617027,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7776786051690578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5676
    },
    {
      "completion_length": 266.07143783569336,
      "epoch": 0.9519258979839893,
      "grad_norm": 0.4137675130766855,
      "kl": 0.101409912109375,
      "learning_rate": 4.998093315331665e-07,
      "loss": 0.0001,
      "reward": 1.7950893491506577,
      "reward_std": 0.057452425360679626,
      "rewards/equation_reward_func": 0.7964286096394062,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 5678
    },
    {
      "completion_length": 256.8973331451416,
      "epoch": 0.9522612012238568,
      "grad_norm": 0.23364485897135776,
      "kl": 0.0838165283203125,
      "learning_rate": 4.998088628322562e-07,
      "loss": 0.0001,
      "reward": 1.8321429342031479,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8321428820490837,
      "rewards/format_reward_func": 1.0,
      "step": 5680
    },
    {
      "completion_length": 269.2500114440918,
      "epoch": 0.9525965044637243,
      "grad_norm": 0.44124717033408517,
      "kl": 0.0906219482421875,
      "learning_rate": 4.998083935561932e-07,
      "loss": 0.0001,
      "reward": 1.7196429148316383,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7241071835160255,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5682
    },
    {
      "completion_length": 265.31251430511475,
      "epoch": 0.952931807703592,
      "grad_norm": 0.1912833560793976,
      "kl": 0.07427978515625,
      "learning_rate": 4.998079237049785e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714525729418,
      "rewards/format_reward_func": 1.0,
      "step": 5684
    },
    {
      "completion_length": 268.4509086608887,
      "epoch": 0.9532671109434595,
      "grad_norm": 0.12658711226962915,
      "kl": 0.06768798828125,
      "learning_rate": 4.998074532786135e-07,
      "loss": 0.0001,
      "reward": 1.7071429416537285,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7071428932249546,
      "rewards/format_reward_func": 1.0,
      "step": 5686
    },
    {
      "completion_length": 283.089298248291,
      "epoch": 0.953602414183327,
      "grad_norm": 0.23460052009656376,
      "kl": 0.073516845703125,
      "learning_rate": 4.998069822770992e-07,
      "loss": 0.0001,
      "reward": 1.7303571924567223,
      "reward_std": 0.06313453428447247,
      "rewards/equation_reward_func": 0.7526785936206579,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 5688
    },
    {
      "completion_length": 281.93751430511475,
      "epoch": 0.9539377174231946,
      "grad_norm": 0.18191125932432262,
      "kl": 0.0916290283203125,
      "learning_rate": 4.998065107004365e-07,
      "loss": 0.0001,
      "reward": 1.6625000536441803,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.6848214641213417,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 5690
    },
    {
      "completion_length": 281.11162281036377,
      "epoch": 0.9542730206630622,
      "grad_norm": 0.2272203925271211,
      "kl": 0.1111297607421875,
      "learning_rate": 4.998060385486265e-07,
      "loss": 0.0001,
      "reward": 1.7339286729693413,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7383928932249546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5692
    },
    {
      "completion_length": 280.6964406967163,
      "epoch": 0.9546083239029297,
      "grad_norm": 0.12683642382177307,
      "kl": 0.086151123046875,
      "learning_rate": 4.998055658216705e-07,
      "loss": 0.0001,
      "reward": 1.7482143267989159,
      "reward_std": 0.0328299580141902,
      "rewards/equation_reward_func": 0.7526785954833031,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5694
    },
    {
      "completion_length": 282.22769260406494,
      "epoch": 0.9549436271427972,
      "grad_norm": 0.33633389268441777,
      "kl": 0.1428070068359375,
      "learning_rate": 4.998050925195694e-07,
      "loss": 0.0001,
      "reward": 1.7125000655651093,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.7348214611411095,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 5696
    },
    {
      "completion_length": 277.4687614440918,
      "epoch": 0.9552789303826649,
      "grad_norm": 0.2182738670209157,
      "kl": 0.1761627197265625,
      "learning_rate": 4.998046186423243e-07,
      "loss": 0.0002,
      "reward": 1.7625000774860382,
      "reward_std": 0.09343911055475473,
      "rewards/equation_reward_func": 0.7758928760886192,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5698
    },
    {
      "completion_length": 272.46876430511475,
      "epoch": 0.9556142336225324,
      "grad_norm": 0.2382541500873255,
      "kl": 0.0994110107421875,
      "learning_rate": 4.998041441899365e-07,
      "loss": 0.0001,
      "reward": 1.7589286416769028,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7633928842842579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5700
    },
    {
      "completion_length": 277.77680110931396,
      "epoch": 0.9559495368623999,
      "grad_norm": 0.17786839241166985,
      "kl": 0.1782073974609375,
      "learning_rate": 4.998036691624069e-07,
      "loss": 0.0002,
      "reward": 1.7571429312229156,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 5702
    },
    {
      "completion_length": 280.6607265472412,
      "epoch": 0.9562848401022674,
      "grad_norm": 0.44365582048278357,
      "kl": 0.2714080810546875,
      "learning_rate": 4.998031935597366e-07,
      "loss": 0.0003,
      "reward": 1.698214329779148,
      "reward_std": 0.09343911055475473,
      "rewards/equation_reward_func": 0.7116071823984385,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5704
    },
    {
      "completion_length": 276.37501335144043,
      "epoch": 0.9566201433421351,
      "grad_norm": 0.15799020884293222,
      "kl": 0.0831298828125,
      "learning_rate": 4.998027173819268e-07,
      "loss": 0.0001,
      "reward": 1.7767857685685158,
      "reward_std": 0.05303300637751818,
      "rewards/equation_reward_func": 0.7901786006987095,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5706
    },
    {
      "completion_length": 274.84822845458984,
      "epoch": 0.9569554465820026,
      "grad_norm": 0.22140325800096408,
      "kl": 0.145477294921875,
      "learning_rate": 4.998022406289784e-07,
      "loss": 0.0001,
      "reward": 1.7053572311997414,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7098214514553547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5708
    },
    {
      "completion_length": 275.5937623977661,
      "epoch": 0.9572907498218701,
      "grad_norm": 0.43758406981647235,
      "kl": 0.108306884765625,
      "learning_rate": 4.998017633008928e-07,
      "loss": 0.0001,
      "reward": 1.6910715103149414,
      "reward_std": 0.0883883461356163,
      "rewards/equation_reward_func": 0.7133928909897804,
      "rewards/format_reward_func": 0.9776785783469677,
      "step": 5710
    },
    {
      "completion_length": 269.6205520629883,
      "epoch": 0.9576260530617378,
      "grad_norm": 0.09696601262183618,
      "kl": 0.07403564453125,
      "learning_rate": 4.998012853976707e-07,
      "loss": 0.0001,
      "reward": 1.737500086426735,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7419643178582191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5712
    },
    {
      "completion_length": 271.7009057998657,
      "epoch": 0.9579613563016053,
      "grad_norm": 0.3847634166372032,
      "kl": 0.0723419189453125,
      "learning_rate": 4.998008069193136e-07,
      "loss": 0.0001,
      "reward": 1.775000050663948,
      "reward_std": 0.11616754438728094,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5714
    },
    {
      "completion_length": 253.88840293884277,
      "epoch": 0.9582966595414728,
      "grad_norm": 0.2766356278864402,
      "kl": 0.07025146484375,
      "learning_rate": 4.998003278658222e-07,
      "loss": 0.0001,
      "reward": 1.7446429207921028,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.7660714499652386,
      "rewards/format_reward_func": 0.9785714447498322,
      "step": 5716
    },
    {
      "completion_length": 269.3750114440918,
      "epoch": 0.9586319627813403,
      "grad_norm": 0.31884468339624056,
      "kl": 0.077911376953125,
      "learning_rate": 4.997998482371981e-07,
      "loss": 0.0001,
      "reward": 1.6964286863803864,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.6964285932481289,
      "rewards/format_reward_func": 1.0,
      "step": 5718
    },
    {
      "completion_length": 267.9509057998657,
      "epoch": 0.958967266021208,
      "grad_norm": 0.15281834104888092,
      "kl": 0.081817626953125,
      "learning_rate": 4.99799368033442e-07,
      "loss": 0.0001,
      "reward": 1.700000062584877,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7089286092668772,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5720
    },
    {
      "completion_length": 265.84376430511475,
      "epoch": 0.9593025692610755,
      "grad_norm": 0.13977553928490638,
      "kl": 0.0774688720703125,
      "learning_rate": 4.997988872545551e-07,
      "loss": 0.0001,
      "reward": 1.7446429207921028,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071801632643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5722
    },
    {
      "completion_length": 258.8437614440918,
      "epoch": 0.959637872500943,
      "grad_norm": 0.0732864545892103,
      "kl": 0.0887908935546875,
      "learning_rate": 4.997984059005386e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 5724
    },
    {
      "completion_length": 256.7321548461914,
      "epoch": 0.9599731757408106,
      "grad_norm": 0.1907927548631042,
      "kl": 0.078094482421875,
      "learning_rate": 4.997979239713935e-07,
      "loss": 0.0001,
      "reward": 1.8214286267757416,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8214286044239998,
      "rewards/format_reward_func": 1.0,
      "step": 5726
    },
    {
      "completion_length": 252.30804443359375,
      "epoch": 0.9603084789806782,
      "grad_norm": 0.5063146778249406,
      "kl": 0.0906982421875,
      "learning_rate": 4.997974414671211e-07,
      "loss": 0.0001,
      "reward": 1.698214367032051,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.702678607776761,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5728
    },
    {
      "completion_length": 259.276798248291,
      "epoch": 0.9606437822205457,
      "grad_norm": 0.3635133661797194,
      "kl": 0.1004638671875,
      "learning_rate": 4.997969583877223e-07,
      "loss": 0.0001,
      "reward": 1.7357143461704254,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7446428909897804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5730
    },
    {
      "completion_length": 252.36162090301514,
      "epoch": 0.9609790854604132,
      "grad_norm": 0.28713514263695467,
      "kl": 0.0994110107421875,
      "learning_rate": 4.997964747331982e-07,
      "loss": 0.0001,
      "reward": 1.742857202887535,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7428571749478579,
      "rewards/format_reward_func": 1.0,
      "step": 5732
    },
    {
      "completion_length": 250.84375953674316,
      "epoch": 0.9613143887002809,
      "grad_norm": 0.2510492779487852,
      "kl": 0.250762939453125,
      "learning_rate": 4.9979599050355e-07,
      "loss": 0.0003,
      "reward": 1.723214365541935,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7276786044239998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5734
    },
    {
      "completion_length": 249.1830472946167,
      "epoch": 0.9616496919401484,
      "grad_norm": 0.22425040779164032,
      "kl": 0.0988311767578125,
      "learning_rate": 4.99795505698779e-07,
      "loss": 0.0001,
      "reward": 1.7486607730388641,
      "reward_std": 0.05240166233852506,
      "rewards/equation_reward_func": 0.7500000223517418,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 5736
    },
    {
      "completion_length": 242.62947845458984,
      "epoch": 0.9619849951800159,
      "grad_norm": 0.23321105934828876,
      "kl": 0.093170166015625,
      "learning_rate": 4.997950203188859e-07,
      "loss": 0.0001,
      "reward": 1.825000062584877,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8250000290572643,
      "rewards/format_reward_func": 1.0,
      "step": 5738
    },
    {
      "completion_length": 252.52233600616455,
      "epoch": 0.9623202984198835,
      "grad_norm": 1.1336488889710032,
      "kl": 0.0902557373046875,
      "learning_rate": 4.997945343638721e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.771428594365716,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5740
    },
    {
      "completion_length": 238.45983123779297,
      "epoch": 0.9626556016597511,
      "grad_norm": 0.08074561960104147,
      "kl": 0.0883331298828125,
      "learning_rate": 4.997940478337387e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.03030457627028227,
      "rewards/equation_reward_func": 0.7660714685916901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5742
    },
    {
      "completion_length": 241.31251049041748,
      "epoch": 0.9629909048996186,
      "grad_norm": 0.16187012849684024,
      "kl": 0.0847930908203125,
      "learning_rate": 4.997935607284869e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.8196428641676903,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5744
    },
    {
      "completion_length": 243.7232255935669,
      "epoch": 0.9633262081394861,
      "grad_norm": 0.7833918591829895,
      "kl": 0.11737060546875,
      "learning_rate": 4.997930730481175e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7660714499652386,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5746
    },
    {
      "completion_length": 254.01786994934082,
      "epoch": 0.9636615113793537,
      "grad_norm": 0.2893503495802617,
      "kl": 0.0744171142578125,
      "learning_rate": 4.99792584792632e-07,
      "loss": 0.0001,
      "reward": 1.7214286178350449,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7303571961820126,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5748
    },
    {
      "completion_length": 247.45090579986572,
      "epoch": 0.9639968146192213,
      "grad_norm": 0.21552202774201076,
      "kl": 0.076690673828125,
      "learning_rate": 4.997920959620312e-07,
      "loss": 0.0001,
      "reward": 1.7517857626080513,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7562500312924385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5750
    },
    {
      "completion_length": 228.21876049041748,
      "epoch": 0.9643321178590888,
      "grad_norm": 0.17294579698170476,
      "kl": 0.0935211181640625,
      "learning_rate": 4.997916065563164e-07,
      "loss": 0.0001,
      "reward": 1.7767857611179352,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7812500298023224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5752
    },
    {
      "completion_length": 237.68304824829102,
      "epoch": 0.9646674210989564,
      "grad_norm": 0.2319434648635915,
      "kl": 0.148101806640625,
      "learning_rate": 4.997911165754888e-07,
      "loss": 0.0001,
      "reward": 1.7357143685221672,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7357142902910709,
      "rewards/format_reward_func": 1.0,
      "step": 5754
    },
    {
      "completion_length": 235.92858123779297,
      "epoch": 0.965002724338824,
      "grad_norm": 0.25800015070654836,
      "kl": 0.1013336181640625,
      "learning_rate": 4.997906260195495e-07,
      "loss": 0.0001,
      "reward": 1.7839286401867867,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7883928865194321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5756
    },
    {
      "completion_length": 227.48661613464355,
      "epoch": 0.9653380275786915,
      "grad_norm": 0.5387041435019756,
      "kl": 0.2335357666015625,
      "learning_rate": 4.997901348884994e-07,
      "loss": 0.0002,
      "reward": 1.8125000670552254,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8169643171131611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5758
    },
    {
      "completion_length": 255.6071538925171,
      "epoch": 0.965673330818559,
      "grad_norm": 0.9829855590547008,
      "kl": 0.515045166015625,
      "learning_rate": 4.997896431823398e-07,
      "loss": 0.0005,
      "reward": 1.6928571909666061,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7107143253087997,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5760
    },
    {
      "completion_length": 249.9062623977661,
      "epoch": 0.9660086340584266,
      "grad_norm": 0.4302051737915668,
      "kl": 0.381072998046875,
      "learning_rate": 4.997891509010719e-07,
      "loss": 0.0004,
      "reward": 1.7964286357164383,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.8053571581840515,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5762
    },
    {
      "completion_length": 251.86161708831787,
      "epoch": 0.9663439372982942,
      "grad_norm": 0.2003740166787335,
      "kl": 0.1812744140625,
      "learning_rate": 4.997886580446968e-07,
      "loss": 0.0002,
      "reward": 1.7642857730388641,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7821428701281548,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5764
    },
    {
      "completion_length": 236.9910831451416,
      "epoch": 0.9666792405381617,
      "grad_norm": 0.30823609277606523,
      "kl": 0.79034423828125,
      "learning_rate": 4.997881646132154e-07,
      "loss": 0.0008,
      "reward": 1.7732143551111221,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.777678593993187,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5766
    },
    {
      "completion_length": 258.9464406967163,
      "epoch": 0.9670145437780292,
      "grad_norm": 0.4599728212819268,
      "kl": 4.020721435546875,
      "learning_rate": 4.997876706066293e-07,
      "loss": 0.004,
      "reward": 1.7410714775323868,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7455357424914837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5768
    },
    {
      "completion_length": 255.64733028411865,
      "epoch": 0.9673498470178968,
      "grad_norm": 1.0446450602846626,
      "kl": 4.951690673828125,
      "learning_rate": 4.997871760249392e-07,
      "loss": 0.005,
      "reward": 1.6772322282195091,
      "reward_std": 0.1433153918478638,
      "rewards/equation_reward_func": 0.7142857499420643,
      "rewards/format_reward_func": 0.9629464447498322,
      "step": 5770
    },
    {
      "completion_length": 256.3928699493408,
      "epoch": 0.9676851502577644,
      "grad_norm": 0.652762162620934,
      "kl": 1.116546630859375,
      "learning_rate": 4.997866808681464e-07,
      "loss": 0.0011,
      "reward": 1.6901786252856255,
      "reward_std": 0.14773480826988816,
      "rewards/equation_reward_func": 0.7366071753203869,
      "rewards/format_reward_func": 0.9535714499652386,
      "step": 5772
    },
    {
      "completion_length": 260.3928680419922,
      "epoch": 0.9680204534976319,
      "grad_norm": 0.4629856359890017,
      "kl": 1.8438873291015625,
      "learning_rate": 4.997861851362522e-07,
      "loss": 0.0018,
      "reward": 1.7361607551574707,
      "reward_std": 0.12058695964515209,
      "rewards/equation_reward_func": 0.7598214652389288,
      "rewards/format_reward_func": 0.9763392992317677,
      "step": 5774
    },
    {
      "completion_length": 258.7410840988159,
      "epoch": 0.9683557567374995,
      "grad_norm": 0.2744137856877816,
      "kl": 0.482940673828125,
      "learning_rate": 4.997856888292575e-07,
      "loss": 0.0005,
      "reward": 1.7428571954369545,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5776
    },
    {
      "completion_length": 255.73661613464355,
      "epoch": 0.968691059977367,
      "grad_norm": 0.1992131837536013,
      "kl": 0.2537994384765625,
      "learning_rate": 4.997851919471634e-07,
      "loss": 0.0003,
      "reward": 1.6647322177886963,
      "reward_std": 0.07007933338172734,
      "rewards/equation_reward_func": 0.6794643215835094,
      "rewards/format_reward_func": 0.9852678664028645,
      "step": 5778
    },
    {
      "completion_length": 243.08929824829102,
      "epoch": 0.9690263632172346,
      "grad_norm": 0.2374438613904606,
      "kl": 0.1378021240234375,
      "learning_rate": 4.997846944899713e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.0858629671856761,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5780
    },
    {
      "completion_length": 230.6116189956665,
      "epoch": 0.9693616664571021,
      "grad_norm": 0.19207024544444537,
      "kl": 0.1244354248046875,
      "learning_rate": 4.997841964576822e-07,
      "loss": 0.0001,
      "reward": 1.7482143267989159,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7526786141097546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5782
    },
    {
      "completion_length": 234.8303680419922,
      "epoch": 0.9696969696969697,
      "grad_norm": 0.29969401860216555,
      "kl": 0.1541748046875,
      "learning_rate": 4.997836978502973e-07,
      "loss": 0.0002,
      "reward": 1.7196429297327995,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7330357395112514,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5784
    },
    {
      "completion_length": 235.4687614440918,
      "epoch": 0.9700322729368372,
      "grad_norm": 0.0050095061490047785,
      "kl": 0.1041717529296875,
      "learning_rate": 4.997831986678177e-07,
      "loss": 0.0001,
      "reward": 1.7397321984171867,
      "reward_std": 0.034723992459475994,
      "rewards/equation_reward_func": 0.745535746216774,
      "rewards/format_reward_func": 0.994196429848671,
      "step": 5786
    },
    {
      "completion_length": 233.54465007781982,
      "epoch": 0.9703675761767048,
      "grad_norm": 0.4226043906249522,
      "kl": 0.1141204833984375,
      "learning_rate": 4.997826989102445e-07,
      "loss": 0.0001,
      "reward": 1.7517857775092125,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500331550837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5788
    },
    {
      "completion_length": 235.62501049041748,
      "epoch": 0.9707028794165724,
      "grad_norm": 0.14606378420437877,
      "kl": 0.11199951171875,
      "learning_rate": 4.997821985775789e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 5790
    },
    {
      "completion_length": 224.12054443359375,
      "epoch": 0.9710381826564399,
      "grad_norm": 0.4096506818257736,
      "kl": 0.1255035400390625,
      "learning_rate": 4.997816976698222e-07,
      "loss": 0.0001,
      "reward": 1.8000000417232513,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 5792
    },
    {
      "completion_length": 231.3482255935669,
      "epoch": 0.9713734858963075,
      "grad_norm": 0.3583328866620858,
      "kl": 0.137237548828125,
      "learning_rate": 4.997811961869754e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7589285895228386,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5794
    },
    {
      "completion_length": 228.71429538726807,
      "epoch": 0.971708789136175,
      "grad_norm": 0.20583662808869987,
      "kl": 0.106781005859375,
      "learning_rate": 4.997806941290396e-07,
      "loss": 0.0001,
      "reward": 1.7142857909202576,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857406288385,
      "rewards/format_reward_func": 1.0,
      "step": 5796
    },
    {
      "completion_length": 222.55804634094238,
      "epoch": 0.9720440923760426,
      "grad_norm": 0.17532440930536886,
      "kl": 0.09112548828125,
      "learning_rate": 4.99780191496016e-07,
      "loss": 0.0001,
      "reward": 1.725000061094761,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.725000036880374,
      "rewards/format_reward_func": 1.0,
      "step": 5798
    },
    {
      "completion_length": 235.87501049041748,
      "epoch": 0.9723793956159101,
      "grad_norm": 0.31365165812364126,
      "kl": 0.09649658203125,
      "learning_rate": 4.997796882879058e-07,
      "loss": 0.0001,
      "reward": 1.764285758137703,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7732143122702837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5800
    },
    {
      "completion_length": 223.37500953674316,
      "epoch": 0.9727146988557777,
      "grad_norm": 0.3106940012934583,
      "kl": 0.104827880859375,
      "learning_rate": 4.997791845047102e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 5802
    },
    {
      "completion_length": 218.8616180419922,
      "epoch": 0.9730500020956453,
      "grad_norm": 0.22578401002909218,
      "kl": 0.09271240234375,
      "learning_rate": 4.997786801464303e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714626312256,
      "rewards/format_reward_func": 1.0,
      "step": 5804
    },
    {
      "completion_length": 220.16518592834473,
      "epoch": 0.9733853053355128,
      "grad_norm": 0.331889824862444,
      "kl": 0.08624267578125,
      "learning_rate": 4.997781752130673e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428962051868,
      "rewards/format_reward_func": 1.0,
      "step": 5806
    },
    {
      "completion_length": 224.34822463989258,
      "epoch": 0.9737206085753803,
      "grad_norm": 0.21284460310280223,
      "kl": 0.1049041748046875,
      "learning_rate": 4.997776697046223e-07,
      "loss": 0.0001,
      "reward": 1.8125000521540642,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8169643059372902,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5808
    },
    {
      "completion_length": 221.4553680419922,
      "epoch": 0.9740559118152479,
      "grad_norm": 0.20490578484408217,
      "kl": 0.07861328125,
      "learning_rate": 4.997771636210965e-07,
      "loss": 0.0001,
      "reward": 1.783928632736206,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7883928865194321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5810
    },
    {
      "completion_length": 225.59822463989258,
      "epoch": 0.9743912150551155,
      "grad_norm": 0.33365690223263317,
      "kl": 0.0732269287109375,
      "learning_rate": 4.99776656962491e-07,
      "loss": 0.0001,
      "reward": 1.7892857566475868,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857436090708,
      "rewards/format_reward_func": 1.0,
      "step": 5812
    },
    {
      "completion_length": 225.36161422729492,
      "epoch": 0.974726518294983,
      "grad_norm": 0.2679487209690672,
      "kl": 0.0830841064453125,
      "learning_rate": 4.997761497288071e-07,
      "loss": 0.0001,
      "reward": 1.7607143372297287,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7696428969502449,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5814
    },
    {
      "completion_length": 218.46429634094238,
      "epoch": 0.9750618215348505,
      "grad_norm": 0.3324314294312522,
      "kl": 0.0901641845703125,
      "learning_rate": 4.997756419200458e-07,
      "loss": 0.0001,
      "reward": 1.796428643167019,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 5816
    },
    {
      "completion_length": 229.84375858306885,
      "epoch": 0.9753971247747182,
      "grad_norm": 0.18266863033969682,
      "kl": 0.075103759765625,
      "learning_rate": 4.997751335362085e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 5818
    },
    {
      "completion_length": 225.60715293884277,
      "epoch": 0.9757324280145857,
      "grad_norm": 0.3062540378594742,
      "kl": 0.079925537109375,
      "learning_rate": 4.997746245772962e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143074274063,
      "rewards/format_reward_func": 1.0,
      "step": 5820
    },
    {
      "completion_length": 232.3035831451416,
      "epoch": 0.9760677312544532,
      "grad_norm": 0.26651813698518534,
      "kl": 0.07159423828125,
      "learning_rate": 4.9977411504331e-07,
      "loss": 0.0001,
      "reward": 1.6892857998609543,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.6892857421189547,
      "rewards/format_reward_func": 1.0,
      "step": 5822
    },
    {
      "completion_length": 228.92411708831787,
      "epoch": 0.9764030344943208,
      "grad_norm": 0.2276887505675821,
      "kl": 0.0889892578125,
      "learning_rate": 4.997736049342512e-07,
      "loss": 0.0001,
      "reward": 1.7642857655882835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 5824
    },
    {
      "completion_length": 233.46875953674316,
      "epoch": 0.9767383377341884,
      "grad_norm": 0.07011180696246816,
      "kl": 0.0858306884765625,
      "learning_rate": 4.997730942501211e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 5826
    },
    {
      "completion_length": 234.75893878936768,
      "epoch": 0.9770736409740559,
      "grad_norm": 0.5499560677708243,
      "kl": 0.0884246826171875,
      "learning_rate": 4.997725829909205e-07,
      "loss": 0.0001,
      "reward": 1.755357213318348,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.759821455925703,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5828
    },
    {
      "completion_length": 228.28125953674316,
      "epoch": 0.9774089442139234,
      "grad_norm": 0.2858952293699924,
      "kl": 0.0711822509765625,
      "learning_rate": 4.99772071156651e-07,
      "loss": 0.0001,
      "reward": 1.760714367032051,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7607143204659224,
      "rewards/format_reward_func": 1.0,
      "step": 5830
    },
    {
      "completion_length": 233.47768878936768,
      "epoch": 0.9777442474537911,
      "grad_norm": 0.252328997483877,
      "kl": 0.0720367431640625,
      "learning_rate": 4.997715587473135e-07,
      "loss": 0.0001,
      "reward": 1.7267857864499092,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.731250025331974,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5832
    },
    {
      "completion_length": 234.97322273254395,
      "epoch": 0.9780795506936586,
      "grad_norm": 0.2657323747803113,
      "kl": 0.0772247314453125,
      "learning_rate": 4.997710457629092e-07,
      "loss": 0.0001,
      "reward": 1.7446429133415222,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071857511997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5834
    },
    {
      "completion_length": 228.12054443359375,
      "epoch": 0.9784148539335261,
      "grad_norm": 0.00524528276891892,
      "kl": 0.0808563232421875,
      "learning_rate": 4.997705322034394e-07,
      "loss": 0.0001,
      "reward": 1.7732143625617027,
      "reward_std": 0.027779195457696915,
      "rewards/equation_reward_func": 0.7776786014437675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5836
    },
    {
      "completion_length": 238.71876049041748,
      "epoch": 0.9787501571733936,
      "grad_norm": 0.28480684344520696,
      "kl": 0.076904296875,
      "learning_rate": 4.997700180689053e-07,
      "loss": 0.0001,
      "reward": 1.7964286357164383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7964285872876644,
      "rewards/format_reward_func": 1.0,
      "step": 5838
    },
    {
      "completion_length": 227.2366180419922,
      "epoch": 0.9790854604132613,
      "grad_norm": 0.3515029048482449,
      "kl": 0.078582763671875,
      "learning_rate": 4.99769503359308e-07,
      "loss": 0.0001,
      "reward": 1.7357143759727478,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 5840
    },
    {
      "completion_length": 234.6830472946167,
      "epoch": 0.9794207636531288,
      "grad_norm": 0.1499024663881233,
      "kl": 0.073211669921875,
      "learning_rate": 4.997689880746486e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714730620384,
      "rewards/format_reward_func": 1.0,
      "step": 5842
    },
    {
      "completion_length": 229.09376049041748,
      "epoch": 0.9797560668929963,
      "grad_norm": 0.22930478631284873,
      "kl": 0.05889892578125,
      "learning_rate": 4.997684722149284e-07,
      "loss": 0.0001,
      "reward": 1.753571480512619,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714600235224,
      "rewards/format_reward_func": 1.0,
      "step": 5844
    },
    {
      "completion_length": 228.95983219146729,
      "epoch": 0.980091370132864,
      "grad_norm": 0.20276460790670736,
      "kl": 0.09027099609375,
      "learning_rate": 4.997679557801487e-07,
      "loss": 0.0001,
      "reward": 1.814285784959793,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857365310192,
      "rewards/format_reward_func": 1.0,
      "step": 5846
    },
    {
      "completion_length": 223.62947463989258,
      "epoch": 0.9804266733727315,
      "grad_norm": 0.23756297896683548,
      "kl": 0.0641937255859375,
      "learning_rate": 4.997674387703104e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 5848
    },
    {
      "completion_length": 214.40626049041748,
      "epoch": 0.980761976612599,
      "grad_norm": 0.24790898194269112,
      "kl": 0.0617218017578125,
      "learning_rate": 4.997669211854148e-07,
      "loss": 0.0001,
      "reward": 1.8214286491274834,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8214285895228386,
      "rewards/format_reward_func": 1.0,
      "step": 5850
    },
    {
      "completion_length": 229.6919755935669,
      "epoch": 0.9810972798524665,
      "grad_norm": 0.45248769291328733,
      "kl": 0.089996337890625,
      "learning_rate": 4.997664030254634e-07,
      "loss": 0.0001,
      "reward": 1.7446429505944252,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7482143118977547,
      "rewards/format_reward_func": 0.9964285790920258,
      "step": 5852
    },
    {
      "completion_length": 233.15625858306885,
      "epoch": 0.9814325830923342,
      "grad_norm": 0.17702506587319505,
      "kl": 0.0703277587890625,
      "learning_rate": 4.99765884290457e-07,
      "loss": 0.0001,
      "reward": 1.7517857775092125,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7651786021888256,
      "rewards/format_reward_func": 0.9866071455180645,
      "step": 5854
    },
    {
      "completion_length": 231.97322463989258,
      "epoch": 0.9817678863322017,
      "grad_norm": 0.25418976598809995,
      "kl": 0.06585693359375,
      "learning_rate": 4.997653649803968e-07,
      "loss": 0.0001,
      "reward": 1.74642863124609,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.746428593993187,
      "rewards/format_reward_func": 1.0,
      "step": 5856
    },
    {
      "completion_length": 234.37054634094238,
      "epoch": 0.9821031895720692,
      "grad_norm": 0.18552869148200038,
      "kl": 0.09344482421875,
      "learning_rate": 4.997648450952842e-07,
      "loss": 0.0001,
      "reward": 1.792857214808464,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571589291096,
      "rewards/format_reward_func": 1.0,
      "step": 5858
    },
    {
      "completion_length": 245.8839406967163,
      "epoch": 0.9824384928119368,
      "grad_norm": 0.2988328236976148,
      "kl": 0.0713958740234375,
      "learning_rate": 4.997643246351204e-07,
      "loss": 0.0001,
      "reward": 1.7303571999073029,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7348214760422707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5860
    },
    {
      "completion_length": 236.5625123977661,
      "epoch": 0.9827737960518044,
      "grad_norm": 0.09087717881506024,
      "kl": 0.06475830078125,
      "learning_rate": 4.997638035999065e-07,
      "loss": 0.0001,
      "reward": 1.7285715341567993,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714671015739,
      "rewards/format_reward_func": 1.0,
      "step": 5862
    },
    {
      "completion_length": 240.37501335144043,
      "epoch": 0.9831090992916719,
      "grad_norm": 0.2330972837811373,
      "kl": 0.073760986328125,
      "learning_rate": 4.997632819896437e-07,
      "loss": 0.0001,
      "reward": 1.7928572446107864,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7928571626543999,
      "rewards/format_reward_func": 1.0,
      "step": 5864
    },
    {
      "completion_length": 242.43750953674316,
      "epoch": 0.9834444025315394,
      "grad_norm": 0.3031159260506758,
      "kl": 0.0743408203125,
      "learning_rate": 4.997627598043331e-07,
      "loss": 0.0001,
      "reward": 1.7392858043313026,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857447266579,
      "rewards/format_reward_func": 1.0,
      "step": 5866
    },
    {
      "completion_length": 228.8616180419922,
      "epoch": 0.983779705771407,
      "grad_norm": 0.21641965125787252,
      "kl": 0.082244873046875,
      "learning_rate": 4.997622370439762e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 5868
    },
    {
      "completion_length": 243.7634048461914,
      "epoch": 0.9841150090112746,
      "grad_norm": 0.17336499937615538,
      "kl": 0.068206787109375,
      "learning_rate": 4.99761713708574e-07,
      "loss": 0.0001,
      "reward": 1.742857202887535,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571656346321,
      "rewards/format_reward_func": 1.0,
      "step": 5870
    },
    {
      "completion_length": 241.12947463989258,
      "epoch": 0.9844503122511421,
      "grad_norm": 0.2895336684874953,
      "kl": 0.0819244384765625,
      "learning_rate": 4.997611897981277e-07,
      "loss": 0.0001,
      "reward": 1.7821428999304771,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7910714596509933,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5872
    },
    {
      "completion_length": 241.39733123779297,
      "epoch": 0.9847856154910097,
      "grad_norm": 0.2139574550323431,
      "kl": 0.0740966796875,
      "learning_rate": 4.997606653126385e-07,
      "loss": 0.0001,
      "reward": 1.7928572222590446,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 5874
    },
    {
      "completion_length": 245.07143783569336,
      "epoch": 0.9851209187308773,
      "grad_norm": 0.24078819208661117,
      "kl": 0.0866851806640625,
      "learning_rate": 4.997601402521077e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 5876
    },
    {
      "completion_length": 227.65179634094238,
      "epoch": 0.9854562219707448,
      "grad_norm": 0.3018082502223864,
      "kl": 0.0716400146484375,
      "learning_rate": 4.997596146165363e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571827709675,
      "rewards/format_reward_func": 1.0,
      "step": 5878
    },
    {
      "completion_length": 232.21429824829102,
      "epoch": 0.9857915252106123,
      "grad_norm": 0.20631451236813453,
      "kl": 0.0742034912109375,
      "learning_rate": 4.997590884059259e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 5880
    },
    {
      "completion_length": 245.1116180419922,
      "epoch": 0.9861268284504799,
      "grad_norm": 0.14566184117270678,
      "kl": 0.078857421875,
      "learning_rate": 4.997585616202773e-07,
      "loss": 0.0001,
      "reward": 1.714285783469677,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7142857499420643,
      "rewards/format_reward_func": 1.0,
      "step": 5882
    },
    {
      "completion_length": 245.8571538925171,
      "epoch": 0.9864621316903475,
      "grad_norm": 0.2400514185909936,
      "kl": 0.0743560791015625,
      "learning_rate": 4.99758034259592e-07,
      "loss": 0.0001,
      "reward": 1.7589286416769028,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7633928805589676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5884
    },
    {
      "completion_length": 241.3348331451416,
      "epoch": 0.986797434930215,
      "grad_norm": 0.10029168277200791,
      "kl": 0.0720062255859375,
      "learning_rate": 4.997575063238711e-07,
      "loss": 0.0001,
      "reward": 1.7214286252856255,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7214286047965288,
      "rewards/format_reward_func": 1.0,
      "step": 5886
    },
    {
      "completion_length": 239.9732255935669,
      "epoch": 0.9871327381700825,
      "grad_norm": 0.12171211237035984,
      "kl": 0.077362060546875,
      "learning_rate": 4.997569778131157e-07,
      "loss": 0.0001,
      "reward": 1.7839286178350449,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5888
    },
    {
      "completion_length": 251.11608600616455,
      "epoch": 0.9874680414099501,
      "grad_norm": 0.24615144405117734,
      "kl": 0.1072845458984375,
      "learning_rate": 4.997564487273272e-07,
      "loss": 0.0001,
      "reward": 1.7125000730156898,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7258929051458836,
      "rewards/format_reward_func": 0.9866071455180645,
      "step": 5890
    },
    {
      "completion_length": 248.27233409881592,
      "epoch": 0.9878033446498177,
      "grad_norm": 0.10878122748316436,
      "kl": 0.0915069580078125,
      "learning_rate": 4.997559190665067e-07,
      "loss": 0.0001,
      "reward": 1.7714286595582962,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285887777805,
      "rewards/format_reward_func": 1.0,
      "step": 5892
    },
    {
      "completion_length": 260.90626335144043,
      "epoch": 0.9881386478896852,
      "grad_norm": 0.24469718671854498,
      "kl": 0.10359954833984375,
      "learning_rate": 4.997553888306556e-07,
      "loss": 0.0001,
      "reward": 1.733928643167019,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7383928894996643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5894
    },
    {
      "completion_length": 263.89287090301514,
      "epoch": 0.9884739511295528,
      "grad_norm": 0.19164770454787758,
      "kl": 0.0981903076171875,
      "learning_rate": 4.997548580197749e-07,
      "loss": 0.0001,
      "reward": 1.7553572058677673,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7598214522004128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5896
    },
    {
      "completion_length": 262.14733028411865,
      "epoch": 0.9888092543694204,
      "grad_norm": 0.11700179725728932,
      "kl": 0.09405517578125,
      "learning_rate": 4.99754326633866e-07,
      "loss": 0.0001,
      "reward": 1.8035714700818062,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.8125000223517418,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5898
    },
    {
      "completion_length": 249.94197463989258,
      "epoch": 0.9891445576092879,
      "grad_norm": 0.1817638086218183,
      "kl": 0.14984130859375,
      "learning_rate": 4.997537946729298e-07,
      "loss": 0.0001,
      "reward": 1.7732143625617027,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7776785977184772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5900
    },
    {
      "completion_length": 266.526798248291,
      "epoch": 0.9894798608491554,
      "grad_norm": 0.3553538368445335,
      "kl": 0.1319122314453125,
      "learning_rate": 4.99753262136968e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.767857164144516,
      "rewards/format_reward_func": 1.0,
      "step": 5902
    },
    {
      "completion_length": 255.9375123977661,
      "epoch": 0.989815164089023,
      "grad_norm": 0.26831031710499303,
      "kl": 0.145782470703125,
      "learning_rate": 4.997527290259816e-07,
      "loss": 0.0001,
      "reward": 1.8071429058909416,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8160714507102966,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5904
    },
    {
      "completion_length": 268.3571548461914,
      "epoch": 0.9901504673288906,
      "grad_norm": 0.2864201561733845,
      "kl": 0.1101837158203125,
      "learning_rate": 4.997521953399717e-07,
      "loss": 0.0001,
      "reward": 1.766071505844593,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7794643118977547,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5906
    },
    {
      "completion_length": 266.1250114440918,
      "epoch": 0.9904857705687581,
      "grad_norm": 0.3488360998148485,
      "kl": 0.1395721435546875,
      "learning_rate": 4.997516610789397e-07,
      "loss": 0.0001,
      "reward": 1.773214340209961,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.7866071723401546,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5908
    },
    {
      "completion_length": 260.60715103149414,
      "epoch": 0.9908210738086257,
      "grad_norm": 0.13988328840005468,
      "kl": 0.1138763427734375,
      "learning_rate": 4.997511262428867e-07,
      "loss": 0.0001,
      "reward": 1.7500000596046448,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 5910
    },
    {
      "completion_length": 273.7991189956665,
      "epoch": 0.9911563770484932,
      "grad_norm": 0.18774606542998765,
      "kl": 0.15557861328125,
      "learning_rate": 4.997505908318142e-07,
      "loss": 0.0002,
      "reward": 1.7803571969270706,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7848214656114578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5912
    },
    {
      "completion_length": 261.88840770721436,
      "epoch": 0.9914916802883608,
      "grad_norm": 0.20507424623893872,
      "kl": 0.2316436767578125,
      "learning_rate": 4.997500548457231e-07,
      "loss": 0.0002,
      "reward": 1.7325893491506577,
      "reward_std": 0.044825518038123846,
      "rewards/equation_reward_func": 0.7383928745985031,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 5914
    },
    {
      "completion_length": 256.0669755935669,
      "epoch": 0.9918269835282283,
      "grad_norm": 0.3678386165226889,
      "kl": 0.1391754150390625,
      "learning_rate": 4.997495182846147e-07,
      "loss": 0.0001,
      "reward": 1.7348214909434319,
      "reward_std": 0.08207489270716906,
      "rewards/equation_reward_func": 0.741071455180645,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 5916
    },
    {
      "completion_length": 256.1160840988159,
      "epoch": 0.9921622867680959,
      "grad_norm": 0.1756867333928644,
      "kl": 0.1789093017578125,
      "learning_rate": 4.997489811484903e-07,
      "loss": 0.0002,
      "reward": 1.7803572043776512,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7848214581608772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5918
    },
    {
      "completion_length": 269.5178689956665,
      "epoch": 0.9924975900079634,
      "grad_norm": 0.11046519379219497,
      "kl": 0.1328277587890625,
      "learning_rate": 4.997484434373513e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.789285734295845,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5920
    },
    {
      "completion_length": 269.83929538726807,
      "epoch": 0.992832893247831,
      "grad_norm": 0.2843856797083342,
      "kl": 0.4182891845703125,
      "learning_rate": 4.997479051511988e-07,
      "loss": 0.0004,
      "reward": 1.696428656578064,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7053571809083223,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5922
    },
    {
      "completion_length": 266.794659614563,
      "epoch": 0.9931681964876986,
      "grad_norm": 0.25353859287113506,
      "kl": 0.2885894775390625,
      "learning_rate": 4.997473662900339e-07,
      "loss": 0.0003,
      "reward": 1.6517857983708382,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.6562500335276127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5924
    },
    {
      "completion_length": 260.80358505249023,
      "epoch": 0.9935034997275661,
      "grad_norm": 0.17313230378824523,
      "kl": 0.297637939453125,
      "learning_rate": 4.997468268538579e-07,
      "loss": 0.0003,
      "reward": 1.7571429312229156,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7660714574158192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5926
    },
    {
      "completion_length": 249.4107265472412,
      "epoch": 0.9938388029674337,
      "grad_norm": 0.1547458708021722,
      "kl": 0.3666229248046875,
      "learning_rate": 4.997462868426722e-07,
      "loss": 0.0004,
      "reward": 1.826785758137703,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8312500193715096,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5928
    },
    {
      "completion_length": 255.52233219146729,
      "epoch": 0.9941741062073012,
      "grad_norm": 0.39843122540179543,
      "kl": 0.4351806640625,
      "learning_rate": 4.997457462564781e-07,
      "loss": 0.0004,
      "reward": 1.7040179446339607,
      "reward_std": 0.06124049751088023,
      "rewards/equation_reward_func": 0.7098214589059353,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 5930
    },
    {
      "completion_length": 259.84376430511475,
      "epoch": 0.9945094094471688,
      "grad_norm": 0.3779794869916766,
      "kl": 0.6574249267578125,
      "learning_rate": 4.997452050952765e-07,
      "loss": 0.0007,
      "reward": 1.76071435213089,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7696428894996643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5932
    },
    {
      "completion_length": 251.8794765472412,
      "epoch": 0.9948447126870363,
      "grad_norm": 0.8563269925502356,
      "kl": 1.45794677734375,
      "learning_rate": 4.997446633590689e-07,
      "loss": 0.0015,
      "reward": 1.7553571909666061,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214708268642,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5934
    },
    {
      "completion_length": 251.55805110931396,
      "epoch": 0.9951800159269039,
      "grad_norm": 0.2936187191356089,
      "kl": 1.031280517578125,
      "learning_rate": 4.997441210478564e-07,
      "loss": 0.001,
      "reward": 1.7142857611179352,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7232143320143223,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5936
    },
    {
      "completion_length": 260.50447368621826,
      "epoch": 0.9955153191667715,
      "grad_norm": 0.28716438779501635,
      "kl": 0.818145751953125,
      "learning_rate": 4.997435781616405e-07,
      "loss": 0.0008,
      "reward": 1.7267857939004898,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7312500290572643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5938
    },
    {
      "completion_length": 256.3616180419922,
      "epoch": 0.995850622406639,
      "grad_norm": 0.23298759856279558,
      "kl": 1.7165985107421875,
      "learning_rate": 4.997430347004221e-07,
      "loss": 0.0017,
      "reward": 1.7642857655882835,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7732143178582191,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5940
    },
    {
      "completion_length": 256.57143783569336,
      "epoch": 0.9961859256465065,
      "grad_norm": 0.2095096831056827,
      "kl": 0.1193695068359375,
      "learning_rate": 4.997424906642028e-07,
      "loss": 0.0001,
      "reward": 1.8200893253087997,
      "reward_std": 0.06250318605452776,
      "rewards/equation_reward_func": 0.8214285895228386,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 5942
    },
    {
      "completion_length": 253.92858219146729,
      "epoch": 0.9965212288863741,
      "grad_norm": 0.22191173301560937,
      "kl": 0.5313568115234375,
      "learning_rate": 4.997419460529836e-07,
      "loss": 0.0005,
      "reward": 1.7160715237259865,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7205357477068901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5944
    },
    {
      "completion_length": 252.2946548461914,
      "epoch": 0.9968565321262417,
      "grad_norm": 0.2167018970654993,
      "kl": 1.0382080078125,
      "learning_rate": 4.997414008667658e-07,
      "loss": 0.001,
      "reward": 1.7982143461704254,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8026785887777805,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5946
    },
    {
      "completion_length": 262.90626525878906,
      "epoch": 0.9971918353661092,
      "grad_norm": 0.25514814703256206,
      "kl": 0.0827789306640625,
      "learning_rate": 4.997408551055508e-07,
      "loss": 0.0001,
      "reward": 1.7839286476373672,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7875000238418579,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 5948
    },
    {
      "completion_length": 261.5714406967163,
      "epoch": 0.9975271386059767,
      "grad_norm": 0.30817589784765403,
      "kl": 0.498565673828125,
      "learning_rate": 4.997403087693398e-07,
      "loss": 0.0005,
      "reward": 1.7821429297327995,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 5950
    },
    {
      "completion_length": 264.19197845458984,
      "epoch": 0.9978624418458444,
      "grad_norm": 0.21263202583302027,
      "kl": 0.1584625244140625,
      "learning_rate": 4.997397618581339e-07,
      "loss": 0.0002,
      "reward": 1.7767857611179352,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7812500204890966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5952
    },
    {
      "completion_length": 259.008939743042,
      "epoch": 0.9981977450857119,
      "grad_norm": 0.2297520903477404,
      "kl": 0.1841583251953125,
      "learning_rate": 4.997392143719344e-07,
      "loss": 0.0002,
      "reward": 1.7107143625617027,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7107143327593803,
      "rewards/format_reward_func": 1.0,
      "step": 5954
    },
    {
      "completion_length": 255.6785831451416,
      "epoch": 0.9985330483255794,
      "grad_norm": 0.23026073378599074,
      "kl": 0.18896484375,
      "learning_rate": 4.997386663107428e-07,
      "loss": 0.0002,
      "reward": 1.7428572177886963,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571823984385,
      "rewards/format_reward_func": 1.0,
      "step": 5956
    },
    {
      "completion_length": 266.88394260406494,
      "epoch": 0.998868351565447,
      "grad_norm": 0.2934772523750861,
      "kl": 0.1302490234375,
      "learning_rate": 4.997381176745602e-07,
      "loss": 0.0001,
      "reward": 1.7839286178350449,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5958
    },
    {
      "completion_length": 262.446439743042,
      "epoch": 0.9992036548053146,
      "grad_norm": 0.26250342499542473,
      "kl": 0.257049560546875,
      "learning_rate": 4.997375684633878e-07,
      "loss": 0.0003,
      "reward": 1.782142922282219,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7821428682655096,
      "rewards/format_reward_func": 1.0,
      "step": 5960
    },
    {
      "completion_length": 253.3616180419922,
      "epoch": 0.9995389580451821,
      "grad_norm": 0.35762220838323594,
      "kl": 0.0915374755859375,
      "learning_rate": 4.997370186772269e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5962
    },
    {
      "completion_length": 263.8259029388428,
      "epoch": 0.9998742612850496,
      "grad_norm": 1.4570787798699594,
      "kl": 0.1656646728515625,
      "learning_rate": 4.997364683160787e-07,
      "loss": 0.0002,
      "reward": 1.7321429327130318,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428880095482,
      "rewards/format_reward_func": 1.0,
      "step": 5964
    },
    {
      "completion_length": 265.9805304787376,
      "epoch": 1.0003353032398676,
      "grad_norm": 0.9384003688196095,
      "kl": 0.10651189630681818,
      "learning_rate": 4.997359173799447e-07,
      "loss": 0.0001,
      "reward": 1.6987013925205579,
      "reward_std": 0.08815876665440472,
      "rewards/equation_reward_func": 0.7116883397102356,
      "rewards/format_reward_func": 0.9870129932056774,
      "step": 5966
    },
    {
      "completion_length": 259.83037090301514,
      "epoch": 1.000670606479735,
      "grad_norm": 1.2250441001337247,
      "kl": 0.1012420654296875,
      "learning_rate": 4.99735365868826e-07,
      "loss": 0.0001,
      "reward": 1.769642949104309,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7741071656346321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5968
    },
    {
      "completion_length": 265.8125104904175,
      "epoch": 1.0010059097196027,
      "grad_norm": 0.21174169445063962,
      "kl": 0.10089111328125,
      "learning_rate": 4.997348137827238e-07,
      "loss": 0.0001,
      "reward": 1.714285783469677,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7142857573926449,
      "rewards/format_reward_func": 1.0,
      "step": 5970
    },
    {
      "completion_length": 264.96876430511475,
      "epoch": 1.0013412129594703,
      "grad_norm": 0.309308577606721,
      "kl": 0.06874847412109375,
      "learning_rate": 4.997342611216395e-07,
      "loss": 0.0001,
      "reward": 1.7767857760190964,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7812500335276127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5972
    },
    {
      "completion_length": 258.415189743042,
      "epoch": 1.0016765161993377,
      "grad_norm": 0.9905876695648793,
      "kl": 0.0795135498046875,
      "learning_rate": 4.997337078855744e-07,
      "loss": 0.0001,
      "reward": 1.792857214808464,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 5974
    },
    {
      "completion_length": 266.07144355773926,
      "epoch": 1.0020118194392054,
      "grad_norm": 0.2839674817964923,
      "kl": 0.1131744384765625,
      "learning_rate": 4.997331540745296e-07,
      "loss": 0.0001,
      "reward": 1.7303571924567223,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7437500394880772,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5976
    },
    {
      "completion_length": 276.91519355773926,
      "epoch": 1.0023471226790728,
      "grad_norm": 0.24495313830148246,
      "kl": 0.123016357421875,
      "learning_rate": 4.997325996885066e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 5978
    },
    {
      "completion_length": 273.0178699493408,
      "epoch": 1.0026824259189404,
      "grad_norm": 1.3292883047967456,
      "kl": 0.136383056640625,
      "learning_rate": 4.997320447275065e-07,
      "loss": 0.0001,
      "reward": 1.71830365806818,
      "reward_std": 0.05492704384960234,
      "rewards/equation_reward_func": 0.7241071686148643,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 5980
    },
    {
      "completion_length": 271.33929538726807,
      "epoch": 1.003017729158808,
      "grad_norm": 1.1772382051916397,
      "kl": 0.2193603515625,
      "learning_rate": 4.997314891915307e-07,
      "loss": 0.0002,
      "reward": 1.7607143595814705,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7785714454948902,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5982
    },
    {
      "completion_length": 260.65179347991943,
      "epoch": 1.0033530323986755,
      "grad_norm": 4.171729150636543,
      "kl": 0.216400146484375,
      "learning_rate": 4.997309330805803e-07,
      "loss": 0.0002,
      "reward": 1.7625000551342964,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643200933933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 5984
    },
    {
      "completion_length": 264.4196529388428,
      "epoch": 1.003688335638543,
      "grad_norm": 0.38616155970757654,
      "kl": 0.297698974609375,
      "learning_rate": 4.997303763946568e-07,
      "loss": 0.0003,
      "reward": 1.7160715013742447,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.7294643148779869,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 5986
    },
    {
      "completion_length": 261.3169755935669,
      "epoch": 1.0040236388784107,
      "grad_norm": 0.21821546768331762,
      "kl": 0.239715576171875,
      "learning_rate": 4.997298191337613e-07,
      "loss": 0.0002,
      "reward": 1.7611608058214188,
      "reward_std": 0.05492704384960234,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 5988
    },
    {
      "completion_length": 272.94197845458984,
      "epoch": 1.0043589421182781,
      "grad_norm": 0.8574313762374961,
      "kl": 0.2301483154296875,
      "learning_rate": 4.997292612978954e-07,
      "loss": 0.0002,
      "reward": 1.7214286103844643,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7303571626543999,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5990
    },
    {
      "completion_length": 282.87947273254395,
      "epoch": 1.0046942453581458,
      "grad_norm": 0.41062082585236326,
      "kl": 0.18109130859375,
      "learning_rate": 4.997287028870599e-07,
      "loss": 0.0002,
      "reward": 1.739285796880722,
      "reward_std": 0.09596449136734009,
      "rewards/equation_reward_func": 0.7571428753435612,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5992
    },
    {
      "completion_length": 264.7321548461914,
      "epoch": 1.0050295485980134,
      "grad_norm": 0.22685254042735117,
      "kl": 0.5179901123046875,
      "learning_rate": 4.997281439012564e-07,
      "loss": 0.0005,
      "reward": 1.8035714849829674,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.8125000335276127,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 5994
    },
    {
      "completion_length": 269.5223331451416,
      "epoch": 1.0053648518378808,
      "grad_norm": 0.22963070499995214,
      "kl": 0.550018310546875,
      "learning_rate": 4.997275843404861e-07,
      "loss": 0.0006,
      "reward": 1.7142857760190964,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7321428991854191,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 5996
    },
    {
      "completion_length": 265.7946557998657,
      "epoch": 1.0057001550777485,
      "grad_norm": 0.4467582667821596,
      "kl": 0.439666748046875,
      "learning_rate": 4.997270242047504e-07,
      "loss": 0.0004,
      "reward": 1.7343750521540642,
      "reward_std": 0.0675539500080049,
      "rewards/equation_reward_func": 0.7535714637488127,
      "rewards/format_reward_func": 0.980803582817316,
      "step": 5998
    },
    {
      "completion_length": 262.388409614563,
      "epoch": 1.006035458317616,
      "grad_norm": 0.21615161312001333,
      "kl": 0.2450103759765625,
      "learning_rate": 4.997264634940503e-07,
      "loss": 0.0002,
      "reward": 1.773214340209961,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.777678620070219,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6000
    },
    {
      "completion_length": 250.29911994934082,
      "epoch": 1.0063707615574835,
      "grad_norm": 0.34505426050760507,
      "kl": 1.3549957275390625,
      "learning_rate": 4.997259022083875e-07,
      "loss": 0.0014,
      "reward": 1.76071435213089,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 6002
    },
    {
      "completion_length": 256.5000104904175,
      "epoch": 1.0067060647973511,
      "grad_norm": 0.23308189879398827,
      "kl": 0.4465484619140625,
      "learning_rate": 4.99725340347763e-07,
      "loss": 0.0004,
      "reward": 1.7714286372065544,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 6004
    },
    {
      "completion_length": 250.5669765472412,
      "epoch": 1.0070413680372186,
      "grad_norm": 0.23353505975057562,
      "kl": 0.1022491455078125,
      "learning_rate": 4.997247779121782e-07,
      "loss": 0.0001,
      "reward": 1.8357143178582191,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.8446428813040257,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6006
    },
    {
      "completion_length": 260.1562614440918,
      "epoch": 1.0073766712770862,
      "grad_norm": 0.34613575609921376,
      "kl": 0.1776885986328125,
      "learning_rate": 4.997242149016343e-07,
      "loss": 0.0002,
      "reward": 1.7477679252624512,
      "reward_std": 0.08396892924793065,
      "rewards/equation_reward_func": 0.7553571686148643,
      "rewards/format_reward_func": 0.9924107231199741,
      "step": 6008
    },
    {
      "completion_length": 260.95983505249023,
      "epoch": 1.0077119745169538,
      "grad_norm": 0.209932605339094,
      "kl": 0.2499847412109375,
      "learning_rate": 4.997236513161327e-07,
      "loss": 0.0003,
      "reward": 1.741071492433548,
      "reward_std": 0.08333758544176817,
      "rewards/equation_reward_func": 0.7633928693830967,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 6010
    },
    {
      "completion_length": 268.901798248291,
      "epoch": 1.0080472777568212,
      "grad_norm": 0.23702142547618296,
      "kl": 0.272735595703125,
      "learning_rate": 4.997230871556747e-07,
      "loss": 0.0003,
      "reward": 1.751785770058632,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500387430191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6012
    },
    {
      "completion_length": 258.6875123977661,
      "epoch": 1.0083825809966889,
      "grad_norm": 0.2647906060959816,
      "kl": 0.30157470703125,
      "learning_rate": 4.997225224202616e-07,
      "loss": 0.0003,
      "reward": 1.7517857924103737,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500294297934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6014
    },
    {
      "completion_length": 260.34822368621826,
      "epoch": 1.0087178842365565,
      "grad_norm": 0.3469783106868168,
      "kl": 0.2371063232421875,
      "learning_rate": 4.997219571098945e-07,
      "loss": 0.0002,
      "reward": 1.7589286267757416,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7633928842842579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6016
    },
    {
      "completion_length": 251.0714406967163,
      "epoch": 1.009053187476424,
      "grad_norm": 0.2878495464486915,
      "kl": 0.1557464599609375,
      "learning_rate": 4.997213912245751e-07,
      "loss": 0.0002,
      "reward": 1.7678572461009026,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 6018
    },
    {
      "completion_length": 253.09376049041748,
      "epoch": 1.0093884907162916,
      "grad_norm": 0.31678984513529884,
      "kl": 0.5877227783203125,
      "learning_rate": 4.997208247643042e-07,
      "loss": 0.0006,
      "reward": 1.7678571864962578,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571790456772,
      "rewards/format_reward_func": 1.0,
      "step": 6020
    },
    {
      "completion_length": 251.00001335144043,
      "epoch": 1.0097237939561592,
      "grad_norm": 0.2262559946347966,
      "kl": 0.2086334228515625,
      "learning_rate": 4.997202577290836e-07,
      "loss": 0.0002,
      "reward": 1.7607143595814705,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7696428932249546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6022
    },
    {
      "completion_length": 259.9598340988159,
      "epoch": 1.0100590971960266,
      "grad_norm": 0.3477392927745284,
      "kl": 0.3122711181640625,
      "learning_rate": 4.997196901189142e-07,
      "loss": 0.0003,
      "reward": 1.7696429342031479,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7741071730852127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6024
    },
    {
      "completion_length": 251.7098331451416,
      "epoch": 1.0103944004358942,
      "grad_norm": 0.20619517269581278,
      "kl": 0.207794189453125,
      "learning_rate": 4.997191219337976e-07,
      "loss": 0.0002,
      "reward": 1.7339286506175995,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7383928876370192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6026
    },
    {
      "completion_length": 267.39733600616455,
      "epoch": 1.0107297036757619,
      "grad_norm": 0.28115504056281226,
      "kl": 0.31494140625,
      "learning_rate": 4.99718553173735e-07,
      "loss": 0.0003,
      "reward": 1.7214286401867867,
      "reward_std": 0.0909137288108468,
      "rewards/equation_reward_func": 0.739285746589303,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6028
    },
    {
      "completion_length": 250.46876049041748,
      "epoch": 1.0110650069156293,
      "grad_norm": 0.08975104121727806,
      "kl": 0.3542633056640625,
      "learning_rate": 4.997179838387276e-07,
      "loss": 0.0004,
      "reward": 1.7642857730388641,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 6030
    },
    {
      "completion_length": 249.0714406967163,
      "epoch": 1.011400310155497,
      "grad_norm": 0.2578463046422513,
      "kl": 0.090484619140625,
      "learning_rate": 4.997174139287768e-07,
      "loss": 0.0001,
      "reward": 1.7964286357164383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.796428594738245,
      "rewards/format_reward_func": 1.0,
      "step": 6032
    },
    {
      "completion_length": 248.33036994934082,
      "epoch": 1.0117356133953643,
      "grad_norm": 0.30858155230260276,
      "kl": 0.078338623046875,
      "learning_rate": 4.997168434438841e-07,
      "loss": 0.0001,
      "reward": 1.7321429401636124,
      "reward_std": 0.0656599160283804,
      "rewards/equation_reward_func": 0.7410714663565159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6034
    },
    {
      "completion_length": 252.12054538726807,
      "epoch": 1.012070916635232,
      "grad_norm": 0.26137515161726865,
      "kl": 0.0858306884765625,
      "learning_rate": 4.997162723840505e-07,
      "loss": 0.0001,
      "reward": 1.7910714596509933,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7955357432365417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6036
    },
    {
      "completion_length": 246.0669765472412,
      "epoch": 1.0124062198750996,
      "grad_norm": 0.30419280294145973,
      "kl": 0.07515716552734375,
      "learning_rate": 4.997157007492775e-07,
      "loss": 0.0001,
      "reward": 1.7660714909434319,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7705357372760773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6038
    },
    {
      "completion_length": 250.9241189956665,
      "epoch": 1.012741523114967,
      "grad_norm": 0.2789904085399853,
      "kl": 0.1070098876953125,
      "learning_rate": 4.997151285395663e-07,
      "loss": 0.0001,
      "reward": 1.739285796880722,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857521772385,
      "rewards/format_reward_func": 1.0,
      "step": 6040
    },
    {
      "completion_length": 257.3526906967163,
      "epoch": 1.0130768263548346,
      "grad_norm": 0.29540488870553583,
      "kl": 0.211273193359375,
      "learning_rate": 4.997145557549184e-07,
      "loss": 0.0002,
      "reward": 1.6821429580450058,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.6821429021656513,
      "rewards/format_reward_func": 1.0,
      "step": 6042
    },
    {
      "completion_length": 263.98661613464355,
      "epoch": 1.0134121295947023,
      "grad_norm": 0.3503033270367516,
      "kl": 0.429779052734375,
      "learning_rate": 4.99713982395335e-07,
      "loss": 0.0004,
      "reward": 1.7642857506871223,
      "reward_std": 0.04040610231459141,
      "rewards/equation_reward_func": 0.77321432903409,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6044
    },
    {
      "completion_length": 251.9509048461914,
      "epoch": 1.0137474328345697,
      "grad_norm": 0.2594043398085691,
      "kl": 0.1715240478515625,
      "learning_rate": 4.997134084608174e-07,
      "loss": 0.0002,
      "reward": 1.85714291036129,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8571428693830967,
      "rewards/format_reward_func": 1.0,
      "step": 6046
    },
    {
      "completion_length": 255.3482265472412,
      "epoch": 1.0140827360744373,
      "grad_norm": 0.3322364759657405,
      "kl": 0.13507080078125,
      "learning_rate": 4.997128339513669e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 6048
    },
    {
      "completion_length": 253.30805110931396,
      "epoch": 1.014418039314305,
      "grad_norm": 0.5803981729663658,
      "kl": 0.480499267578125,
      "learning_rate": 4.997122588669849e-07,
      "loss": 0.0005,
      "reward": 1.7535715028643608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714693367481,
      "rewards/format_reward_func": 1.0,
      "step": 6050
    },
    {
      "completion_length": 252.18751049041748,
      "epoch": 1.0147533425541724,
      "grad_norm": 0.40183967799838594,
      "kl": 0.190338134765625,
      "learning_rate": 4.997116832076727e-07,
      "loss": 0.0002,
      "reward": 1.7571429535746574,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 6052
    },
    {
      "completion_length": 246.7321548461914,
      "epoch": 1.01508864579404,
      "grad_norm": 0.26403663378360054,
      "kl": 0.138153076171875,
      "learning_rate": 4.997111069734316e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.8053571693599224,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6054
    },
    {
      "completion_length": 250.02679920196533,
      "epoch": 1.0154239490339076,
      "grad_norm": 0.18947696729731334,
      "kl": 0.109039306640625,
      "learning_rate": 4.997105301642629e-07,
      "loss": 0.0001,
      "reward": 1.7660714983940125,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7705357372760773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6056
    },
    {
      "completion_length": 257.2767972946167,
      "epoch": 1.015759252273775,
      "grad_norm": 0.43595958723248723,
      "kl": 0.13599395751953125,
      "learning_rate": 4.99709952780168e-07,
      "loss": 0.0001,
      "reward": 1.7392857745289803,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857447266579,
      "rewards/format_reward_func": 1.0,
      "step": 6058
    },
    {
      "completion_length": 258.1741189956665,
      "epoch": 1.0160945555136427,
      "grad_norm": 0.261448358162978,
      "kl": 0.150146484375,
      "learning_rate": 4.997093748211482e-07,
      "loss": 0.0002,
      "reward": 1.73214291036129,
      "reward_std": 0.04545686487108469,
      "rewards/equation_reward_func": 0.7410714700818062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6060
    },
    {
      "completion_length": 256.78126430511475,
      "epoch": 1.01642985875351,
      "grad_norm": 0.18200663054584304,
      "kl": 0.103240966796875,
      "learning_rate": 4.997087962872049e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 6062
    },
    {
      "completion_length": 258.3571557998657,
      "epoch": 1.0167651619933777,
      "grad_norm": 0.1819383632703182,
      "kl": 0.17974853515625,
      "learning_rate": 4.997082171783393e-07,
      "loss": 0.0002,
      "reward": 1.7732143625617027,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.777678593993187,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6064
    },
    {
      "completion_length": 259.5134057998657,
      "epoch": 1.0171004652332454,
      "grad_norm": 0.3211241713421729,
      "kl": 0.13037109375,
      "learning_rate": 4.997076374945529e-07,
      "loss": 0.0001,
      "reward": 1.7500000521540642,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6066
    },
    {
      "completion_length": 255.58036994934082,
      "epoch": 1.0174357684731128,
      "grad_norm": 0.2425993917665553,
      "kl": 0.2032470703125,
      "learning_rate": 4.997070572358468e-07,
      "loss": 0.0002,
      "reward": 1.814285770058632,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8142857439815998,
      "rewards/format_reward_func": 1.0,
      "step": 6068
    },
    {
      "completion_length": 264.77680110931396,
      "epoch": 1.0177710717129804,
      "grad_norm": 0.15083247763046945,
      "kl": 0.17120361328125,
      "learning_rate": 4.997064764022225e-07,
      "loss": 0.0002,
      "reward": 1.780357226729393,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6070
    },
    {
      "completion_length": 259.620548248291,
      "epoch": 1.018106374952848,
      "grad_norm": 0.5226944898181175,
      "kl": 0.14019775390625,
      "learning_rate": 4.997058949936814e-07,
      "loss": 0.0001,
      "reward": 1.7482143342494965,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.7616071663796902,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6072
    },
    {
      "completion_length": 267.4955520629883,
      "epoch": 1.0184416781927155,
      "grad_norm": 0.29505621677078775,
      "kl": 0.1959228515625,
      "learning_rate": 4.997053130102247e-07,
      "loss": 0.0002,
      "reward": 1.7071429267525673,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7160714529454708,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6074
    },
    {
      "completion_length": 270.7053699493408,
      "epoch": 1.018776981432583,
      "grad_norm": 0.5437437458100716,
      "kl": 0.2197265625,
      "learning_rate": 4.997047304518538e-07,
      "loss": 0.0002,
      "reward": 1.7482143566012383,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7526785954833031,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6076
    },
    {
      "completion_length": 268.59822368621826,
      "epoch": 1.0191122846724507,
      "grad_norm": 0.24741324747983043,
      "kl": 0.14093017578125,
      "learning_rate": 4.997041473185699e-07,
      "loss": 0.0001,
      "reward": 1.6857143640518188,
      "reward_std": 0.09091372694820166,
      "rewards/equation_reward_func": 0.6946428865194321,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6078
    },
    {
      "completion_length": 279.3928680419922,
      "epoch": 1.0194475879123182,
      "grad_norm": 0.16270285483118307,
      "kl": 0.12310791015625,
      "learning_rate": 4.997035636103746e-07,
      "loss": 0.0001,
      "reward": 1.7808035910129547,
      "reward_std": 0.0776554774492979,
      "rewards/equation_reward_func": 0.800000037997961,
      "rewards/format_reward_func": 0.980803582817316,
      "step": 6080
    },
    {
      "completion_length": 270.41518783569336,
      "epoch": 1.0197828911521858,
      "grad_norm": 0.35956905028353797,
      "kl": 0.286376953125,
      "learning_rate": 4.997029793272691e-07,
      "loss": 0.0003,
      "reward": 1.703571505844593,
      "reward_std": 0.08586296625435352,
      "rewards/equation_reward_func": 0.7125000394880772,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6082
    },
    {
      "completion_length": 270.30358505249023,
      "epoch": 1.0201181943920532,
      "grad_norm": 0.47292226180050445,
      "kl": 0.266021728515625,
      "learning_rate": 4.997023944692547e-07,
      "loss": 0.0003,
      "reward": 1.7482143491506577,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7616071775555611,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6084
    },
    {
      "completion_length": 251.9107265472412,
      "epoch": 1.0204534976319208,
      "grad_norm": 0.2832647131892392,
      "kl": 0.1192626953125,
      "learning_rate": 4.997018090363329e-07,
      "loss": 0.0001,
      "reward": 1.7732143327593803,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7866071555763483,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6086
    },
    {
      "completion_length": 264.56697845458984,
      "epoch": 1.0207888008717885,
      "grad_norm": 0.31822589986037786,
      "kl": 0.26104736328125,
      "learning_rate": 4.997012230285048e-07,
      "loss": 0.0003,
      "reward": 1.6875000894069672,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.7098214626312256,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 6088
    },
    {
      "completion_length": 265.49108695983887,
      "epoch": 1.0211241041116559,
      "grad_norm": 0.45547720897693517,
      "kl": 0.40484619140625,
      "learning_rate": 4.997006364457721e-07,
      "loss": 0.0004,
      "reward": 1.7750000655651093,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 6090
    },
    {
      "completion_length": 268.34376430511475,
      "epoch": 1.0214594073515235,
      "grad_norm": 0.2786369185101486,
      "kl": 0.428558349609375,
      "learning_rate": 4.997000492881358e-07,
      "loss": 0.0004,
      "reward": 1.7053572237491608,
      "reward_std": 0.09848987311124802,
      "rewards/equation_reward_func": 0.7366071715950966,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 6092
    },
    {
      "completion_length": 277.3214406967163,
      "epoch": 1.0217947105913912,
      "grad_norm": 0.18411353534994465,
      "kl": 0.46099853515625,
      "learning_rate": 4.996994615555975e-07,
      "loss": 0.0005,
      "reward": 1.7071429193019867,
      "reward_std": 0.10101525392383337,
      "rewards/equation_reward_func": 0.7250000387430191,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6094
    },
    {
      "completion_length": 267.19197845458984,
      "epoch": 1.0221300138312586,
      "grad_norm": 0.3466971216842296,
      "kl": 0.677581787109375,
      "learning_rate": 4.996988732481584e-07,
      "loss": 0.0007,
      "reward": 1.694642923772335,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.716964315623045,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 6096
    },
    {
      "completion_length": 267.0223379135132,
      "epoch": 1.0224653170711262,
      "grad_norm": 0.3621796109181974,
      "kl": 0.21337890625,
      "learning_rate": 4.9969828436582e-07,
      "loss": 0.0002,
      "reward": 1.7410714849829674,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7544643245637417,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6098
    },
    {
      "completion_length": 273.089298248291,
      "epoch": 1.0228006203109938,
      "grad_norm": 0.7042681314100642,
      "kl": 0.7377166748046875,
      "learning_rate": 4.996976949085834e-07,
      "loss": 0.0007,
      "reward": 1.6803572326898575,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.6937500275671482,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6100
    },
    {
      "completion_length": 267.9687614440918,
      "epoch": 1.0231359235508612,
      "grad_norm": 0.2904338249092064,
      "kl": 0.147735595703125,
      "learning_rate": 4.996971048764502e-07,
      "loss": 0.0001,
      "reward": 1.7142857909202576,
      "reward_std": 0.1060660183429718,
      "rewards/equation_reward_func": 0.7410714644938707,
      "rewards/format_reward_func": 0.9732142947614193,
      "step": 6102
    },
    {
      "completion_length": 254.40179443359375,
      "epoch": 1.0234712267907289,
      "grad_norm": 0.1715708017841068,
      "kl": 0.1373291015625,
      "learning_rate": 4.996965142694217e-07,
      "loss": 0.0001,
      "reward": 1.8035714849829674,
      "reward_std": 0.025253813713788986,
      "rewards/equation_reward_func": 0.8125000223517418,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6104
    },
    {
      "completion_length": 260.66518783569336,
      "epoch": 1.0238065300305965,
      "grad_norm": 0.5298721752119621,
      "kl": 0.3925323486328125,
      "learning_rate": 4.996959230874993e-07,
      "loss": 0.0004,
      "reward": 1.7607143446803093,
      "reward_std": 0.09596449043601751,
      "rewards/equation_reward_func": 0.769642885774374,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6106
    },
    {
      "completion_length": 261.8437614440918,
      "epoch": 1.024141833270464,
      "grad_norm": 0.2597657463578196,
      "kl": 0.9878387451171875,
      "learning_rate": 4.996953313306842e-07,
      "loss": 0.001,
      "reward": 1.7339286357164383,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7473214529454708,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6108
    },
    {
      "completion_length": 258.2544765472412,
      "epoch": 1.0244771365103316,
      "grad_norm": 0.34126144567118943,
      "kl": 0.373443603515625,
      "learning_rate": 4.99694738998978e-07,
      "loss": 0.0004,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 6110
    },
    {
      "completion_length": 248.68304443359375,
      "epoch": 1.024812439750199,
      "grad_norm": 0.191507371496448,
      "kl": 0.4516754150390625,
      "learning_rate": 4.996941460923818e-07,
      "loss": 0.0005,
      "reward": 1.7660714760422707,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7794643137603998,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6112
    },
    {
      "completion_length": 252.77233409881592,
      "epoch": 1.0251477429900666,
      "grad_norm": 0.38964610451237514,
      "kl": 0.4473114013671875,
      "learning_rate": 4.996935526108972e-07,
      "loss": 0.0004,
      "reward": 1.7482143267989159,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7526786085218191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6114
    },
    {
      "completion_length": 251.73215293884277,
      "epoch": 1.0254830462299342,
      "grad_norm": 0.2627876383548723,
      "kl": 0.3182525634765625,
      "learning_rate": 4.996929585545254e-07,
      "loss": 0.0003,
      "reward": 1.7696429193019867,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6116
    },
    {
      "completion_length": 239.15625953674316,
      "epoch": 1.0258183494698017,
      "grad_norm": 0.11734356018251145,
      "kl": 0.720947265625,
      "learning_rate": 4.996923639232678e-07,
      "loss": 0.0007,
      "reward": 1.7642857730388641,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 6118
    },
    {
      "completion_length": 232.41518878936768,
      "epoch": 1.0261536527096693,
      "grad_norm": 0.3020438778673639,
      "kl": 0.3007354736328125,
      "learning_rate": 4.996917687171258e-07,
      "loss": 0.0003,
      "reward": 1.7839286476373672,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6120
    },
    {
      "completion_length": 234.4821548461914,
      "epoch": 1.026488955949537,
      "grad_norm": 0.19764513890719873,
      "kl": 0.581573486328125,
      "learning_rate": 4.996911729361009e-07,
      "loss": 0.0006,
      "reward": 1.8357143551111221,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8357143141329288,
      "rewards/format_reward_func": 1.0,
      "step": 6122
    },
    {
      "completion_length": 230.95983219146729,
      "epoch": 1.0268242591894043,
      "grad_norm": 0.1907246795303604,
      "kl": 0.1136474609375,
      "learning_rate": 4.996905765801942e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 6124
    },
    {
      "completion_length": 233.60715198516846,
      "epoch": 1.027159562429272,
      "grad_norm": 0.22052651145584928,
      "kl": 0.1941375732421875,
      "learning_rate": 4.996899796494073e-07,
      "loss": 0.0002,
      "reward": 1.7928571999073029,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 6126
    },
    {
      "completion_length": 233.07590198516846,
      "epoch": 1.0274948656691396,
      "grad_norm": 0.28840565132762075,
      "kl": 0.187591552734375,
      "learning_rate": 4.996893821437414e-07,
      "loss": 0.0002,
      "reward": 1.6821429654955864,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6821428909897804,
      "rewards/format_reward_func": 1.0,
      "step": 6128
    },
    {
      "completion_length": 222.7991189956665,
      "epoch": 1.027830168909007,
      "grad_norm": 0.22193940299452425,
      "kl": 0.144378662109375,
      "learning_rate": 4.99688784063198e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428999304771,
      "rewards/format_reward_func": 1.0,
      "step": 6130
    },
    {
      "completion_length": 230.36608219146729,
      "epoch": 1.0281654721488747,
      "grad_norm": 0.5075373073783743,
      "kl": 0.3329925537109375,
      "learning_rate": 4.996881854077785e-07,
      "loss": 0.0003,
      "reward": 1.7571429088711739,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.757142897695303,
      "rewards/format_reward_func": 1.0,
      "step": 6132
    },
    {
      "completion_length": 231.11608219146729,
      "epoch": 1.0285007753887423,
      "grad_norm": 0.20879696471969295,
      "kl": 0.253082275390625,
      "learning_rate": 4.996875861774842e-07,
      "loss": 0.0003,
      "reward": 1.7517857924103737,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500312924385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6134
    },
    {
      "completion_length": 230.20090579986572,
      "epoch": 1.0288360786286097,
      "grad_norm": 0.38774526912973895,
      "kl": 0.451629638671875,
      "learning_rate": 4.996869863723165e-07,
      "loss": 0.0005,
      "reward": 1.7267857789993286,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7401785980910063,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6136
    },
    {
      "completion_length": 220.5357265472412,
      "epoch": 1.0291713818684773,
      "grad_norm": 0.27093543124287117,
      "kl": 0.1378173828125,
      "learning_rate": 4.996863859922767e-07,
      "loss": 0.0001,
      "reward": 1.7464286610484123,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.746428593993187,
      "rewards/format_reward_func": 1.0,
      "step": 6138
    },
    {
      "completion_length": 231.71876049041748,
      "epoch": 1.0295066851083448,
      "grad_norm": 0.6042941302054783,
      "kl": 0.576324462890625,
      "learning_rate": 4.996857850373663e-07,
      "loss": 0.0006,
      "reward": 1.7147322073578835,
      "reward_std": 0.04987628059461713,
      "rewards/equation_reward_func": 0.7205357514321804,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 6140
    },
    {
      "completion_length": 229.93751049041748,
      "epoch": 1.0298419883482124,
      "grad_norm": 0.40842733005443116,
      "kl": 0.823272705078125,
      "learning_rate": 4.996851835075867e-07,
      "loss": 0.0008,
      "reward": 1.6964286342263222,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7142857573926449,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6142
    },
    {
      "completion_length": 221.98215198516846,
      "epoch": 1.03017729158808,
      "grad_norm": 0.24154587050830953,
      "kl": 0.40399169921875,
      "learning_rate": 4.996845814029392e-07,
      "loss": 0.0004,
      "reward": 1.7535714879631996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 6144
    },
    {
      "completion_length": 218.56697750091553,
      "epoch": 1.0305125948279474,
      "grad_norm": 0.2808529909155721,
      "kl": 1.314208984375,
      "learning_rate": 4.996839787234252e-07,
      "loss": 0.0013,
      "reward": 1.7839286476373672,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6146
    },
    {
      "completion_length": 231.55804824829102,
      "epoch": 1.030847898067815,
      "grad_norm": 0.20719697823983504,
      "kl": 0.54595947265625,
      "learning_rate": 4.996833754690461e-07,
      "loss": 0.0005,
      "reward": 1.7214286625385284,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7303571794182062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6148
    },
    {
      "completion_length": 228.35715293884277,
      "epoch": 1.0311832013076827,
      "grad_norm": 0.32793316359403357,
      "kl": 0.9630889892578125,
      "learning_rate": 4.996827716398033e-07,
      "loss": 0.001,
      "reward": 1.7482143566012383,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6150
    },
    {
      "completion_length": 223.99554538726807,
      "epoch": 1.0315185045475501,
      "grad_norm": 0.13363658578744522,
      "kl": 0.293060302734375,
      "learning_rate": 4.996821672356982e-07,
      "loss": 0.0003,
      "reward": 1.7303571999073029,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7348214760422707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6152
    },
    {
      "completion_length": 219.50000858306885,
      "epoch": 1.0318538077874178,
      "grad_norm": 0.233703696734875,
      "kl": 0.246307373046875,
      "learning_rate": 4.996815622567322e-07,
      "loss": 0.0002,
      "reward": 1.7785715013742447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 6154
    },
    {
      "completion_length": 222.09822368621826,
      "epoch": 1.0321891110272854,
      "grad_norm": 0.16851135813365517,
      "kl": 1.2260894775390625,
      "learning_rate": 4.996809567029067e-07,
      "loss": 0.0012,
      "reward": 1.766071505844593,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6156
    },
    {
      "completion_length": 222.87947463989258,
      "epoch": 1.0325244142671528,
      "grad_norm": 0.3325081112548761,
      "kl": 0.226348876953125,
      "learning_rate": 4.996803505742229e-07,
      "loss": 0.0002,
      "reward": 1.7589286491274834,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7633929029107094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6158
    },
    {
      "completion_length": 215.20090293884277,
      "epoch": 1.0328597175070204,
      "grad_norm": 0.3898566757276845,
      "kl": 0.326934814453125,
      "learning_rate": 4.996797438706825e-07,
      "loss": 0.0003,
      "reward": 1.742857202887535,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7517857439815998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6160
    },
    {
      "completion_length": 216.13840293884277,
      "epoch": 1.033195020746888,
      "grad_norm": 0.29160838237888526,
      "kl": 0.46624755859375,
      "learning_rate": 4.996791365922867e-07,
      "loss": 0.0005,
      "reward": 1.6821429431438446,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.6821428947150707,
      "rewards/format_reward_func": 1.0,
      "step": 6162
    },
    {
      "completion_length": 207.17411708831787,
      "epoch": 1.0335303239867555,
      "grad_norm": 0.29139948351310624,
      "kl": 0.110260009765625,
      "learning_rate": 4.99678528739037e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714715719223,
      "rewards/format_reward_func": 1.0,
      "step": 6164
    },
    {
      "completion_length": 207.4062614440918,
      "epoch": 1.0338656272266231,
      "grad_norm": 0.4263314352373076,
      "kl": 0.1416778564453125,
      "learning_rate": 4.996779203109347e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 6166
    },
    {
      "completion_length": 208.258939743042,
      "epoch": 1.0342009304664905,
      "grad_norm": 0.20562322052764345,
      "kl": 0.4632568359375,
      "learning_rate": 4.996773113079813e-07,
      "loss": 0.0005,
      "reward": 1.8000000566244125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000156462193,
      "rewards/format_reward_func": 1.0,
      "step": 6168
    },
    {
      "completion_length": 214.71875953674316,
      "epoch": 1.0345362337063582,
      "grad_norm": 0.4857726664048818,
      "kl": 1.15155029296875,
      "learning_rate": 4.996767017301781e-07,
      "loss": 0.0012,
      "reward": 1.7892857566475868,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7982143051922321,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6170
    },
    {
      "completion_length": 208.66965198516846,
      "epoch": 1.0348715369462258,
      "grad_norm": 0.8186370164924367,
      "kl": 0.657196044921875,
      "learning_rate": 4.996760915775266e-07,
      "loss": 0.0007,
      "reward": 1.7321429327130318,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.732142873108387,
      "rewards/format_reward_func": 1.0,
      "step": 6172
    },
    {
      "completion_length": 201.98661613464355,
      "epoch": 1.0352068401860932,
      "grad_norm": 0.09891019549043235,
      "kl": 0.2893218994140625,
      "learning_rate": 4.996754808500282e-07,
      "loss": 0.0003,
      "reward": 1.7357143387198448,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7357143331319094,
      "rewards/format_reward_func": 1.0,
      "step": 6174
    },
    {
      "completion_length": 206.6696538925171,
      "epoch": 1.0355421434259608,
      "grad_norm": 0.3504160565048967,
      "kl": 0.156005859375,
      "learning_rate": 4.996748695476843e-07,
      "loss": 0.0002,
      "reward": 1.7892857864499092,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857529222965,
      "rewards/format_reward_func": 1.0,
      "step": 6176
    },
    {
      "completion_length": 211.74554443359375,
      "epoch": 1.0358774466658285,
      "grad_norm": 0.280817911357908,
      "kl": 0.1046600341796875,
      "learning_rate": 4.996742576704961e-07,
      "loss": 0.0001,
      "reward": 1.7964286506175995,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 6178
    },
    {
      "completion_length": 216.66072273254395,
      "epoch": 1.036212749905696,
      "grad_norm": 0.15394694928327,
      "kl": 0.117523193359375,
      "learning_rate": 4.996736452184653e-07,
      "loss": 0.0001,
      "reward": 1.7178572118282318,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7178571820259094,
      "rewards/format_reward_func": 1.0,
      "step": 6180
    },
    {
      "completion_length": 215.63393783569336,
      "epoch": 1.0365480531455635,
      "grad_norm": 0.1798703112423341,
      "kl": 0.0845489501953125,
      "learning_rate": 4.996730321915932e-07,
      "loss": 0.0001,
      "reward": 1.735714353621006,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143145054579,
      "rewards/format_reward_func": 1.0,
      "step": 6182
    },
    {
      "completion_length": 211.61161518096924,
      "epoch": 1.0368833563854312,
      "grad_norm": 0.26618095980539785,
      "kl": 0.10028076171875,
      "learning_rate": 4.996724185898812e-07,
      "loss": 0.0001,
      "reward": 1.8160714879631996,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.820535734295845,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6184
    },
    {
      "completion_length": 218.41072368621826,
      "epoch": 1.0372186596252986,
      "grad_norm": 0.23902131457280135,
      "kl": 0.090423583984375,
      "learning_rate": 4.996718044133306e-07,
      "loss": 0.0001,
      "reward": 1.8017857447266579,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.8062500357627869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6186
    },
    {
      "completion_length": 216.80357933044434,
      "epoch": 1.0375539628651662,
      "grad_norm": 0.6333183597495273,
      "kl": 0.0997314453125,
      "learning_rate": 4.996711896619432e-07,
      "loss": 0.0001,
      "reward": 1.7928571924567223,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 6188
    },
    {
      "completion_length": 218.7901906967163,
      "epoch": 1.0378892661050338,
      "grad_norm": 0.24219539600599282,
      "kl": 0.0910186767578125,
      "learning_rate": 4.9967057433572e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 6190
    },
    {
      "completion_length": 225.9419765472412,
      "epoch": 1.0382245693449013,
      "grad_norm": 0.38545151734950256,
      "kl": 0.108154296875,
      "learning_rate": 4.996699584346625e-07,
      "loss": 0.0001,
      "reward": 1.762500062584877,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6192
    },
    {
      "completion_length": 221.77679634094238,
      "epoch": 1.038559872584769,
      "grad_norm": 0.004252728264381877,
      "kl": 0.094970703125,
      "learning_rate": 4.996693419587723e-07,
      "loss": 0.0001,
      "reward": 1.7178572118282318,
      "reward_std": 0.005050762556493282,
      "rewards/equation_reward_func": 0.7178571671247482,
      "rewards/format_reward_func": 1.0,
      "step": 6194
    },
    {
      "completion_length": 222.7991180419922,
      "epoch": 1.0388951758246363,
      "grad_norm": 0.20910697584743013,
      "kl": 0.100341796875,
      "learning_rate": 4.996687249080507e-07,
      "loss": 0.0001,
      "reward": 1.8464286178350449,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8464286029338837,
      "rewards/format_reward_func": 1.0,
      "step": 6196
    },
    {
      "completion_length": 236.7991180419922,
      "epoch": 1.039230479064504,
      "grad_norm": 0.4548612314796721,
      "kl": 0.1113128662109375,
      "learning_rate": 4.99668107282499e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428880095482,
      "rewards/format_reward_func": 1.0,
      "step": 6198
    },
    {
      "completion_length": 221.1875114440918,
      "epoch": 1.0395657823043716,
      "grad_norm": 0.21093026878166982,
      "kl": 0.09210205078125,
      "learning_rate": 4.996674890821189e-07,
      "loss": 0.0001,
      "reward": 1.7500000894069672,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000279396772,
      "rewards/format_reward_func": 1.0,
      "step": 6200
    },
    {
      "completion_length": 222.7991180419922,
      "epoch": 1.039901085544239,
      "grad_norm": 0.04538039287639752,
      "kl": 0.0979766845703125,
      "learning_rate": 4.996668703069115e-07,
      "loss": 0.0001,
      "reward": 1.7464286088943481,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.7553571797907352,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6202
    },
    {
      "completion_length": 226.7991180419922,
      "epoch": 1.0402363887841066,
      "grad_norm": 0.2985622998817586,
      "kl": 0.099334716796875,
      "learning_rate": 4.996662509568785e-07,
      "loss": 0.0001,
      "reward": 1.764285795390606,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 6204
    },
    {
      "completion_length": 224.3705472946167,
      "epoch": 1.0405716920239743,
      "grad_norm": 0.3199587341752373,
      "kl": 0.118927001953125,
      "learning_rate": 4.996656310320213e-07,
      "loss": 0.0001,
      "reward": 1.7321429550647736,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321428768336773,
      "rewards/format_reward_func": 1.0,
      "step": 6206
    },
    {
      "completion_length": 240.6160831451416,
      "epoch": 1.0409069952638417,
      "grad_norm": 0.1554368722368833,
      "kl": 0.094329833984375,
      "learning_rate": 4.996650105323411e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 6208
    },
    {
      "completion_length": 233.73661518096924,
      "epoch": 1.0412422985037093,
      "grad_norm": 0.2616894712735538,
      "kl": 0.0997772216796875,
      "learning_rate": 4.996643894578396e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 6210
    },
    {
      "completion_length": 233.00893783569336,
      "epoch": 1.041577601743577,
      "grad_norm": 0.32797188169474895,
      "kl": 0.0866241455078125,
      "learning_rate": 4.996637678085181e-07,
      "loss": 0.0001,
      "reward": 1.725000075995922,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000461935997,
      "rewards/format_reward_func": 1.0,
      "step": 6212
    },
    {
      "completion_length": 213.83929538726807,
      "epoch": 1.0419129049834444,
      "grad_norm": 0.27628963395743616,
      "kl": 0.085784912109375,
      "learning_rate": 4.99663145584378e-07,
      "loss": 0.0001,
      "reward": 1.742857202887535,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571693599224,
      "rewards/format_reward_func": 1.0,
      "step": 6214
    },
    {
      "completion_length": 221.89733123779297,
      "epoch": 1.042248208223312,
      "grad_norm": 0.08516641647525623,
      "kl": 0.090362548828125,
      "learning_rate": 4.996625227854207e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7428571693599224,
      "rewards/format_reward_func": 1.0,
      "step": 6216
    },
    {
      "completion_length": 232.78126049041748,
      "epoch": 1.0425835114631794,
      "grad_norm": 0.2787046385004818,
      "kl": 0.0954742431640625,
      "learning_rate": 4.996618994116478e-07,
      "loss": 0.0001,
      "reward": 1.7267857864499092,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7312500346451998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6218
    },
    {
      "completion_length": 230.34376049041748,
      "epoch": 1.042918814703047,
      "grad_norm": 0.1534465803849126,
      "kl": 0.095611572265625,
      "learning_rate": 4.996612754630605e-07,
      "loss": 0.0001,
      "reward": 1.7125000730156898,
      "reward_std": 0.0328299580141902,
      "rewards/equation_reward_func": 0.7169643342494965,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6220
    },
    {
      "completion_length": 241.18751049041748,
      "epoch": 1.0432541179429147,
      "grad_norm": 0.2836709467862103,
      "kl": 0.087158203125,
      "learning_rate": 4.996606509396605e-07,
      "loss": 0.0001,
      "reward": 1.7107143923640251,
      "reward_std": 0.0858629634603858,
      "rewards/equation_reward_func": 0.7107143141329288,
      "rewards/format_reward_func": 1.0,
      "step": 6222
    },
    {
      "completion_length": 237.9196548461914,
      "epoch": 1.043589421182782,
      "grad_norm": 0.5012585855206936,
      "kl": 0.09515380859375,
      "learning_rate": 4.996600258414491e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286126196384,
      "rewards/format_reward_func": 1.0,
      "step": 6224
    },
    {
      "completion_length": 234.18751049041748,
      "epoch": 1.0439247244226497,
      "grad_norm": 0.2459476407984449,
      "kl": 0.0809173583984375,
      "learning_rate": 4.996594001684278e-07,
      "loss": 0.0001,
      "reward": 1.7714286595582962,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 6226
    },
    {
      "completion_length": 238.39733028411865,
      "epoch": 1.0442600276625174,
      "grad_norm": 0.14081129195328745,
      "kl": 0.07952880859375,
      "learning_rate": 4.996587739205979e-07,
      "loss": 0.0001,
      "reward": 1.751785770058632,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7562500182539225,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6228
    },
    {
      "completion_length": 237.11608219146729,
      "epoch": 1.0445953309023848,
      "grad_norm": 0.1969990280030385,
      "kl": 0.078887939453125,
      "learning_rate": 4.996581470979609e-07,
      "loss": 0.0001,
      "reward": 1.6928572207689285,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.6928571742027998,
      "rewards/format_reward_func": 1.0,
      "step": 6230
    },
    {
      "completion_length": 235.27679824829102,
      "epoch": 1.0449306341422524,
      "grad_norm": 0.10241176001437838,
      "kl": 0.077056884765625,
      "learning_rate": 4.996575197005184e-07,
      "loss": 0.0001,
      "reward": 1.8053571805357933,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8098214603960514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6232
    },
    {
      "completion_length": 233.43304634094238,
      "epoch": 1.04526593738212,
      "grad_norm": 0.43947870911254105,
      "kl": 0.0736846923828125,
      "learning_rate": 4.996568917282718e-07,
      "loss": 0.0001,
      "reward": 1.7500000894069672,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 6234
    },
    {
      "completion_length": 244.62947750091553,
      "epoch": 1.0456012406219874,
      "grad_norm": 0.22402884796565656,
      "kl": 0.0749969482421875,
      "learning_rate": 4.996562631812222e-07,
      "loss": 0.0001,
      "reward": 1.7178572192788124,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7178571857511997,
      "rewards/format_reward_func": 1.0,
      "step": 6236
    },
    {
      "completion_length": 244.3259038925171,
      "epoch": 1.045936543861855,
      "grad_norm": 0.275003424803172,
      "kl": 0.0874786376953125,
      "learning_rate": 4.996556340593715e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 6238
    },
    {
      "completion_length": 235.7321538925171,
      "epoch": 1.0462718471017227,
      "grad_norm": 0.22154834415764207,
      "kl": 0.0706024169921875,
      "learning_rate": 4.99655004362721e-07,
      "loss": 0.0001,
      "reward": 1.703571505844593,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.703571455553174,
      "rewards/format_reward_func": 1.0,
      "step": 6240
    },
    {
      "completion_length": 236.3437623977661,
      "epoch": 1.0466071503415901,
      "grad_norm": 0.39231322722768996,
      "kl": 0.07466888427734375,
      "learning_rate": 4.99654374091272e-07,
      "loss": 0.0001,
      "reward": 1.7522322162985802,
      "reward_std": 0.0473508988507092,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 6242
    },
    {
      "completion_length": 235.2857265472412,
      "epoch": 1.0469424535814578,
      "grad_norm": 0.23871235936534596,
      "kl": 0.067901611328125,
      "learning_rate": 4.996537432450261e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 6244
    },
    {
      "completion_length": 245.44197463989258,
      "epoch": 1.0472777568213252,
      "grad_norm": 0.13381440765275113,
      "kl": 0.06072998046875,
      "learning_rate": 4.996531118239848e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 6246
    },
    {
      "completion_length": 249.54911518096924,
      "epoch": 1.0476130600611928,
      "grad_norm": 0.17442154972673185,
      "kl": 0.071014404296875,
      "learning_rate": 4.996524798281494e-07,
      "loss": 0.0001,
      "reward": 1.6750000640749931,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6750000417232513,
      "rewards/format_reward_func": 1.0,
      "step": 6248
    },
    {
      "completion_length": 236.6116180419922,
      "epoch": 1.0479483633010604,
      "grad_norm": 0.1054946432492003,
      "kl": 0.06951904296875,
      "learning_rate": 4.996518472575214e-07,
      "loss": 0.0001,
      "reward": 1.803571492433548,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 6250
    },
    {
      "completion_length": 242.00001049041748,
      "epoch": 1.0482836665409279,
      "grad_norm": 0.21478339945524438,
      "kl": 0.0652008056640625,
      "learning_rate": 4.996512141121023e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 6252
    },
    {
      "completion_length": 236.27679538726807,
      "epoch": 1.0486189697807955,
      "grad_norm": 0.2525516533339607,
      "kl": 0.079071044921875,
      "learning_rate": 4.996505803918936e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 6254
    },
    {
      "completion_length": 242.00894260406494,
      "epoch": 1.0489542730206631,
      "grad_norm": 0.19916546296729642,
      "kl": 0.071990966796875,
      "learning_rate": 4.996499460968966e-07,
      "loss": 0.0001,
      "reward": 1.816071480512619,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8205357417464256,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6256
    },
    {
      "completion_length": 247.7187614440918,
      "epoch": 1.0492895762605305,
      "grad_norm": 0.5188461088710983,
      "kl": 0.518463134765625,
      "learning_rate": 4.996493112271129e-07,
      "loss": 0.0005,
      "reward": 1.751785784959793,
      "reward_std": 0.10859139822423458,
      "rewards/equation_reward_func": 0.765178594738245,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6258
    },
    {
      "completion_length": 250.15626049041748,
      "epoch": 1.0496248795003982,
      "grad_norm": 0.2216511665362915,
      "kl": 0.06902313232421875,
      "learning_rate": 4.996486757825439e-07,
      "loss": 0.0001,
      "reward": 1.7160715013742447,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7205357663333416,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6260
    },
    {
      "completion_length": 255.28572463989258,
      "epoch": 1.0499601827402658,
      "grad_norm": 0.3203314600754042,
      "kl": 0.109161376953125,
      "learning_rate": 4.996480397631911e-07,
      "loss": 0.0001,
      "reward": 1.7214286476373672,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7303571775555611,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6262
    },
    {
      "completion_length": 242.7857265472412,
      "epoch": 1.0502954859801332,
      "grad_norm": 0.302882520820471,
      "kl": 0.0699310302734375,
      "learning_rate": 4.996474031690561e-07,
      "loss": 0.0001,
      "reward": 1.8053572103381157,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8098214529454708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6264
    },
    {
      "completion_length": 254.93304538726807,
      "epoch": 1.0506307892200009,
      "grad_norm": 0.24384354682916257,
      "kl": 0.0830535888671875,
      "learning_rate": 4.996467660001399e-07,
      "loss": 0.0001,
      "reward": 1.744642935693264,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6266
    },
    {
      "completion_length": 249.7187623977661,
      "epoch": 1.0509660924598685,
      "grad_norm": 0.3250066916248128,
      "kl": 0.0789337158203125,
      "learning_rate": 4.996461282564446e-07,
      "loss": 0.0001,
      "reward": 1.708928644657135,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7223214507102966,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6268
    },
    {
      "completion_length": 246.92411994934082,
      "epoch": 1.051301395699736,
      "grad_norm": 0.2190682285191929,
      "kl": 0.084503173828125,
      "learning_rate": 4.996454899379711e-07,
      "loss": 0.0001,
      "reward": 1.7339286506175995,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7383928820490837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6270
    },
    {
      "completion_length": 259.10715198516846,
      "epoch": 1.0516366989396035,
      "grad_norm": 0.27884125227950174,
      "kl": 0.0814361572265625,
      "learning_rate": 4.996448510447213e-07,
      "loss": 0.0001,
      "reward": 1.705357238650322,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.709821455180645,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6272
    },
    {
      "completion_length": 242.6116180419922,
      "epoch": 1.051972002179471,
      "grad_norm": 0.26117811635112054,
      "kl": 0.0708770751953125,
      "learning_rate": 4.996442115766965e-07,
      "loss": 0.0001,
      "reward": 1.7857143580913544,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7946428880095482,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6274
    },
    {
      "completion_length": 253.24555110931396,
      "epoch": 1.0523073054193386,
      "grad_norm": 0.20232160553432557,
      "kl": 0.0785369873046875,
      "learning_rate": 4.996435715338981e-07,
      "loss": 0.0001,
      "reward": 1.7803571969270706,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7937500290572643,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6276
    },
    {
      "completion_length": 250.5491180419922,
      "epoch": 1.0526426086592062,
      "grad_norm": 0.28832289892932167,
      "kl": 0.06787109375,
      "learning_rate": 4.996429309163276e-07,
      "loss": 0.0001,
      "reward": 1.7392858043313026,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7660714611411095,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 6278
    },
    {
      "completion_length": 247.56697940826416,
      "epoch": 1.0529779118990736,
      "grad_norm": 0.23480186692840185,
      "kl": 0.0709991455078125,
      "learning_rate": 4.996422897239866e-07,
      "loss": 0.0001,
      "reward": 1.7321429252624512,
      "reward_std": 0.10606601741164923,
      "rewards/equation_reward_func": 0.7589286006987095,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 6280
    },
    {
      "completion_length": 257.071439743042,
      "epoch": 1.0533132151389413,
      "grad_norm": 0.3109790026339082,
      "kl": 0.0759124755859375,
      "learning_rate": 4.996416479568763e-07,
      "loss": 0.0001,
      "reward": 1.7428572177886963,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7517857290804386,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6282
    },
    {
      "completion_length": 243.58483219146729,
      "epoch": 1.053648518378809,
      "grad_norm": 0.21916995248392268,
      "kl": 0.06939697265625,
      "learning_rate": 4.996410056149986e-07,
      "loss": 0.0001,
      "reward": 1.7589286640286446,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928842842579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6284
    },
    {
      "completion_length": 251.3303680419922,
      "epoch": 1.0539838216186763,
      "grad_norm": 0.1085527793568352,
      "kl": 0.07000732421875,
      "learning_rate": 4.996403626983547e-07,
      "loss": 0.0001,
      "reward": 1.7107143476605415,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7196428924798965,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6286
    },
    {
      "completion_length": 257.58483695983887,
      "epoch": 1.054319124858544,
      "grad_norm": 0.27881947506588706,
      "kl": 0.0749053955078125,
      "learning_rate": 4.996397192069461e-07,
      "loss": 0.0001,
      "reward": 1.6857143566012383,
      "reward_std": 0.11111677996814251,
      "rewards/equation_reward_func": 0.7125000320374966,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 6288
    },
    {
      "completion_length": 251.14733219146729,
      "epoch": 1.0546544280984116,
      "grad_norm": 0.27651003259180323,
      "kl": 0.0708160400390625,
      "learning_rate": 4.996390751407741e-07,
      "loss": 0.0001,
      "reward": 1.7500000596046448,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 6290
    },
    {
      "completion_length": 243.77679538726807,
      "epoch": 1.054989731338279,
      "grad_norm": 0.21381289177800553,
      "kl": 0.0681915283203125,
      "learning_rate": 4.996384304998406e-07,
      "loss": 0.0001,
      "reward": 1.7803571969270706,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7848214525729418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6292
    },
    {
      "completion_length": 246.5089406967163,
      "epoch": 1.0553250345781466,
      "grad_norm": 0.22910722937156328,
      "kl": 0.0751953125,
      "learning_rate": 4.99637785284147e-07,
      "loss": 0.0001,
      "reward": 1.775000050663948,
      "reward_std": 0.04545686487108469,
      "rewards/equation_reward_func": 0.7839286141097546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6294
    },
    {
      "completion_length": 259.4241180419922,
      "epoch": 1.0556603378180143,
      "grad_norm": 0.7797186367045612,
      "kl": 0.1149749755859375,
      "learning_rate": 4.996371394936944e-07,
      "loss": 0.0001,
      "reward": 1.7732143551111221,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7776786014437675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6296
    },
    {
      "completion_length": 250.4285831451416,
      "epoch": 1.0559956410578817,
      "grad_norm": 1.2601252607687894,
      "kl": 0.0680389404296875,
      "learning_rate": 4.996364931284847e-07,
      "loss": 0.0001,
      "reward": 1.773214340209961,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7776785995811224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6298
    },
    {
      "completion_length": 238.5803680419922,
      "epoch": 1.0563309442977493,
      "grad_norm": 0.2717918614938203,
      "kl": 0.078369140625,
      "learning_rate": 4.996358461885192e-07,
      "loss": 0.0001,
      "reward": 1.7910714969038963,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7955357395112514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6300
    },
    {
      "completion_length": 242.43304634094238,
      "epoch": 1.0566662475376167,
      "grad_norm": 0.09746477369521411,
      "kl": 0.079925537109375,
      "learning_rate": 4.996351986737993e-07,
      "loss": 0.0001,
      "reward": 1.7303572073578835,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7348214611411095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6302
    },
    {
      "completion_length": 245.0178680419922,
      "epoch": 1.0570015507774844,
      "grad_norm": 0.13468086713952232,
      "kl": 0.076995849609375,
      "learning_rate": 4.996345505843268e-07,
      "loss": 0.0001,
      "reward": 1.7589286342263222,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7633928842842579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6304
    },
    {
      "completion_length": 250.1875114440918,
      "epoch": 1.057336854017352,
      "grad_norm": 0.18329931190892818,
      "kl": 0.0720367431640625,
      "learning_rate": 4.996339019201029e-07,
      "loss": 0.0001,
      "reward": 1.7446429207921028,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6306
    },
    {
      "completion_length": 238.9419765472412,
      "epoch": 1.0576721572572194,
      "grad_norm": 0.20137418059888398,
      "kl": 0.0643157958984375,
      "learning_rate": 4.996332526811292e-07,
      "loss": 0.0001,
      "reward": 1.773214340209961,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7776786163449287,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6308
    },
    {
      "completion_length": 246.35715103149414,
      "epoch": 1.058007460497087,
      "grad_norm": 0.16433502622545273,
      "kl": 0.0754547119140625,
      "learning_rate": 4.996326028674073e-07,
      "loss": 0.0001,
      "reward": 1.716071493923664,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7205357421189547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6310
    },
    {
      "completion_length": 247.68751335144043,
      "epoch": 1.0583427637369547,
      "grad_norm": 0.19212147064966212,
      "kl": 0.0781707763671875,
      "learning_rate": 4.996319524789385e-07,
      "loss": 0.0001,
      "reward": 1.7589286267757416,
      "reward_std": 0.017677669413387775,
      "rewards/equation_reward_func": 0.7633928880095482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6312
    },
    {
      "completion_length": 242.81697273254395,
      "epoch": 1.058678066976822,
      "grad_norm": 0.09429431247365397,
      "kl": 0.0767822265625,
      "learning_rate": 4.996313015157244e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 6314
    },
    {
      "completion_length": 245.39733123779297,
      "epoch": 1.0590133702166897,
      "grad_norm": 0.18421668642270214,
      "kl": 0.09881591796875,
      "learning_rate": 4.996306499777665e-07,
      "loss": 0.0001,
      "reward": 1.762500062584877,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7758928686380386,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6316
    },
    {
      "completion_length": 245.1116189956665,
      "epoch": 1.0593486734565574,
      "grad_norm": 0.3807004269832346,
      "kl": 0.0757904052734375,
      "learning_rate": 4.996299978650662e-07,
      "loss": 0.0001,
      "reward": 1.760714367032051,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 6318
    },
    {
      "completion_length": 251.22322463989258,
      "epoch": 1.0596839766964248,
      "grad_norm": 0.2114666808769925,
      "kl": 0.078765869140625,
      "learning_rate": 4.996293451776252e-07,
      "loss": 0.0001,
      "reward": 1.7500000298023224,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.7678571939468384,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6320
    },
    {
      "completion_length": 246.852689743042,
      "epoch": 1.0600192799362924,
      "grad_norm": 0.16649081805675256,
      "kl": 0.0686492919921875,
      "learning_rate": 4.996286919154449e-07,
      "loss": 0.0001,
      "reward": 1.7625000700354576,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6322
    },
    {
      "completion_length": 247.61608123779297,
      "epoch": 1.06035458317616,
      "grad_norm": 0.30420478683403074,
      "kl": 0.0810699462890625,
      "learning_rate": 4.996280380785267e-07,
      "loss": 0.0001,
      "reward": 1.7410715073347092,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.745535746216774,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6324
    },
    {
      "completion_length": 251.58483219146729,
      "epoch": 1.0606898864160275,
      "grad_norm": 0.23160704711198593,
      "kl": 0.0971832275390625,
      "learning_rate": 4.996273836668723e-07,
      "loss": 0.0001,
      "reward": 1.7571429163217545,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7660714648663998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6326
    },
    {
      "completion_length": 244.9642972946167,
      "epoch": 1.061025189655895,
      "grad_norm": 0.3936892033816459,
      "kl": 0.0736236572265625,
      "learning_rate": 4.996267286804831e-07,
      "loss": 0.0001,
      "reward": 1.7446429282426834,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7491071652621031,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6328
    },
    {
      "completion_length": 241.1116180419922,
      "epoch": 1.0613604928957625,
      "grad_norm": 0.1383559369834235,
      "kl": 0.0700225830078125,
      "learning_rate": 4.996260731193606e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 6330
    },
    {
      "completion_length": 245.1473331451416,
      "epoch": 1.0616957961356301,
      "grad_norm": 0.3744539358250854,
      "kl": 0.0670928955078125,
      "learning_rate": 4.996254169835063e-07,
      "loss": 0.0001,
      "reward": 1.8160715028643608,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.8205357491970062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6332
    },
    {
      "completion_length": 247.540189743042,
      "epoch": 1.0620310993754978,
      "grad_norm": 0.4886645560494357,
      "kl": 0.0726165771484375,
      "learning_rate": 4.996247602729216e-07,
      "loss": 0.0001,
      "reward": 1.7267857640981674,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.731250025331974,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6334
    },
    {
      "completion_length": 251.43304538726807,
      "epoch": 1.0623664026153652,
      "grad_norm": 0.1689562542398881,
      "kl": 0.0782623291015625,
      "learning_rate": 4.996241029876084e-07,
      "loss": 0.0001,
      "reward": 1.7910714820027351,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7955357432365417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6336
    },
    {
      "completion_length": 263.352689743042,
      "epoch": 1.0627017058552328,
      "grad_norm": 0.26788844139713214,
      "kl": 0.13543701171875,
      "learning_rate": 4.996234451275678e-07,
      "loss": 0.0001,
      "reward": 1.728571504354477,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6338
    },
    {
      "completion_length": 247.23661994934082,
      "epoch": 1.0630370090951005,
      "grad_norm": 0.19132039665083433,
      "kl": 0.1452789306640625,
      "learning_rate": 4.996227866928016e-07,
      "loss": 0.0001,
      "reward": 1.7285715267062187,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7375000230967999,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6340
    },
    {
      "completion_length": 258.4419746398926,
      "epoch": 1.0633723123349679,
      "grad_norm": 0.19874075640970054,
      "kl": 0.1968841552734375,
      "learning_rate": 4.996221276833111e-07,
      "loss": 0.0002,
      "reward": 1.7607143446803093,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7696428820490837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6342
    },
    {
      "completion_length": 251.9464406967163,
      "epoch": 1.0637076155748355,
      "grad_norm": 0.23396514743791882,
      "kl": 0.1824951171875,
      "learning_rate": 4.996214680990979e-07,
      "loss": 0.0002,
      "reward": 1.7767857983708382,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7812500111758709,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6344
    },
    {
      "completion_length": 238.31251049041748,
      "epoch": 1.0640429188147031,
      "grad_norm": 0.43415076467434777,
      "kl": 0.073333740234375,
      "learning_rate": 4.996208079401635e-07,
      "loss": 0.0001,
      "reward": 1.7839286550879478,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7883928865194321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6346
    },
    {
      "completion_length": 245.0089406967163,
      "epoch": 1.0643782220545706,
      "grad_norm": 0.4964206120947802,
      "kl": 0.758087158203125,
      "learning_rate": 4.996201472065096e-07,
      "loss": 0.0008,
      "reward": 1.7321429401636124,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7410714626312256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6348
    },
    {
      "completion_length": 258.4107275009155,
      "epoch": 1.0647135252944382,
      "grad_norm": 0.2686395974838121,
      "kl": 0.7075042724609375,
      "learning_rate": 4.996194858981374e-07,
      "loss": 0.0007,
      "reward": 1.7535714954137802,
      "reward_std": 0.0656599160283804,
      "rewards/equation_reward_func": 0.7625000365078449,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6350
    },
    {
      "completion_length": 241.78572463989258,
      "epoch": 1.0650488285343056,
      "grad_norm": 0.20670819187440403,
      "kl": 0.2355194091796875,
      "learning_rate": 4.996188240150486e-07,
      "loss": 0.0002,
      "reward": 1.8000000566244125,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 6352
    },
    {
      "completion_length": 258.058048248291,
      "epoch": 1.0653841317741732,
      "grad_norm": 0.24306248609979544,
      "kl": 0.6288604736328125,
      "learning_rate": 4.996181615572449e-07,
      "loss": 0.0006,
      "reward": 1.7375000938773155,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7419643141329288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6354
    },
    {
      "completion_length": 239.61608028411865,
      "epoch": 1.0657194350140409,
      "grad_norm": 0.28623101171516113,
      "kl": 0.140655517578125,
      "learning_rate": 4.996174985247275e-07,
      "loss": 0.0001,
      "reward": 1.7178572192788124,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7267857454717159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6356
    },
    {
      "completion_length": 244.5759048461914,
      "epoch": 1.0660547382539083,
      "grad_norm": 0.18153326342052306,
      "kl": 0.0954132080078125,
      "learning_rate": 4.996168349174981e-07,
      "loss": 0.0001,
      "reward": 1.7196429297327995,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7241071779280901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6358
    },
    {
      "completion_length": 247.27679634094238,
      "epoch": 1.066390041493776,
      "grad_norm": 0.35221345372591045,
      "kl": 0.691741943359375,
      "learning_rate": 4.996161707355582e-07,
      "loss": 0.0007,
      "reward": 1.7500000670552254,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7589286081492901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6360
    },
    {
      "completion_length": 255.2366189956665,
      "epoch": 1.0667253447336436,
      "grad_norm": 0.5279583818578518,
      "kl": 0.78302001953125,
      "learning_rate": 4.996155059789093e-07,
      "loss": 0.0008,
      "reward": 1.6785714775323868,
      "reward_std": 0.11111677903681993,
      "rewards/equation_reward_func": 0.7053571660071611,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 6362
    },
    {
      "completion_length": 256.92858123779297,
      "epoch": 1.067060647973511,
      "grad_norm": 0.2476382497789468,
      "kl": 0.3397369384765625,
      "learning_rate": 4.996148406475529e-07,
      "loss": 0.0003,
      "reward": 1.7464286461472511,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 6364
    },
    {
      "completion_length": 251.4017972946167,
      "epoch": 1.0673959512133786,
      "grad_norm": 0.22177943064580702,
      "kl": 1.80938720703125,
      "learning_rate": 4.996141747414906e-07,
      "loss": 0.0018,
      "reward": 1.7928571924567223,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.8017857410013676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6366
    },
    {
      "completion_length": 249.25447750091553,
      "epoch": 1.0677312544532462,
      "grad_norm": 0.2940764790633873,
      "kl": 0.3301849365234375,
      "learning_rate": 4.99613508260724e-07,
      "loss": 0.0003,
      "reward": 1.7321429029107094,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7410714533179998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6368
    },
    {
      "completion_length": 262.54465770721436,
      "epoch": 1.0680665576931136,
      "grad_norm": 0.36158647929313403,
      "kl": 0.313018798828125,
      "learning_rate": 4.996128412052546e-07,
      "loss": 0.0003,
      "reward": 1.7517857626080513,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7651786096394062,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6370
    },
    {
      "completion_length": 265.00894260406494,
      "epoch": 1.0684018609329813,
      "grad_norm": 0.2477564882494541,
      "kl": 0.3001556396484375,
      "learning_rate": 4.996121735750838e-07,
      "loss": 0.0003,
      "reward": 1.7071429267525673,
      "reward_std": 0.1010152529925108,
      "rewards/equation_reward_func": 0.7250000275671482,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6372
    },
    {
      "completion_length": 253.47322463989258,
      "epoch": 1.068737164172849,
      "grad_norm": 0.21786370922386733,
      "kl": 0.0959014892578125,
      "learning_rate": 4.996115053702132e-07,
      "loss": 0.0001,
      "reward": 1.7375000640749931,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643197208643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6374
    },
    {
      "completion_length": 252.44644165039062,
      "epoch": 1.0690724674127163,
      "grad_norm": 0.16300399710153712,
      "kl": 0.1212158203125,
      "learning_rate": 4.996108365906444e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7910714522004128,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6376
    },
    {
      "completion_length": 255.8482255935669,
      "epoch": 1.069407770652584,
      "grad_norm": 0.4202332630677491,
      "kl": 0.266632080078125,
      "learning_rate": 4.996101672363789e-07,
      "loss": 0.0003,
      "reward": 1.7464286461472511,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7464286163449287,
      "rewards/format_reward_func": 1.0,
      "step": 6378
    },
    {
      "completion_length": 250.80358219146729,
      "epoch": 1.0697430738924516,
      "grad_norm": 0.21634007380193282,
      "kl": 0.1393280029296875,
      "learning_rate": 4.996094973074182e-07,
      "loss": 0.0001,
      "reward": 1.7910715118050575,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7955357432365417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6380
    },
    {
      "completion_length": 240.21429634094238,
      "epoch": 1.070078377132319,
      "grad_norm": 0.2353217723544025,
      "kl": 0.1108551025390625,
      "learning_rate": 4.996088268037641e-07,
      "loss": 0.0001,
      "reward": 1.7517857924103737,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500312924385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6382
    },
    {
      "completion_length": 254.6250114440918,
      "epoch": 1.0704136803721866,
      "grad_norm": 0.22977953063269516,
      "kl": 0.0945281982421875,
      "learning_rate": 4.996081557254177e-07,
      "loss": 0.0001,
      "reward": 1.7696429193019867,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7741071600466967,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6384
    },
    {
      "completion_length": 244.19643592834473,
      "epoch": 1.070748983612054,
      "grad_norm": 0.22856891401563256,
      "kl": 0.212188720703125,
      "learning_rate": 4.996074840723809e-07,
      "loss": 0.0002,
      "reward": 1.7642858251929283,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 6386
    },
    {
      "completion_length": 236.9062623977661,
      "epoch": 1.0710842868519217,
      "grad_norm": 0.20213076895736504,
      "kl": 0.105621337890625,
      "learning_rate": 4.99606811844655e-07,
      "loss": 0.0001,
      "reward": 1.7785715162754059,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714380443096,
      "rewards/format_reward_func": 1.0,
      "step": 6388
    },
    {
      "completion_length": 248.64733409881592,
      "epoch": 1.0714195900917893,
      "grad_norm": 0.16060784507836098,
      "kl": 0.07293701171875,
      "learning_rate": 4.996061390422418e-07,
      "loss": 0.0001,
      "reward": 1.8035714998841286,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8125000335276127,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6390
    },
    {
      "completion_length": 237.47322273254395,
      "epoch": 1.0717548933316567,
      "grad_norm": 0.3744916426993643,
      "kl": 0.1375274658203125,
      "learning_rate": 4.996054656651427e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7732143066823483,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6392
    },
    {
      "completion_length": 240.9419765472412,
      "epoch": 1.0720901965715244,
      "grad_norm": 0.5524514166335937,
      "kl": 0.1202850341796875,
      "learning_rate": 4.996047917133591e-07,
      "loss": 0.0001,
      "reward": 1.7482143566012383,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7616071775555611,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6394
    },
    {
      "completion_length": 247.79465293884277,
      "epoch": 1.072425499811392,
      "grad_norm": 0.1931412468701336,
      "kl": 0.1047821044921875,
      "learning_rate": 4.996041171868929e-07,
      "loss": 0.0001,
      "reward": 1.7214286476373672,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7214286103844643,
      "rewards/format_reward_func": 1.0,
      "step": 6396
    },
    {
      "completion_length": 231.8973331451416,
      "epoch": 1.0727608030512594,
      "grad_norm": 0.16039896848106713,
      "kl": 0.07208251953125,
      "learning_rate": 4.996034420857454e-07,
      "loss": 0.0001,
      "reward": 1.7839286401867867,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6398
    },
    {
      "completion_length": 247.7634038925171,
      "epoch": 1.073096106291127,
      "grad_norm": 0.40971345021003636,
      "kl": 0.078338623046875,
      "learning_rate": 4.996027664099183e-07,
      "loss": 0.0001,
      "reward": 1.7785715162754059,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 6400
    },
    {
      "completion_length": 248.1741189956665,
      "epoch": 1.0734314095309947,
      "grad_norm": 0.1190091501059919,
      "kl": 0.104644775390625,
      "learning_rate": 4.996020901594131e-07,
      "loss": 0.0001,
      "reward": 1.748214341700077,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7526785992085934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6402
    },
    {
      "completion_length": 244.5491180419922,
      "epoch": 1.073766712770862,
      "grad_norm": 0.20764049742946922,
      "kl": 0.074798583984375,
      "learning_rate": 4.996014133342314e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285798370838,
      "rewards/format_reward_func": 1.0,
      "step": 6404
    },
    {
      "completion_length": 246.4017972946167,
      "epoch": 1.0741020160107297,
      "grad_norm": 0.21780243516392667,
      "kl": 0.067718505859375,
      "learning_rate": 4.996007359343745e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 6406
    },
    {
      "completion_length": 246.27679920196533,
      "epoch": 1.0744373192505972,
      "grad_norm": 0.48129432871493416,
      "kl": 0.08587646484375,
      "learning_rate": 4.996000579598442e-07,
      "loss": 0.0001,
      "reward": 1.7071429565548897,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7160714659839869,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6408
    },
    {
      "completion_length": 245.6562614440918,
      "epoch": 1.0747726224904648,
      "grad_norm": 0.24065879335741383,
      "kl": 0.0702972412109375,
      "learning_rate": 4.995993794106421e-07,
      "loss": 0.0001,
      "reward": 1.8000000715255737,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 6410
    },
    {
      "completion_length": 247.36608123779297,
      "epoch": 1.0751079257303324,
      "grad_norm": 0.26497871394053646,
      "kl": 0.0687713623046875,
      "learning_rate": 4.995987002867698e-07,
      "loss": 0.0001,
      "reward": 1.7553572207689285,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7598214596509933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6412
    },
    {
      "completion_length": 248.27679824829102,
      "epoch": 1.0754432289701998,
      "grad_norm": 0.41136003473797084,
      "kl": 0.0745391845703125,
      "learning_rate": 4.995980205882286e-07,
      "loss": 0.0001,
      "reward": 1.7392858043313026,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 1.0,
      "step": 6414
    },
    {
      "completion_length": 247.2678689956665,
      "epoch": 1.0757785322100675,
      "grad_norm": 0.21398005997336908,
      "kl": 0.064544677734375,
      "learning_rate": 4.995973403150202e-07,
      "loss": 0.0001,
      "reward": 1.719642959535122,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.724107176065445,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6416
    },
    {
      "completion_length": 234.8259038925171,
      "epoch": 1.076113835449935,
      "grad_norm": 0.25680062865685876,
      "kl": 0.0615081787109375,
      "learning_rate": 4.995966594671462e-07,
      "loss": 0.0001,
      "reward": 1.8071429207921028,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.8071428835391998,
      "rewards/format_reward_func": 1.0,
      "step": 6418
    },
    {
      "completion_length": 239.17858409881592,
      "epoch": 1.0764491386898025,
      "grad_norm": 0.3022971347656876,
      "kl": 0.05865478515625,
      "learning_rate": 4.995959780446081e-07,
      "loss": 0.0001,
      "reward": 1.7687500417232513,
      "reward_std": 0.04419417306780815,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 6420
    },
    {
      "completion_length": 240.0357255935669,
      "epoch": 1.0767844419296702,
      "grad_norm": 0.3100097102368012,
      "kl": 0.065399169921875,
      "learning_rate": 4.995952960474076e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.800000037997961,
      "rewards/format_reward_func": 1.0,
      "step": 6422
    },
    {
      "completion_length": 250.1116189956665,
      "epoch": 1.0771197451695378,
      "grad_norm": 0.12420128973486144,
      "kl": 0.0625457763671875,
      "learning_rate": 4.995946134755462e-07,
      "loss": 0.0001,
      "reward": 1.7500000521540642,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 6424
    },
    {
      "completion_length": 252.18751335144043,
      "epoch": 1.0774550484094052,
      "grad_norm": 0.16978968685993892,
      "kl": 0.067596435546875,
      "learning_rate": 4.995939303290253e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 1.0,
      "step": 6426
    },
    {
      "completion_length": 254.5044755935669,
      "epoch": 1.0777903516492728,
      "grad_norm": 0.33574965100851467,
      "kl": 0.064117431640625,
      "learning_rate": 4.995932466078468e-07,
      "loss": 0.0001,
      "reward": 1.762500062584877,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7669643275439739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6428
    },
    {
      "completion_length": 249.9241180419922,
      "epoch": 1.0781256548891403,
      "grad_norm": 0.22794782342785525,
      "kl": 0.06793212890625,
      "learning_rate": 4.995925623120121e-07,
      "loss": 0.0001,
      "reward": 1.6964286491274834,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7053571809083223,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6430
    },
    {
      "completion_length": 251.34376335144043,
      "epoch": 1.0784609581290079,
      "grad_norm": 0.21343348359286007,
      "kl": 0.064117431640625,
      "learning_rate": 4.995918774415226e-07,
      "loss": 0.0001,
      "reward": 1.7964286357164383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 6432
    },
    {
      "completion_length": 250.5178680419922,
      "epoch": 1.0787962613688755,
      "grad_norm": 0.2154335777824777,
      "kl": 0.073577880859375,
      "learning_rate": 4.995911919963802e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 6434
    },
    {
      "completion_length": 256.2455520629883,
      "epoch": 1.079131564608743,
      "grad_norm": 0.22964136065354926,
      "kl": 0.0595703125,
      "learning_rate": 4.995905059765863e-07,
      "loss": 0.0001,
      "reward": 1.79464291036129,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.8080357350409031,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6436
    },
    {
      "completion_length": 255.1294765472412,
      "epoch": 1.0794668678486106,
      "grad_norm": 0.30185869514240066,
      "kl": 0.069580078125,
      "learning_rate": 4.995898193821426e-07,
      "loss": 0.0001,
      "reward": 1.753571517765522,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 6438
    },
    {
      "completion_length": 254.9241180419922,
      "epoch": 1.0798021710884782,
      "grad_norm": 0.22874675324827642,
      "kl": 0.060516357421875,
      "learning_rate": 4.995891322130505e-07,
      "loss": 0.0001,
      "reward": 1.7625000774860382,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643014669418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6440
    },
    {
      "completion_length": 245.75447845458984,
      "epoch": 1.0801374743283456,
      "grad_norm": 0.15348513336418879,
      "kl": 0.058837890625,
      "learning_rate": 4.995884444693117e-07,
      "loss": 0.0001,
      "reward": 1.835714340209961,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8357143104076385,
      "rewards/format_reward_func": 1.0,
      "step": 6442
    },
    {
      "completion_length": 247.2366189956665,
      "epoch": 1.0804727775682132,
      "grad_norm": 0.17123096708578153,
      "kl": 0.0656585693359375,
      "learning_rate": 4.995877561509278e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964286170899868,
      "rewards/format_reward_func": 1.0,
      "step": 6444
    },
    {
      "completion_length": 257.4598340988159,
      "epoch": 1.0808080808080809,
      "grad_norm": 0.28229072762750085,
      "kl": 0.069366455078125,
      "learning_rate": 4.995870672579003e-07,
      "loss": 0.0001,
      "reward": 1.7361607998609543,
      "reward_std": 0.10038390709087253,
      "rewards/equation_reward_func": 0.7419643122702837,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 6446
    },
    {
      "completion_length": 249.0089406967163,
      "epoch": 1.0811433840479483,
      "grad_norm": 0.2814686950997999,
      "kl": 0.08123779296875,
      "learning_rate": 4.995863777902308e-07,
      "loss": 0.0001,
      "reward": 1.8196429088711739,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8241071663796902,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6448
    },
    {
      "completion_length": 251.23215293884277,
      "epoch": 1.081478687287816,
      "grad_norm": 0.18263971276236252,
      "kl": 0.0647430419921875,
      "learning_rate": 4.99585687747921e-07,
      "loss": 0.0001,
      "reward": 1.7839286401867867,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6450
    },
    {
      "completion_length": 246.1071548461914,
      "epoch": 1.0818139905276836,
      "grad_norm": 0.3340344630893179,
      "kl": 0.07000732421875,
      "learning_rate": 4.995849971309723e-07,
      "loss": 0.0001,
      "reward": 1.7785715088248253,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 1.0,
      "step": 6452
    },
    {
      "completion_length": 245.1875114440918,
      "epoch": 1.082149293767551,
      "grad_norm": 0.23800132416609462,
      "kl": 0.0767669677734375,
      "learning_rate": 4.995843059393865e-07,
      "loss": 0.0001,
      "reward": 1.7410714775323868,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357499420643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6454
    },
    {
      "completion_length": 247.5000114440918,
      "epoch": 1.0824845970074186,
      "grad_norm": 0.19794952988984094,
      "kl": 0.07769775390625,
      "learning_rate": 4.995836141731651e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285887777805,
      "rewards/format_reward_func": 1.0,
      "step": 6456
    },
    {
      "completion_length": 244.90179634094238,
      "epoch": 1.0828199002472862,
      "grad_norm": 0.2173504470847555,
      "kl": 0.0712127685546875,
      "learning_rate": 4.995829218323096e-07,
      "loss": 0.0001,
      "reward": 1.7053572162985802,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7098214700818062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6458
    },
    {
      "completion_length": 258.9375123977661,
      "epoch": 1.0831552034871537,
      "grad_norm": 0.24258856009385474,
      "kl": 0.0692596435546875,
      "learning_rate": 4.995822289168216e-07,
      "loss": 0.0001,
      "reward": 1.7035715207457542,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.703571442514658,
      "rewards/format_reward_func": 1.0,
      "step": 6460
    },
    {
      "completion_length": 254.15626525878906,
      "epoch": 1.0834905067270213,
      "grad_norm": 0.21434048158714838,
      "kl": 0.0685882568359375,
      "learning_rate": 4.99581535426703e-07,
      "loss": 0.0001,
      "reward": 1.7196429297327995,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7241071909666061,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6462
    },
    {
      "completion_length": 255.51786708831787,
      "epoch": 1.0838258099668887,
      "grad_norm": 0.322625348089809,
      "kl": 0.066741943359375,
      "learning_rate": 4.99580841361955e-07,
      "loss": 0.0001,
      "reward": 1.7303572371602058,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7348214536905289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6464
    },
    {
      "completion_length": 262.401798248291,
      "epoch": 1.0841611132067563,
      "grad_norm": 0.18824960966371346,
      "kl": 0.0705108642578125,
      "learning_rate": 4.995801467225794e-07,
      "loss": 0.0001,
      "reward": 1.7375000640749931,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.7508928924798965,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6466
    },
    {
      "completion_length": 253.3259048461914,
      "epoch": 1.084496416446624,
      "grad_norm": 0.32009667898891864,
      "kl": 0.074462890625,
      "learning_rate": 4.995794515085778e-07,
      "loss": 0.0001,
      "reward": 1.7517857775092125,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500238418579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6468
    },
    {
      "completion_length": 252.25001525878906,
      "epoch": 1.0848317196864914,
      "grad_norm": 0.23876316643511034,
      "kl": 0.069305419921875,
      "learning_rate": 4.995787557199518e-07,
      "loss": 0.0001,
      "reward": 1.764285795390606,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7732143066823483,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6470
    },
    {
      "completion_length": 255.30804920196533,
      "epoch": 1.085167022926359,
      "grad_norm": 0.26642096787601005,
      "kl": 0.069610595703125,
      "learning_rate": 4.995780593567028e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 6472
    },
    {
      "completion_length": 252.88394260406494,
      "epoch": 1.0855023261662267,
      "grad_norm": 0.24414306302956112,
      "kl": 0.0724029541015625,
      "learning_rate": 4.995773624188328e-07,
      "loss": 0.0001,
      "reward": 1.7267858013510704,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7312500216066837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6474
    },
    {
      "completion_length": 255.62054538726807,
      "epoch": 1.085837629406094,
      "grad_norm": 0.250032728392339,
      "kl": 0.0707550048828125,
      "learning_rate": 4.995766649063429e-07,
      "loss": 0.0001,
      "reward": 1.7964286133646965,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 6476
    },
    {
      "completion_length": 256.508939743042,
      "epoch": 1.0861729326459617,
      "grad_norm": 0.1228238932830406,
      "kl": 0.0816497802734375,
      "learning_rate": 4.995759668192352e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143107801676,
      "rewards/format_reward_func": 1.0,
      "step": 6478
    },
    {
      "completion_length": 253.83483409881592,
      "epoch": 1.0865082358858293,
      "grad_norm": 0.2071801861333888,
      "kl": 0.0679779052734375,
      "learning_rate": 4.995752681575111e-07,
      "loss": 0.0001,
      "reward": 1.8160714730620384,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8205357268452644,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6480
    },
    {
      "completion_length": 241.84822368621826,
      "epoch": 1.0868435391256968,
      "grad_norm": 0.17132538782288873,
      "kl": 0.0664825439453125,
      "learning_rate": 4.995745689211721e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000134110451,
      "rewards/format_reward_func": 1.0,
      "step": 6482
    },
    {
      "completion_length": 251.61608123779297,
      "epoch": 1.0871788423655644,
      "grad_norm": 0.33335513174173054,
      "kl": 0.077301025390625,
      "learning_rate": 4.9957386911022e-07,
      "loss": 0.0001,
      "reward": 1.7107143849134445,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7196428813040257,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6484
    },
    {
      "completion_length": 249.34375953674316,
      "epoch": 1.0875141456054318,
      "grad_norm": 0.1509868126855817,
      "kl": 0.0727386474609375,
      "learning_rate": 4.995731687246562e-07,
      "loss": 0.0001,
      "reward": 1.814285770058632,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857439815998,
      "rewards/format_reward_func": 1.0,
      "step": 6486
    },
    {
      "completion_length": 263.23662090301514,
      "epoch": 1.0878494488452994,
      "grad_norm": 0.14210317501246147,
      "kl": 0.07989501953125,
      "learning_rate": 4.995724677644824e-07,
      "loss": 0.0001,
      "reward": 1.7107143625617027,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7196428924798965,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6488
    },
    {
      "completion_length": 258.1696548461914,
      "epoch": 1.088184752085167,
      "grad_norm": 0.22951210067149835,
      "kl": 0.075286865234375,
      "learning_rate": 4.995717662297003e-07,
      "loss": 0.0001,
      "reward": 1.7178572565317154,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7178571857511997,
      "rewards/format_reward_func": 1.0,
      "step": 6490
    },
    {
      "completion_length": 256.8303689956665,
      "epoch": 1.0885200553250345,
      "grad_norm": 0.21521536442832762,
      "kl": 0.0674591064453125,
      "learning_rate": 4.995710641203115e-07,
      "loss": 0.0001,
      "reward": 1.7410714849829674,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7455357424914837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6492
    },
    {
      "completion_length": 252.008939743042,
      "epoch": 1.0888553585649021,
      "grad_norm": 0.17696140360889953,
      "kl": 0.073944091796875,
      "learning_rate": 4.995703614363176e-07,
      "loss": 0.0001,
      "reward": 1.7410714998841286,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357406288385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6494
    },
    {
      "completion_length": 276.0759048461914,
      "epoch": 1.0891906618047698,
      "grad_norm": 0.29492164732292375,
      "kl": 0.0858001708984375,
      "learning_rate": 4.995696581777201e-07,
      "loss": 0.0001,
      "reward": 1.7196429371833801,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7241071723401546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6496
    },
    {
      "completion_length": 265.07144260406494,
      "epoch": 1.0895259650446372,
      "grad_norm": 0.20700246572065972,
      "kl": 0.098724365234375,
      "learning_rate": 4.995689543445209e-07,
      "loss": 0.0001,
      "reward": 1.7785714641213417,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7875000238418579,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6498
    },
    {
      "completion_length": 267.6875104904175,
      "epoch": 1.0898612682845048,
      "grad_norm": 0.1563277942876201,
      "kl": 0.0779876708984375,
      "learning_rate": 4.995682499367212e-07,
      "loss": 0.0001,
      "reward": 1.7892857491970062,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.807142885401845,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6500
    },
    {
      "completion_length": 260.4017972946167,
      "epoch": 1.0901965715243724,
      "grad_norm": 0.16573974603677033,
      "kl": 0.0820465087890625,
      "learning_rate": 4.99567544954323e-07,
      "loss": 0.0001,
      "reward": 1.812500037252903,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8169643133878708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6502
    },
    {
      "completion_length": 270.37054920196533,
      "epoch": 1.0905318747642399,
      "grad_norm": 0.20563901893599482,
      "kl": 0.08184814453125,
      "learning_rate": 4.995668393973277e-07,
      "loss": 0.0001,
      "reward": 1.7089286521077156,
      "reward_std": 0.10859140008687973,
      "rewards/equation_reward_func": 0.7401786111295223,
      "rewards/format_reward_func": 0.9687500111758709,
      "step": 6504
    },
    {
      "completion_length": 276.2589387893677,
      "epoch": 1.0908671780041075,
      "grad_norm": 0.7781279067739908,
      "kl": 0.1058349609375,
      "learning_rate": 4.99566133265737e-07,
      "loss": 0.0001,
      "reward": 1.712500050663948,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7348214518278837,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 6506
    },
    {
      "completion_length": 284.8482255935669,
      "epoch": 1.0912024812439751,
      "grad_norm": 0.30134766414761066,
      "kl": 0.1042633056640625,
      "learning_rate": 4.995654265595526e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7767857424914837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6508
    },
    {
      "completion_length": 289.0535879135132,
      "epoch": 1.0915377844838425,
      "grad_norm": 0.25376529695446304,
      "kl": 0.114471435546875,
      "learning_rate": 4.99564719278776e-07,
      "loss": 0.0001,
      "reward": 1.7160714715719223,
      "reward_std": 0.07828682288527489,
      "rewards/equation_reward_func": 0.738392885774374,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 6510
    },
    {
      "completion_length": 284.9821557998657,
      "epoch": 1.0918730877237102,
      "grad_norm": 0.42895058427759786,
      "kl": 0.13214111328125,
      "learning_rate": 4.99564011423409e-07,
      "loss": 0.0001,
      "reward": 1.6714286506175995,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.6892857495695353,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6512
    },
    {
      "completion_length": 283.6562614440918,
      "epoch": 1.0922083909635778,
      "grad_norm": 0.6980135663023137,
      "kl": 0.156036376953125,
      "learning_rate": 4.99563302993453e-07,
      "loss": 0.0002,
      "reward": 1.6982143893837929,
      "reward_std": 0.12374368775635958,
      "rewards/equation_reward_func": 0.7294643074274063,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 6514
    },
    {
      "completion_length": 284.52233600616455,
      "epoch": 1.0925436942034452,
      "grad_norm": 0.44354741258299424,
      "kl": 0.1544189453125,
      "learning_rate": 4.995625939889097e-07,
      "loss": 0.0002,
      "reward": 1.7008929252624512,
      "reward_std": 0.059346460737288,
      "rewards/equation_reward_func": 0.7116071954369545,
      "rewards/format_reward_func": 0.9892857223749161,
      "step": 6516
    },
    {
      "completion_length": 278.3214406967163,
      "epoch": 1.0928789974433128,
      "grad_norm": 0.3003540847071273,
      "kl": 0.144378662109375,
      "learning_rate": 4.995618844097808e-07,
      "loss": 0.0001,
      "reward": 1.7767857909202576,
      "reward_std": 0.10354063473641872,
      "rewards/equation_reward_func": 0.7901785857975483,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6518
    },
    {
      "completion_length": 300.0803680419922,
      "epoch": 1.0932143006831803,
      "grad_norm": 0.6405706414392106,
      "kl": 0.28515625,
      "learning_rate": 4.99561174256068e-07,
      "loss": 0.0003,
      "reward": 1.687500074505806,
      "reward_std": 0.08838834799826145,
      "rewards/equation_reward_func": 0.7098214626312256,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 6520
    },
    {
      "completion_length": 290.151798248291,
      "epoch": 1.093549603923048,
      "grad_norm": 0.21560716139636138,
      "kl": 0.2230987548828125,
      "learning_rate": 4.995604635277728e-07,
      "loss": 0.0002,
      "reward": 1.707142949104309,
      "reward_std": 0.11616754159331322,
      "rewards/equation_reward_func": 0.7428571730852127,
      "rewards/format_reward_func": 0.9642857313156128,
      "step": 6522
    },
    {
      "completion_length": 288.7901945114136,
      "epoch": 1.0938849071629155,
      "grad_norm": 0.6662672959830765,
      "kl": 0.8878631591796875,
      "learning_rate": 4.995597522248968e-07,
      "loss": 0.0009,
      "reward": 1.7214286401867867,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7392857372760773,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6524
    },
    {
      "completion_length": 289.4553689956665,
      "epoch": 1.094220210402783,
      "grad_norm": 0.1884437199547336,
      "kl": 0.4522247314453125,
      "learning_rate": 4.995590403474419e-07,
      "loss": 0.0005,
      "reward": 1.707142896950245,
      "reward_std": 0.08081220462918282,
      "rewards/equation_reward_func": 0.7339286208152771,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 6526
    },
    {
      "completion_length": 268.06251335144043,
      "epoch": 1.0945555136426506,
      "grad_norm": 0.29136324362216554,
      "kl": 0.8277587890625,
      "learning_rate": 4.995583278954095e-07,
      "loss": 0.0008,
      "reward": 1.782142922282219,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7910714484751225,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6528
    },
    {
      "completion_length": 282.5134105682373,
      "epoch": 1.0948908168825182,
      "grad_norm": 0.13064029811812,
      "kl": 1.827606201171875,
      "learning_rate": 4.995576148688012e-07,
      "loss": 0.0018,
      "reward": 1.771428607404232,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7982143238186836,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 6530
    },
    {
      "completion_length": 274.4553699493408,
      "epoch": 1.0952261201223856,
      "grad_norm": 0.2776119596095246,
      "kl": 0.695587158203125,
      "learning_rate": 4.995569012676189e-07,
      "loss": 0.0007,
      "reward": 1.7375000640749931,
      "reward_std": 0.08838834799826145,
      "rewards/equation_reward_func": 0.7508928775787354,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6532
    },
    {
      "completion_length": 283.30358695983887,
      "epoch": 1.0955614233622533,
      "grad_norm": 0.2657775533041802,
      "kl": 0.5340576171875,
      "learning_rate": 4.995561870918639e-07,
      "loss": 0.0005,
      "reward": 1.7482143715023994,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7526785992085934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6534
    },
    {
      "completion_length": 290.11162281036377,
      "epoch": 1.095896726602121,
      "grad_norm": 0.3284617239203021,
      "kl": 0.455780029296875,
      "learning_rate": 4.995554723415382e-07,
      "loss": 0.0005,
      "reward": 1.7071429342031479,
      "reward_std": 0.10606601741164923,
      "rewards/equation_reward_func": 0.733928594738245,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 6536
    },
    {
      "completion_length": 285.56251525878906,
      "epoch": 1.0962320298419883,
      "grad_norm": 0.4737519246446728,
      "kl": 0.477142333984375,
      "learning_rate": 4.995547570166432e-07,
      "loss": 0.0005,
      "reward": 1.7482143715023994,
      "reward_std": 0.10859139822423458,
      "rewards/equation_reward_func": 0.779464315623045,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 6538
    },
    {
      "completion_length": 281.23662281036377,
      "epoch": 1.096567333081856,
      "grad_norm": 0.28880568251512756,
      "kl": 0.977142333984375,
      "learning_rate": 4.995540411171805e-07,
      "loss": 0.001,
      "reward": 1.767857201397419,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.776785746216774,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6540
    },
    {
      "completion_length": 289.55358505249023,
      "epoch": 1.0969026363217234,
      "grad_norm": 0.022334262039546134,
      "kl": 0.261566162109375,
      "learning_rate": 4.99553324643152e-07,
      "loss": 0.0003,
      "reward": 1.757142923772335,
      "reward_std": 0.04040610231459141,
      "rewards/equation_reward_func": 0.7660714536905289,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6542
    },
    {
      "completion_length": 285.9241199493408,
      "epoch": 1.097237939561591,
      "grad_norm": 0.30625968284332616,
      "kl": 0.1597900390625,
      "learning_rate": 4.995526075945593e-07,
      "loss": 0.0002,
      "reward": 1.7553571835160255,
      "reward_std": 0.09343910869210958,
      "rewards/equation_reward_func": 0.7776786051690578,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 6544
    },
    {
      "completion_length": 279.4241199493408,
      "epoch": 1.0975732428014586,
      "grad_norm": 0.4419334505325685,
      "kl": 0.6954498291015625,
      "learning_rate": 4.995518899714039e-07,
      "loss": 0.0007,
      "reward": 1.685714341700077,
      "reward_std": 0.07071067858487368,
      "rewards/equation_reward_func": 0.7035714499652386,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6546
    },
    {
      "completion_length": 286.0134086608887,
      "epoch": 1.097908546041326,
      "grad_norm": 0.33027778382925066,
      "kl": 0.388275146484375,
      "learning_rate": 4.995511717736875e-07,
      "loss": 0.0004,
      "reward": 1.73214291036129,
      "reward_std": 0.11616754066199064,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6548
    },
    {
      "completion_length": 283.43750953674316,
      "epoch": 1.0982438492811937,
      "grad_norm": 0.21767882315718848,
      "kl": 0.23004150390625,
      "learning_rate": 4.995504530014117e-07,
      "loss": 0.0002,
      "reward": 1.7500000670552254,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7589286006987095,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6550
    },
    {
      "completion_length": 281.70983123779297,
      "epoch": 1.0985791525210613,
      "grad_norm": 0.3029158393739477,
      "kl": 0.1859130859375,
      "learning_rate": 4.995497336545783e-07,
      "loss": 0.0002,
      "reward": 1.7410715073347092,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357611179352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6552
    },
    {
      "completion_length": 270.4107246398926,
      "epoch": 1.0989144557609287,
      "grad_norm": 0.37787958558581103,
      "kl": 0.186492919921875,
      "learning_rate": 4.995490137331889e-07,
      "loss": 0.0002,
      "reward": 1.7946429178118706,
      "reward_std": 0.08838834706693888,
      "rewards/equation_reward_func": 0.8080357275903225,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6554
    },
    {
      "completion_length": 288.4419765472412,
      "epoch": 1.0992497590007964,
      "grad_norm": 0.26048348704828717,
      "kl": 0.204254150390625,
      "learning_rate": 4.995482932372451e-07,
      "loss": 0.0002,
      "reward": 1.698214367032051,
      "reward_std": 0.12879444938153028,
      "rewards/equation_reward_func": 0.7294643111526966,
      "rewards/format_reward_func": 0.9687500111758709,
      "step": 6556
    },
    {
      "completion_length": 261.23662090301514,
      "epoch": 1.099585062240664,
      "grad_norm": 0.551212163479299,
      "kl": 0.3712005615234375,
      "learning_rate": 4.995475721667486e-07,
      "loss": 0.0004,
      "reward": 1.7089286148548126,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7223214562982321,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6558
    },
    {
      "completion_length": 249.2544755935669,
      "epoch": 1.0999203654805314,
      "grad_norm": 0.20253441295934366,
      "kl": 0.2565765380859375,
      "learning_rate": 4.995468505217011e-07,
      "loss": 0.0003,
      "reward": 1.769642911851406,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7741071656346321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6560
    },
    {
      "completion_length": 258.9598340988159,
      "epoch": 1.100255668720399,
      "grad_norm": 0.21418411741040597,
      "kl": 0.1841278076171875,
      "learning_rate": 4.995461283021043e-07,
      "loss": 0.0002,
      "reward": 1.6607143729925156,
      "reward_std": 0.0656599160283804,
      "rewards/equation_reward_func": 0.6785714607685804,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6562
    },
    {
      "completion_length": 253.91519165039062,
      "epoch": 1.1005909719602665,
      "grad_norm": 0.2070990208135219,
      "kl": 0.144287109375,
      "learning_rate": 4.995454055079597e-07,
      "loss": 0.0001,
      "reward": 1.721428632736206,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7214286178350449,
      "rewards/format_reward_func": 1.0,
      "step": 6564
    },
    {
      "completion_length": 246.5803689956665,
      "epoch": 1.100926275200134,
      "grad_norm": 0.13879373930903302,
      "kl": 0.237457275390625,
      "learning_rate": 4.995446821392691e-07,
      "loss": 0.0002,
      "reward": 1.7857143431901932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.785714328289032,
      "rewards/format_reward_func": 1.0,
      "step": 6566
    },
    {
      "completion_length": 242.7142972946167,
      "epoch": 1.1012615784400017,
      "grad_norm": 0.26167857128278654,
      "kl": 0.146392822265625,
      "learning_rate": 4.99543958196034e-07,
      "loss": 0.0001,
      "reward": 1.7285715118050575,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.728571455925703,
      "rewards/format_reward_func": 1.0,
      "step": 6568
    },
    {
      "completion_length": 248.7500123977661,
      "epoch": 1.1015968816798691,
      "grad_norm": 0.15531658303917262,
      "kl": 0.11407470703125,
      "learning_rate": 4.995432336782563e-07,
      "loss": 0.0001,
      "reward": 1.7660714983940125,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6570
    },
    {
      "completion_length": 235.06251049041748,
      "epoch": 1.1019321849197368,
      "grad_norm": 0.8186170595241872,
      "kl": 0.1658935546875,
      "learning_rate": 4.995425085859375e-07,
      "loss": 0.0002,
      "reward": 1.8142857775092125,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8142857439815998,
      "rewards/format_reward_func": 1.0,
      "step": 6572
    },
    {
      "completion_length": 239.82590293884277,
      "epoch": 1.1022674881596044,
      "grad_norm": 0.06946096262257632,
      "kl": 0.1508331298828125,
      "learning_rate": 4.995417829190793e-07,
      "loss": 0.0002,
      "reward": 1.8464286103844643,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8464285880327225,
      "rewards/format_reward_func": 1.0,
      "step": 6574
    },
    {
      "completion_length": 241.59375953674316,
      "epoch": 1.1026027913994718,
      "grad_norm": 0.3313378435357141,
      "kl": 0.213287353515625,
      "learning_rate": 4.995410566776835e-07,
      "loss": 0.0002,
      "reward": 1.703571505844593,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7035714667290449,
      "rewards/format_reward_func": 1.0,
      "step": 6576
    },
    {
      "completion_length": 233.88394165039062,
      "epoch": 1.1029380946393395,
      "grad_norm": 0.16843392243853206,
      "kl": 0.14697265625,
      "learning_rate": 4.995403298617516e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.767857164144516,
      "rewards/format_reward_func": 1.0,
      "step": 6578
    },
    {
      "completion_length": 239.76786994934082,
      "epoch": 1.103273397879207,
      "grad_norm": 0.3046427178111775,
      "kl": 0.140228271484375,
      "learning_rate": 4.995396024712853e-07,
      "loss": 0.0001,
      "reward": 1.7250000834465027,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7250000331550837,
      "rewards/format_reward_func": 1.0,
      "step": 6580
    },
    {
      "completion_length": 243.0803680419922,
      "epoch": 1.1036087011190745,
      "grad_norm": 0.21699272384070514,
      "kl": 0.19512939453125,
      "learning_rate": 4.995388745062864e-07,
      "loss": 0.0002,
      "reward": 1.7678572162985802,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571604192257,
      "rewards/format_reward_func": 1.0,
      "step": 6582
    },
    {
      "completion_length": 245.22769165039062,
      "epoch": 1.1039440043589421,
      "grad_norm": 0.1718854405192041,
      "kl": 0.1443328857421875,
      "learning_rate": 4.995381459667564e-07,
      "loss": 0.0001,
      "reward": 1.7428572326898575,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571730852127,
      "rewards/format_reward_func": 1.0,
      "step": 6584
    },
    {
      "completion_length": 234.7098331451416,
      "epoch": 1.1042793075988098,
      "grad_norm": 0.22912566502299336,
      "kl": 0.168731689453125,
      "learning_rate": 4.995374168526971e-07,
      "loss": 0.0002,
      "reward": 1.7571429163217545,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 6586
    },
    {
      "completion_length": 235.71429634094238,
      "epoch": 1.1046146108386772,
      "grad_norm": 0.1733917636861512,
      "kl": 0.141448974609375,
      "learning_rate": 4.995366871641101e-07,
      "loss": 0.0001,
      "reward": 1.7982143312692642,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8026785962283611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6588
    },
    {
      "completion_length": 242.3392972946167,
      "epoch": 1.1049499140785448,
      "grad_norm": 0.3394077171058364,
      "kl": 0.211273193359375,
      "learning_rate": 4.995359569009972e-07,
      "loss": 0.0002,
      "reward": 1.7125000953674316,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7169643230736256,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6590
    },
    {
      "completion_length": 252.6875123977661,
      "epoch": 1.1052852173184124,
      "grad_norm": 0.43081293111981867,
      "kl": 0.11590576171875,
      "learning_rate": 4.995352260633601e-07,
      "loss": 0.0001,
      "reward": 1.7285715192556381,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7285714484751225,
      "rewards/format_reward_func": 1.0,
      "step": 6592
    },
    {
      "completion_length": 258.4330472946167,
      "epoch": 1.1056205205582799,
      "grad_norm": 0.21628723638974706,
      "kl": 0.13763427734375,
      "learning_rate": 4.995344946512002e-07,
      "loss": 0.0001,
      "reward": 1.7392858043313026,
      "reward_std": 0.05555839091539383,
      "rewards/equation_reward_func": 0.7482143007218838,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6594
    },
    {
      "completion_length": 248.11608123779297,
      "epoch": 1.1059558237981475,
      "grad_norm": 0.5966538770183953,
      "kl": 0.261688232421875,
      "learning_rate": 4.995337626645195e-07,
      "loss": 0.0003,
      "reward": 1.7875000685453415,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.791964303702116,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6596
    },
    {
      "completion_length": 245.98215103149414,
      "epoch": 1.106291127038015,
      "grad_norm": 0.157717130031815,
      "kl": 0.097625732421875,
      "learning_rate": 4.995330301033195e-07,
      "loss": 0.0001,
      "reward": 1.7625000700354576,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7669643107801676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6598
    },
    {
      "completion_length": 248.59376335144043,
      "epoch": 1.1066264302778825,
      "grad_norm": 0.13235538653295403,
      "kl": 0.130096435546875,
      "learning_rate": 4.995322969676019e-07,
      "loss": 0.0001,
      "reward": 1.719642922282219,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7241071723401546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6600
    },
    {
      "completion_length": 255.5089406967163,
      "epoch": 1.1069617335177502,
      "grad_norm": 0.3280297518641213,
      "kl": 0.145782470703125,
      "learning_rate": 4.995315632573684e-07,
      "loss": 0.0001,
      "reward": 1.6982143893837929,
      "reward_std": 0.11364216171205044,
      "rewards/equation_reward_func": 0.7205357421189547,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 6602
    },
    {
      "completion_length": 270.5000104904175,
      "epoch": 1.1072970367576176,
      "grad_norm": 0.9859893448344605,
      "kl": 0.204925537109375,
      "learning_rate": 4.995308289726209e-07,
      "loss": 0.0002,
      "reward": 1.7214286476373672,
      "reward_std": 0.09091373067349195,
      "rewards/equation_reward_func": 0.7392857521772385,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6604
    },
    {
      "completion_length": 254.0848331451416,
      "epoch": 1.1076323399974852,
      "grad_norm": 0.6059061751603121,
      "kl": 0.307647705078125,
      "learning_rate": 4.995300941133608e-07,
      "loss": 0.0003,
      "reward": 1.7660714983940125,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7705357372760773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6606
    },
    {
      "completion_length": 274.86608600616455,
      "epoch": 1.1079676432373529,
      "grad_norm": 0.2292386723821532,
      "kl": 0.272247314453125,
      "learning_rate": 4.995293586795899e-07,
      "loss": 0.0003,
      "reward": 1.7553572282195091,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.759821455925703,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6608
    },
    {
      "completion_length": 261.2946557998657,
      "epoch": 1.1083029464772203,
      "grad_norm": 0.2272666943279464,
      "kl": 0.20709228515625,
      "learning_rate": 4.995286226713099e-07,
      "loss": 0.0002,
      "reward": 1.7071429565548897,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7160714566707611,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6610
    },
    {
      "completion_length": 253.86608123779297,
      "epoch": 1.108638249717088,
      "grad_norm": 0.4846628000520213,
      "kl": 0.3332977294921875,
      "learning_rate": 4.995278860885225e-07,
      "loss": 0.0003,
      "reward": 1.6946429386734962,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.6991071794182062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6612
    },
    {
      "completion_length": 254.2321538925171,
      "epoch": 1.1089735529569555,
      "grad_norm": 0.16874754553226004,
      "kl": 0.1689453125,
      "learning_rate": 4.995271489312294e-07,
      "loss": 0.0002,
      "reward": 1.7933036237955093,
      "reward_std": 0.05997780850157142,
      "rewards/equation_reward_func": 0.808035746216774,
      "rewards/format_reward_func": 0.9852678664028645,
      "step": 6614
    },
    {
      "completion_length": 246.46429538726807,
      "epoch": 1.109308856196823,
      "grad_norm": 0.2300480690557061,
      "kl": 0.7288818359375,
      "learning_rate": 4.995264111994322e-07,
      "loss": 0.0007,
      "reward": 1.7285714969038963,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7375000342726707,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6616
    },
    {
      "completion_length": 255.94197463989258,
      "epoch": 1.1096441594366906,
      "grad_norm": 0.2096849116943943,
      "kl": 1.5742645263671875,
      "learning_rate": 4.995256728931328e-07,
      "loss": 0.0016,
      "reward": 1.7254464998841286,
      "reward_std": 0.054927044780924916,
      "rewards/equation_reward_func": 0.7401785962283611,
      "rewards/format_reward_func": 0.9852678664028645,
      "step": 6618
    },
    {
      "completion_length": 253.44197273254395,
      "epoch": 1.109979462676558,
      "grad_norm": 0.24599453467498117,
      "kl": 0.540283203125,
      "learning_rate": 4.995249340123328e-07,
      "loss": 0.0005,
      "reward": 1.801785759627819,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8062500283122063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6620
    },
    {
      "completion_length": 254.8750114440918,
      "epoch": 1.1103147659164256,
      "grad_norm": 0.3624082535916845,
      "kl": 0.9485015869140625,
      "learning_rate": 4.995241945570339e-07,
      "loss": 0.001,
      "reward": 1.7517857775092125,
      "reward_std": 0.1287944484502077,
      "rewards/equation_reward_func": 0.7830357439815998,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 6622
    },
    {
      "completion_length": 249.16072463989258,
      "epoch": 1.1106500691562933,
      "grad_norm": 0.22190804563509325,
      "kl": 0.233489990234375,
      "learning_rate": 4.995234545272377e-07,
      "loss": 0.0002,
      "reward": 1.7892858013510704,
      "reward_std": 0.08586296439170837,
      "rewards/equation_reward_func": 0.7982143126428127,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6624
    },
    {
      "completion_length": 266.75894355773926,
      "epoch": 1.1109853723961607,
      "grad_norm": 0.2184835683267264,
      "kl": 0.2601318359375,
      "learning_rate": 4.99522713922946e-07,
      "loss": 0.0003,
      "reward": 1.6803572326898575,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.684821480885148,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6626
    },
    {
      "completion_length": 258.70536518096924,
      "epoch": 1.1113206756360283,
      "grad_norm": 0.23099437396258055,
      "kl": 0.1270751953125,
      "learning_rate": 4.995219727441605e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.07576144114136696,
      "rewards/equation_reward_func": 0.7910714522004128,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6628
    },
    {
      "completion_length": 249.26340579986572,
      "epoch": 1.111655978875896,
      "grad_norm": 0.15913812225492693,
      "kl": 0.1038665771484375,
      "learning_rate": 4.995212309908829e-07,
      "loss": 0.0001,
      "reward": 1.7321429029107094,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7500000242143869,
      "rewards/format_reward_func": 0.9821428619325161,
      "step": 6630
    },
    {
      "completion_length": 255.02233505249023,
      "epoch": 1.1119912821157634,
      "grad_norm": 0.1314757051272552,
      "kl": 0.1993408203125,
      "learning_rate": 4.995204886631149e-07,
      "loss": 0.0002,
      "reward": 1.753571517765522,
      "reward_std": 0.05555839091539383,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6632
    },
    {
      "completion_length": 251.0848331451416,
      "epoch": 1.112326585355631,
      "grad_norm": 0.32806559142892566,
      "kl": 0.133941650390625,
      "learning_rate": 4.995197457608582e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.10101525485515594,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6634
    },
    {
      "completion_length": 255.01787281036377,
      "epoch": 1.1126618885954986,
      "grad_norm": 0.16793208241146898,
      "kl": 0.108489990234375,
      "learning_rate": 4.995190022841146e-07,
      "loss": 0.0001,
      "reward": 1.7500000447034836,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7589286044239998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6636
    },
    {
      "completion_length": 262.2142972946167,
      "epoch": 1.112997191835366,
      "grad_norm": 0.08044673225781292,
      "kl": 0.131439208984375,
      "learning_rate": 4.995182582328858e-07,
      "loss": 0.0001,
      "reward": 1.7125000804662704,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7169643100351095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6638
    },
    {
      "completion_length": 256.00001335144043,
      "epoch": 1.1133324950752337,
      "grad_norm": 0.20721738332140918,
      "kl": 0.11676025390625,
      "learning_rate": 4.995175136071733e-07,
      "loss": 0.0001,
      "reward": 1.7482143491506577,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.752678606659174,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6640
    },
    {
      "completion_length": 246.9419765472412,
      "epoch": 1.1136677983151013,
      "grad_norm": 0.24787722237396217,
      "kl": 0.095733642578125,
      "learning_rate": 4.99516768406979e-07,
      "loss": 0.0001,
      "reward": 1.7982143387198448,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8026786036789417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6642
    },
    {
      "completion_length": 241.3437614440918,
      "epoch": 1.1140031015549687,
      "grad_norm": 0.4045079891732805,
      "kl": 0.08056640625,
      "learning_rate": 4.995160226323046e-07,
      "loss": 0.0001,
      "reward": 1.7803572192788124,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7848214581608772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6644
    },
    {
      "completion_length": 245.88840198516846,
      "epoch": 1.1143384047948364,
      "grad_norm": 0.28886896563636494,
      "kl": 0.080047607421875,
      "learning_rate": 4.99515276283152e-07,
      "loss": 0.0001,
      "reward": 1.7580358013510704,
      "reward_std": 0.08965103607624769,
      "rewards/equation_reward_func": 0.7598214596509933,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 6646
    },
    {
      "completion_length": 260.1384038925171,
      "epoch": 1.114673708034704,
      "grad_norm": 0.18604455411456783,
      "kl": 0.103424072265625,
      "learning_rate": 4.995145293595224e-07,
      "loss": 0.0001,
      "reward": 1.7375000938773155,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7419643178582191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6648
    },
    {
      "completion_length": 244.8169765472412,
      "epoch": 1.1150090112745714,
      "grad_norm": 0.13081924970963807,
      "kl": 0.073028564453125,
      "learning_rate": 4.995137818614181e-07,
      "loss": 0.0001,
      "reward": 1.8642857670783997,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8642857372760773,
      "rewards/format_reward_func": 1.0,
      "step": 6650
    },
    {
      "completion_length": 256.6294765472412,
      "epoch": 1.115344314514439,
      "grad_norm": 0.2613255092581983,
      "kl": 0.088592529296875,
      "learning_rate": 4.995130337888404e-07,
      "loss": 0.0001,
      "reward": 1.7678571864962578,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.776785746216774,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6652
    },
    {
      "completion_length": 253.9241189956665,
      "epoch": 1.1156796177543065,
      "grad_norm": 0.29768289237812334,
      "kl": 0.07568359375,
      "learning_rate": 4.995122851417912e-07,
      "loss": 0.0001,
      "reward": 1.708928681910038,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7223214525729418,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6654
    },
    {
      "completion_length": 242.65625858306885,
      "epoch": 1.116014920994174,
      "grad_norm": 0.2506214712263386,
      "kl": 0.1123504638671875,
      "learning_rate": 4.995115359202722e-07,
      "loss": 0.0001,
      "reward": 1.7214286476373672,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.72142861969769,
      "rewards/format_reward_func": 1.0,
      "step": 6656
    },
    {
      "completion_length": 253.31251049041748,
      "epoch": 1.1163502242340417,
      "grad_norm": 0.2593847018803885,
      "kl": 0.1129302978515625,
      "learning_rate": 4.995107861242852e-07,
      "loss": 0.0001,
      "reward": 1.7660714983940125,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7705357447266579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6658
    },
    {
      "completion_length": 255.15179824829102,
      "epoch": 1.1166855274739091,
      "grad_norm": 0.27640892602323935,
      "kl": 0.07427978515625,
      "learning_rate": 4.995100357538319e-07,
      "loss": 0.0001,
      "reward": 1.7214286625385284,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7214285992085934,
      "rewards/format_reward_func": 1.0,
      "step": 6660
    },
    {
      "completion_length": 258.34376335144043,
      "epoch": 1.1170208307137768,
      "grad_norm": 0.2795999240184237,
      "kl": 0.152252197265625,
      "learning_rate": 4.995092848089139e-07,
      "loss": 0.0002,
      "reward": 1.7357143759727478,
      "reward_std": 0.1010152529925108,
      "rewards/equation_reward_func": 0.7446428909897804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6662
    },
    {
      "completion_length": 248.71876430511475,
      "epoch": 1.1173561339536444,
      "grad_norm": 0.2088474635895226,
      "kl": 0.067962646484375,
      "learning_rate": 4.99508533289533e-07,
      "loss": 0.0001,
      "reward": 1.6928572282195091,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7017857655882835,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6664
    },
    {
      "completion_length": 260.60715675354004,
      "epoch": 1.1176914371935118,
      "grad_norm": 0.23495409011200794,
      "kl": 0.107940673828125,
      "learning_rate": 4.99507781195691e-07,
      "loss": 0.0001,
      "reward": 1.7410714775323868,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7544643208384514,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6666
    },
    {
      "completion_length": 258.20537185668945,
      "epoch": 1.1180267404333795,
      "grad_norm": 0.3012901568012722,
      "kl": 0.18408203125,
      "learning_rate": 4.995070285273895e-07,
      "loss": 0.0002,
      "reward": 1.7660714834928513,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.779464315623045,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6668
    },
    {
      "completion_length": 267.14287090301514,
      "epoch": 1.118362043673247,
      "grad_norm": 0.12807288181741017,
      "kl": 0.08056640625,
      "learning_rate": 4.995062752846304e-07,
      "loss": 0.0001,
      "reward": 1.7303572297096252,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.743750024586916,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6670
    },
    {
      "completion_length": 254.95983219146729,
      "epoch": 1.1186973469131145,
      "grad_norm": 0.27471962269330896,
      "kl": 0.365966796875,
      "learning_rate": 4.995055214674153e-07,
      "loss": 0.0004,
      "reward": 1.7339286282658577,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7473214566707611,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6672
    },
    {
      "completion_length": 245.0714406967163,
      "epoch": 1.1190326501529821,
      "grad_norm": 0.445086464766825,
      "kl": 0.2374267578125,
      "learning_rate": 4.99504767075746e-07,
      "loss": 0.0002,
      "reward": 1.7821429297327995,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428701281548,
      "rewards/format_reward_func": 1.0,
      "step": 6674
    },
    {
      "completion_length": 258.39286613464355,
      "epoch": 1.1193679533928496,
      "grad_norm": 0.41685696367688935,
      "kl": 0.120513916015625,
      "learning_rate": 4.995040121096243e-07,
      "loss": 0.0001,
      "reward": 1.7589286342263222,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7633928917348385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6676
    },
    {
      "completion_length": 242.99554538726807,
      "epoch": 1.1197032566327172,
      "grad_norm": 0.40514511717107204,
      "kl": 0.22406005859375,
      "learning_rate": 4.995032565690517e-07,
      "loss": 0.0002,
      "reward": 1.7160715088248253,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7205357495695353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6678
    },
    {
      "completion_length": 240.9821538925171,
      "epoch": 1.1200385598725848,
      "grad_norm": 0.46798781534682443,
      "kl": 0.207855224609375,
      "learning_rate": 4.995025004540301e-07,
      "loss": 0.0002,
      "reward": 1.7892857640981674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857491970062,
      "rewards/format_reward_func": 1.0,
      "step": 6680
    },
    {
      "completion_length": 256.11161708831787,
      "epoch": 1.1203738631124522,
      "grad_norm": 0.6230530844540287,
      "kl": 0.28582763671875,
      "learning_rate": 4.995017437645613e-07,
      "loss": 0.0003,
      "reward": 1.770089365541935,
      "reward_std": 0.09280776605010033,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 0.9808035790920258,
      "step": 6682
    },
    {
      "completion_length": 236.52679634094238,
      "epoch": 1.1207091663523199,
      "grad_norm": 2.168458827730386,
      "kl": 0.2948150634765625,
      "learning_rate": 4.99500986500647e-07,
      "loss": 0.0003,
      "reward": 1.7080358043313026,
      "reward_std": 0.04924493608996272,
      "rewards/equation_reward_func": 0.7107143364846706,
      "rewards/format_reward_func": 0.9973214343190193,
      "step": 6684
    },
    {
      "completion_length": 235.4553680419922,
      "epoch": 1.1210444695921875,
      "grad_norm": 0.5230753506181978,
      "kl": 0.3092041015625,
      "learning_rate": 4.995002286622888e-07,
      "loss": 0.0003,
      "reward": 1.7428572252392769,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7517857402563095,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6686
    },
    {
      "completion_length": 263.477689743042,
      "epoch": 1.121379772832055,
      "grad_norm": 2.2929950926167426,
      "kl": 1.7777099609375,
      "learning_rate": 4.994994702494887e-07,
      "loss": 0.0018,
      "reward": 1.7089285999536514,
      "reward_std": 0.1439467379823327,
      "rewards/equation_reward_func": 0.7580357454717159,
      "rewards/format_reward_func": 0.9508928805589676,
      "step": 6688
    },
    {
      "completion_length": 332.70983505249023,
      "epoch": 1.1217150760719226,
      "grad_norm": 2.489929363328956,
      "kl": 2.840087890625,
      "learning_rate": 4.994987112622483e-07,
      "loss": 0.0028,
      "reward": 1.1937500424683094,
      "reward_std": 0.4482551934197545,
      "rewards/equation_reward_func": 0.535714304074645,
      "rewards/format_reward_func": 0.6580357477068901,
      "step": 6690
    },
    {
      "completion_length": 297.50001430511475,
      "epoch": 1.1220503793117902,
      "grad_norm": 2.5227119840513015,
      "kl": 1.386444091796875,
      "learning_rate": 4.994979517005693e-07,
      "loss": 0.0014,
      "reward": 1.365625061094761,
      "reward_std": 0.3276682370342314,
      "rewards/equation_reward_func": 0.6062500178813934,
      "rewards/format_reward_func": 0.7593750320374966,
      "step": 6692
    },
    {
      "completion_length": 292.82590770721436,
      "epoch": 1.1223856825516576,
      "grad_norm": 1.663920193770367,
      "kl": 1.7261810302734375,
      "learning_rate": 4.994971915644536e-07,
      "loss": 0.0017,
      "reward": 1.4058036357164383,
      "reward_std": 0.3415578296408057,
      "rewards/equation_reward_func": 0.6482142992317677,
      "rewards/format_reward_func": 0.7575893234461546,
      "step": 6694
    },
    {
      "completion_length": 314.8214454650879,
      "epoch": 1.1227209857915252,
      "grad_norm": 1.2900474738091785,
      "kl": 1.621734619140625,
      "learning_rate": 4.994964308539027e-07,
      "loss": 0.0016,
      "reward": 1.278571479022503,
      "reward_std": 0.3838579710572958,
      "rewards/equation_reward_func": 0.5732143018394709,
      "rewards/format_reward_func": 0.7053571790456772,
      "step": 6696
    },
    {
      "completion_length": 272.94197368621826,
      "epoch": 1.1230562890313927,
      "grad_norm": 1.9977503024608492,
      "kl": 10.36822509765625,
      "learning_rate": 4.994956695689187e-07,
      "loss": 0.0104,
      "reward": 1.4410714656114578,
      "reward_std": 0.3055711481720209,
      "rewards/equation_reward_func": 0.6151786055415869,
      "rewards/format_reward_func": 0.8258929029107094,
      "step": 6698
    },
    {
      "completion_length": 259.0178699493408,
      "epoch": 1.1233915922712603,
      "grad_norm": 1.0655966422208405,
      "kl": 8.722640991210938,
      "learning_rate": 4.99494907709503e-07,
      "loss": 0.0087,
      "reward": 1.5750000774860382,
      "reward_std": 0.2676904257386923,
      "rewards/equation_reward_func": 0.7178571633994579,
      "rewards/format_reward_func": 0.8571428880095482,
      "step": 6700
    },
    {
      "completion_length": 249.94197750091553,
      "epoch": 1.123726895511128,
      "grad_norm": 1.1186796278129825,
      "kl": 7.51885986328125,
      "learning_rate": 4.994941452756575e-07,
      "loss": 0.0075,
      "reward": 1.6553571745753288,
      "reward_std": 0.18940360005944967,
      "rewards/equation_reward_func": 0.7491071708500385,
      "rewards/format_reward_func": 0.906250037252903,
      "step": 6702
    },
    {
      "completion_length": 241.90626335144043,
      "epoch": 1.1240621987509953,
      "grad_norm": 0.26440510902701225,
      "kl": 0.41455078125,
      "learning_rate": 4.994933822673841e-07,
      "loss": 0.0004,
      "reward": 1.6441965103149414,
      "reward_std": 0.0839689327403903,
      "rewards/equation_reward_func": 0.7053571715950966,
      "rewards/format_reward_func": 0.9388393126428127,
      "step": 6704
    },
    {
      "completion_length": 233.44197463989258,
      "epoch": 1.124397501990863,
      "grad_norm": 0.40772070413726497,
      "kl": 0.08148193359375,
      "learning_rate": 4.994926186846844e-07,
      "loss": 0.0001,
      "reward": 1.6924107670783997,
      "reward_std": 0.08270624093711376,
      "rewards/equation_reward_func": 0.7321428768336773,
      "rewards/format_reward_func": 0.9602678790688515,
      "step": 6706
    },
    {
      "completion_length": 224.9196548461914,
      "epoch": 1.1247328052307306,
      "grad_norm": 0.2948453419333735,
      "kl": 0.0688323974609375,
      "learning_rate": 4.994918545275601e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.812500037252903,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 6708
    },
    {
      "completion_length": 232.6696538925171,
      "epoch": 1.125068108470598,
      "grad_norm": 0.23710215383380753,
      "kl": 0.06982421875,
      "learning_rate": 4.994910897960132e-07,
      "loss": 0.0001,
      "reward": 1.7214286401867867,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 0.9642857238650322,
      "step": 6710
    },
    {
      "completion_length": 229.37947463989258,
      "epoch": 1.1254034117104657,
      "grad_norm": 0.28880507537327116,
      "kl": 0.083526611328125,
      "learning_rate": 4.994903244900453e-07,
      "loss": 0.0001,
      "reward": 1.7000000774860382,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7357143275439739,
      "rewards/format_reward_func": 0.9642857313156128,
      "step": 6712
    },
    {
      "completion_length": 234.78572368621826,
      "epoch": 1.1257387149503333,
      "grad_norm": 0.49477930676249415,
      "kl": 0.129913330078125,
      "learning_rate": 4.994895586096581e-07,
      "loss": 0.0001,
      "reward": 1.6857143267989159,
      "reward_std": 0.12626906856894493,
      "rewards/equation_reward_func": 0.7392857354134321,
      "rewards/format_reward_func": 0.9464285969734192,
      "step": 6714
    },
    {
      "completion_length": 234.7366180419922,
      "epoch": 1.1260740181902007,
      "grad_norm": 0.2093629175985944,
      "kl": 0.0768280029296875,
      "learning_rate": 4.994887921548533e-07,
      "loss": 0.0001,
      "reward": 1.7267857566475868,
      "reward_std": 0.10354063659906387,
      "rewards/equation_reward_func": 0.7758928947150707,
      "rewards/format_reward_func": 0.950892873108387,
      "step": 6716
    },
    {
      "completion_length": 226.94643878936768,
      "epoch": 1.1264093214300683,
      "grad_norm": 0.27764295574443226,
      "kl": 0.0676422119140625,
      "learning_rate": 4.994880251256329e-07,
      "loss": 0.0001,
      "reward": 1.6910715103149414,
      "reward_std": 0.07828682195395231,
      "rewards/equation_reward_func": 0.713392885401845,
      "rewards/format_reward_func": 0.9776785783469677,
      "step": 6718
    },
    {
      "completion_length": 252.6562614440918,
      "epoch": 1.126744624669936,
      "grad_norm": 0.7163660640986659,
      "kl": 0.150604248046875,
      "learning_rate": 4.994872575219986e-07,
      "loss": 0.0002,
      "reward": 1.6642857491970062,
      "reward_std": 0.10606601648032665,
      "rewards/equation_reward_func": 0.7267857491970062,
      "rewards/format_reward_func": 0.9375000186264515,
      "step": 6720
    },
    {
      "completion_length": 240.53572368621826,
      "epoch": 1.1270799279098034,
      "grad_norm": 0.23950711514360806,
      "kl": 0.1202392578125,
      "learning_rate": 4.994864893439521e-07,
      "loss": 0.0001,
      "reward": 1.7178572043776512,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7446428909897804,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 6722
    },
    {
      "completion_length": 229.9866189956665,
      "epoch": 1.127415231149671,
      "grad_norm": 0.7546838926343592,
      "kl": 0.2407379150390625,
      "learning_rate": 4.994857205914952e-07,
      "loss": 0.0002,
      "reward": 1.6946429088711739,
      "reward_std": 0.13384521193802357,
      "rewards/equation_reward_func": 0.7437500320374966,
      "rewards/format_reward_func": 0.9508928768336773,
      "step": 6724
    },
    {
      "completion_length": 234.6026906967163,
      "epoch": 1.1277505343895387,
      "grad_norm": 0.2525886081067382,
      "kl": 0.0670013427734375,
      "learning_rate": 4.994849512646297e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.1060660183429718,
      "rewards/equation_reward_func": 0.7875000350177288,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 6726
    },
    {
      "completion_length": 227.79911708831787,
      "epoch": 1.128085837629406,
      "grad_norm": 0.3302151167549251,
      "kl": 0.068359375,
      "learning_rate": 4.994841813633574e-07,
      "loss": 0.0001,
      "reward": 1.73214291036129,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7410714589059353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6728
    },
    {
      "completion_length": 232.83036708831787,
      "epoch": 1.1284211408692737,
      "grad_norm": 0.27840527237917856,
      "kl": 0.071990966796875,
      "learning_rate": 4.994834108876798e-07,
      "loss": 0.0001,
      "reward": 1.734375037252903,
      "reward_std": 0.0738674052990973,
      "rewards/equation_reward_func": 0.7625000439584255,
      "rewards/format_reward_func": 0.9718750081956387,
      "step": 6730
    },
    {
      "completion_length": 229.79911708831787,
      "epoch": 1.128756444109141,
      "grad_norm": 0.37449317940935545,
      "kl": 0.0745849609375,
      "learning_rate": 4.994826398375991e-07,
      "loss": 0.0001,
      "reward": 1.786160796880722,
      "reward_std": 0.09028238197788596,
      "rewards/equation_reward_func": 0.7919643148779869,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 6732
    },
    {
      "completion_length": 231.52233123779297,
      "epoch": 1.1290917473490087,
      "grad_norm": 0.4020724354337358,
      "kl": 0.074310302734375,
      "learning_rate": 4.994818682131168e-07,
      "loss": 0.0001,
      "reward": 1.8138393461704254,
      "reward_std": 0.061240497045218945,
      "rewards/equation_reward_func": 0.8178571611642838,
      "rewards/format_reward_func": 0.995982151478529,
      "step": 6734
    },
    {
      "completion_length": 226.665189743042,
      "epoch": 1.1294270505888764,
      "grad_norm": 0.21480784890133833,
      "kl": 0.0719451904296875,
      "learning_rate": 4.994810960142347e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7625000327825546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6736
    },
    {
      "completion_length": 237.665189743042,
      "epoch": 1.1297623538287438,
      "grad_norm": 0.42019404209191996,
      "kl": 0.0841827392578125,
      "learning_rate": 4.994803232409546e-07,
      "loss": 0.0001,
      "reward": 1.6464286521077156,
      "reward_std": 0.1414213553071022,
      "rewards/equation_reward_func": 0.6821428798139095,
      "rewards/format_reward_func": 0.9642857313156128,
      "step": 6738
    },
    {
      "completion_length": 235.20536708831787,
      "epoch": 1.1300976570686114,
      "grad_norm": 0.22061191284016354,
      "kl": 0.0779876708984375,
      "learning_rate": 4.994795498932784e-07,
      "loss": 0.0001,
      "reward": 1.7214286476373672,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7392857298254967,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6740
    },
    {
      "completion_length": 236.31251049041748,
      "epoch": 1.130432960308479,
      "grad_norm": 0.26718414308599536,
      "kl": 1.054351806640625,
      "learning_rate": 4.994787759712077e-07,
      "loss": 0.0011,
      "reward": 1.7089286297559738,
      "reward_std": 0.108591397292912,
      "rewards/equation_reward_func": 0.7312500327825546,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 6742
    },
    {
      "completion_length": 230.5357255935669,
      "epoch": 1.1307682635483465,
      "grad_norm": 0.30451959283162633,
      "kl": 0.08184814453125,
      "learning_rate": 4.994780014747444e-07,
      "loss": 0.0001,
      "reward": 1.764285758137703,
      "reward_std": 0.08586296439170837,
      "rewards/equation_reward_func": 0.7821429036557674,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6744
    },
    {
      "completion_length": 240.34822845458984,
      "epoch": 1.131103566788214,
      "grad_norm": 0.2806823903013915,
      "kl": 0.736236572265625,
      "learning_rate": 4.994772264038901e-07,
      "loss": 0.0007,
      "reward": 1.7303572371602058,
      "reward_std": 0.0833375845104456,
      "rewards/equation_reward_func": 0.7616071701049805,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 6746
    },
    {
      "completion_length": 242.71429538726807,
      "epoch": 1.1314388700280817,
      "grad_norm": 0.30587482030970475,
      "kl": 0.078155517578125,
      "learning_rate": 4.994764507586468e-07,
      "loss": 0.0001,
      "reward": 1.7696429267525673,
      "reward_std": 0.09343911148607731,
      "rewards/equation_reward_func": 0.7919643111526966,
      "rewards/format_reward_func": 0.9776785783469677,
      "step": 6748
    },
    {
      "completion_length": 220.5759038925171,
      "epoch": 1.1317741732679492,
      "grad_norm": 0.12420241190982315,
      "kl": 0.0816192626953125,
      "learning_rate": 4.994756745390163e-07,
      "loss": 0.0001,
      "reward": 1.7285714894533157,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7464285995811224,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6750
    },
    {
      "completion_length": 229.2544755935669,
      "epoch": 1.1321094765078168,
      "grad_norm": 0.24521204153039233,
      "kl": 0.078582763671875,
      "learning_rate": 4.994748977450002e-07,
      "loss": 0.0001,
      "reward": 1.7053572088479996,
      "reward_std": 0.10354063659906387,
      "rewards/equation_reward_func": 0.7455357387661934,
      "rewards/format_reward_func": 0.9598214440047741,
      "step": 6752
    },
    {
      "completion_length": 236.79465293884277,
      "epoch": 1.1324447797476842,
      "grad_norm": 0.2621989103061339,
      "kl": 0.086334228515625,
      "learning_rate": 4.994741203766006e-07,
      "loss": 0.0001,
      "reward": 1.7267857417464256,
      "reward_std": 0.11364216171205044,
      "rewards/equation_reward_func": 0.7580357305705547,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 6754
    },
    {
      "completion_length": 230.37054538726807,
      "epoch": 1.1327800829875518,
      "grad_norm": 0.35612925150187297,
      "kl": 0.126983642578125,
      "learning_rate": 4.994733424338187e-07,
      "loss": 0.0001,
      "reward": 1.7160715237259865,
      "reward_std": 0.11869292240589857,
      "rewards/equation_reward_func": 0.7294643297791481,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6756
    },
    {
      "completion_length": 232.8303680419922,
      "epoch": 1.1331153862274195,
      "grad_norm": 0.24637191100536926,
      "kl": 0.08123779296875,
      "learning_rate": 4.99472563916657e-07,
      "loss": 0.0001,
      "reward": 1.7357143461704254,
      "reward_std": 0.0909137288108468,
      "rewards/equation_reward_func": 0.753571467474103,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6758
    },
    {
      "completion_length": 218.29911708831787,
      "epoch": 1.1334506894672869,
      "grad_norm": 0.1348445612235299,
      "kl": 0.06597900390625,
      "learning_rate": 4.994717848251168e-07,
      "loss": 0.0001,
      "reward": 1.8089286237955093,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.813392885029316,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6760
    },
    {
      "completion_length": 235.64286708831787,
      "epoch": 1.1337859927071545,
      "grad_norm": 0.25379818975244123,
      "kl": 0.0818634033203125,
      "learning_rate": 4.994710051592001e-07,
      "loss": 0.0001,
      "reward": 1.7446429133415222,
      "reward_std": 0.10354063473641872,
      "rewards/equation_reward_func": 0.7758928909897804,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 6762
    },
    {
      "completion_length": 225.30804824829102,
      "epoch": 1.1341212959470222,
      "grad_norm": 0.29357611013427315,
      "kl": 0.6669158935546875,
      "learning_rate": 4.994702249189087e-07,
      "loss": 0.0007,
      "reward": 1.7607143595814705,
      "reward_std": 0.10606601554900408,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6764
    },
    {
      "completion_length": 226.20982933044434,
      "epoch": 1.1344565991868896,
      "grad_norm": 0.22086866513243852,
      "kl": 0.0677032470703125,
      "learning_rate": 4.994694441042442e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7482143230736256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6766
    },
    {
      "completion_length": 231.54018783569336,
      "epoch": 1.1347919024267572,
      "grad_norm": 0.2603920909323302,
      "kl": 0.0783538818359375,
      "learning_rate": 4.994686627152085e-07,
      "loss": 0.0001,
      "reward": 1.7500000447034836,
      "reward_std": 0.09596449043601751,
      "rewards/equation_reward_func": 0.7767857424914837,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 6768
    },
    {
      "completion_length": 233.05804634094238,
      "epoch": 1.1351272056666248,
      "grad_norm": 0.15957707276356545,
      "kl": 0.07037353515625,
      "learning_rate": 4.994678807518037e-07,
      "loss": 0.0001,
      "reward": 1.787500061094761,
      "reward_std": 0.08838834706693888,
      "rewards/equation_reward_func": 0.8008928820490837,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6770
    },
    {
      "completion_length": 226.86161708831787,
      "epoch": 1.1354625089064923,
      "grad_norm": 0.2632334048762584,
      "kl": 0.074432373046875,
      "learning_rate": 4.994670982140312e-07,
      "loss": 0.0001,
      "reward": 1.7714286521077156,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7803571633994579,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6772
    },
    {
      "completion_length": 232.81250953674316,
      "epoch": 1.1357978121463599,
      "grad_norm": 0.16245051836075658,
      "kl": 0.0787353515625,
      "learning_rate": 4.994663151018929e-07,
      "loss": 0.0001,
      "reward": 1.7589286118745804,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7812500260770321,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 6774
    },
    {
      "completion_length": 237.65179920196533,
      "epoch": 1.1361331153862273,
      "grad_norm": 0.3432961730562157,
      "kl": 1.7099456787109375,
      "learning_rate": 4.994655314153907e-07,
      "loss": 0.0017,
      "reward": 1.6964286267757416,
      "reward_std": 0.12121830601245165,
      "rewards/equation_reward_func": 0.7321428842842579,
      "rewards/format_reward_func": 0.9642857313156128,
      "step": 6776
    },
    {
      "completion_length": 229.12947273254395,
      "epoch": 1.136468418626095,
      "grad_norm": 0.19895687920441474,
      "kl": 0.0804595947265625,
      "learning_rate": 4.994647471545265e-07,
      "loss": 0.0001,
      "reward": 1.7357143461704254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143238186836,
      "rewards/format_reward_func": 1.0,
      "step": 6778
    },
    {
      "completion_length": 226.87947463989258,
      "epoch": 1.1368037218659626,
      "grad_norm": 0.20275366249162494,
      "kl": 0.0865631103515625,
      "learning_rate": 4.994639623193017e-07,
      "loss": 0.0001,
      "reward": 1.78035718947649,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7848214618861675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6780
    },
    {
      "completion_length": 229.3080472946167,
      "epoch": 1.1371390251058302,
      "grad_norm": 0.25465394761532567,
      "kl": 0.08404541015625,
      "learning_rate": 4.994631769097186e-07,
      "loss": 0.0001,
      "reward": 1.6696429252624512,
      "reward_std": 0.09343910869210958,
      "rewards/equation_reward_func": 0.6830357611179352,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6782
    },
    {
      "completion_length": 235.73215198516846,
      "epoch": 1.1374743283456976,
      "grad_norm": 0.2737771638741082,
      "kl": 1.16455078125,
      "learning_rate": 4.994623909257786e-07,
      "loss": 0.0012,
      "reward": 1.7500000447034836,
      "reward_std": 0.08586296811699867,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 0.9642857313156128,
      "step": 6784
    },
    {
      "completion_length": 220.9509048461914,
      "epoch": 1.1378096315855653,
      "grad_norm": 0.19918002938971385,
      "kl": 0.0744476318359375,
      "learning_rate": 4.994616043674837e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7803571745753288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6786
    },
    {
      "completion_length": 217.20536518096924,
      "epoch": 1.1381449348254327,
      "grad_norm": 0.17679503465918528,
      "kl": 0.079254150390625,
      "learning_rate": 4.994608172348356e-07,
      "loss": 0.0001,
      "reward": 1.798214353621006,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.8026786036789417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6788
    },
    {
      "completion_length": 221.59822368621826,
      "epoch": 1.1384802380653003,
      "grad_norm": 0.33131053910015773,
      "kl": 0.076934814453125,
      "learning_rate": 4.994600295278364e-07,
      "loss": 0.0001,
      "reward": 1.7750000730156898,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7750000227242708,
      "rewards/format_reward_func": 1.0,
      "step": 6790
    },
    {
      "completion_length": 213.90626049041748,
      "epoch": 1.138815541305168,
      "grad_norm": 0.3653836980839601,
      "kl": 1.16046142578125,
      "learning_rate": 4.994592412464876e-07,
      "loss": 0.0012,
      "reward": 1.7500000670552254,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7589286155998707,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6792
    },
    {
      "completion_length": 233.4017972946167,
      "epoch": 1.1391508445450353,
      "grad_norm": 0.27747762072907645,
      "kl": 0.084259033203125,
      "learning_rate": 4.994584523907914e-07,
      "loss": 0.0001,
      "reward": 1.655357226729393,
      "reward_std": 0.09848987404257059,
      "rewards/equation_reward_func": 0.6866071820259094,
      "rewards/format_reward_func": 0.9687500111758709,
      "step": 6794
    },
    {
      "completion_length": 226.19643783569336,
      "epoch": 1.139486147784903,
      "grad_norm": 1.1883478179099713,
      "kl": 3.068939208984375,
      "learning_rate": 4.99457662960749e-07,
      "loss": 0.0031,
      "reward": 1.7125000581145287,
      "reward_std": 0.1035406356677413,
      "rewards/equation_reward_func": 0.7526785898953676,
      "rewards/format_reward_func": 0.9598214440047741,
      "step": 6796
    },
    {
      "completion_length": 221.78572750091553,
      "epoch": 1.1398214510247706,
      "grad_norm": 0.2720848237059936,
      "kl": 0.0783538818359375,
      "learning_rate": 4.994568729563628e-07,
      "loss": 0.0001,
      "reward": 1.7803572043776512,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7937500402331352,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6798
    },
    {
      "completion_length": 215.40625953674316,
      "epoch": 1.140156754264638,
      "grad_norm": 0.17525951997006495,
      "kl": 0.084381103515625,
      "learning_rate": 4.994560823776342e-07,
      "loss": 0.0001,
      "reward": 1.7750000730156898,
      "reward_std": 0.08586296625435352,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6800
    },
    {
      "completion_length": 231.19643783569336,
      "epoch": 1.1404920575045057,
      "grad_norm": 0.5392629404213098,
      "kl": 0.669952392578125,
      "learning_rate": 4.994552912245653e-07,
      "loss": 0.0007,
      "reward": 1.7107143625617027,
      "reward_std": 0.11616754252463579,
      "rewards/equation_reward_func": 0.737500037997961,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 6802
    },
    {
      "completion_length": 220.7991180419922,
      "epoch": 1.1408273607443733,
      "grad_norm": 0.38399115209840423,
      "kl": 0.09869384765625,
      "learning_rate": 4.994544994971579e-07,
      "loss": 0.0001,
      "reward": 1.7017858177423477,
      "reward_std": 0.11869292240589857,
      "rewards/equation_reward_func": 0.7151786051690578,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6804
    },
    {
      "completion_length": 210.77679538726807,
      "epoch": 1.1411626639842407,
      "grad_norm": 0.3414302520044539,
      "kl": 0.08026123046875,
      "learning_rate": 4.994537071954136e-07,
      "loss": 0.0001,
      "reward": 1.7946428954601288,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7991071697324514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6806
    },
    {
      "completion_length": 216.79018688201904,
      "epoch": 1.1414979672241083,
      "grad_norm": 0.35112093270640127,
      "kl": 0.0818939208984375,
      "learning_rate": 4.994529143193344e-07,
      "loss": 0.0001,
      "reward": 1.7053571790456772,
      "reward_std": 0.09343910869210958,
      "rewards/equation_reward_func": 0.7187500260770321,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6808
    },
    {
      "completion_length": 216.69197368621826,
      "epoch": 1.1418332704639758,
      "grad_norm": 0.27308493165841274,
      "kl": 0.084320068359375,
      "learning_rate": 4.994521208689222e-07,
      "loss": 0.0001,
      "reward": 1.7107143551111221,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7196428962051868,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6810
    },
    {
      "completion_length": 229.5982255935669,
      "epoch": 1.1421685737038434,
      "grad_norm": 0.33048177298301895,
      "kl": 0.0850372314453125,
      "learning_rate": 4.994513268441786e-07,
      "loss": 0.0001,
      "reward": 1.7267857864499092,
      "reward_std": 0.09343911055475473,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 6812
    },
    {
      "completion_length": 216.61608219146729,
      "epoch": 1.142503876943711,
      "grad_norm": 0.16062072058112123,
      "kl": 0.0807952880859375,
      "learning_rate": 4.994505322451057e-07,
      "loss": 0.0001,
      "reward": 1.7446429282426834,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.74910718947649,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6814
    },
    {
      "completion_length": 212.56697273254395,
      "epoch": 1.1428391801835784,
      "grad_norm": 0.2508972626651724,
      "kl": 0.085968017578125,
      "learning_rate": 4.994497370717051e-07,
      "loss": 0.0001,
      "reward": 1.7517857924103737,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500387430191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6816
    },
    {
      "completion_length": 221.79465293884277,
      "epoch": 1.143174483423446,
      "grad_norm": 0.28485351061728537,
      "kl": 0.0821075439453125,
      "learning_rate": 4.994489413239788e-07,
      "loss": 0.0001,
      "reward": 1.7732143476605415,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7776786051690578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6818
    },
    {
      "completion_length": 217.49554538726807,
      "epoch": 1.1435097866633137,
      "grad_norm": 0.16398489059121946,
      "kl": 0.0821380615234375,
      "learning_rate": 4.994481450019285e-07,
      "loss": 0.0001,
      "reward": 1.817857176065445,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8178571611642838,
      "rewards/format_reward_func": 1.0,
      "step": 6820
    },
    {
      "completion_length": 217.56250858306885,
      "epoch": 1.1438450899031811,
      "grad_norm": 0.3405546953923567,
      "kl": 0.085357666015625,
      "learning_rate": 4.99447348105556e-07,
      "loss": 0.0001,
      "reward": 1.7325893640518188,
      "reward_std": 0.08018085593357682,
      "rewards/equation_reward_func": 0.7473214641213417,
      "rewards/format_reward_func": 0.9852678664028645,
      "step": 6822
    },
    {
      "completion_length": 224.90179634094238,
      "epoch": 1.1441803931430488,
      "grad_norm": 0.2212434884477253,
      "kl": 0.072967529296875,
      "learning_rate": 4.994465506348633e-07,
      "loss": 0.0001,
      "reward": 1.717857226729393,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7267857324331999,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6824
    },
    {
      "completion_length": 216.48661613464355,
      "epoch": 1.1445156963829164,
      "grad_norm": 0.26409679570024003,
      "kl": 0.088470458984375,
      "learning_rate": 4.994457525898521e-07,
      "loss": 0.0001,
      "reward": 1.7482143640518188,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526786010712385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6826
    },
    {
      "completion_length": 223.00000858306885,
      "epoch": 1.1448509996227838,
      "grad_norm": 0.1694539692321961,
      "kl": 0.0811614990234375,
      "learning_rate": 4.994449539705244e-07,
      "loss": 0.0001,
      "reward": 1.8147321939468384,
      "reward_std": 0.03977475711144507,
      "rewards/equation_reward_func": 0.8205357473343611,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 6828
    },
    {
      "completion_length": 217.86608123779297,
      "epoch": 1.1451863028626514,
      "grad_norm": 0.2320165700751102,
      "kl": 0.0828704833984375,
      "learning_rate": 4.994441547768818e-07,
      "loss": 0.0001,
      "reward": 1.801785759627819,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.8062500283122063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6830
    },
    {
      "completion_length": 219.41965103149414,
      "epoch": 1.1455216061025189,
      "grad_norm": 0.3201717426420188,
      "kl": 0.08636474609375,
      "learning_rate": 4.994433550089263e-07,
      "loss": 0.0001,
      "reward": 1.7446429431438446,
      "reward_std": 0.08838834799826145,
      "rewards/equation_reward_func": 0.7580357529222965,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6832
    },
    {
      "completion_length": 215.90179538726807,
      "epoch": 1.1458569093423865,
      "grad_norm": 0.1761618860877667,
      "kl": 0.084991455078125,
      "learning_rate": 4.994425546666597e-07,
      "loss": 0.0001,
      "reward": 1.803571492433548,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8125000260770321,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6834
    },
    {
      "completion_length": 222.77679538726807,
      "epoch": 1.1461922125822541,
      "grad_norm": 0.23635481786912677,
      "kl": 0.087860107421875,
      "learning_rate": 4.99441753750084e-07,
      "loss": 0.0001,
      "reward": 1.7879464998841286,
      "reward_std": 0.06755395047366619,
      "rewards/equation_reward_func": 0.7982143200933933,
      "rewards/format_reward_func": 0.9897321462631226,
      "step": 6836
    },
    {
      "completion_length": 225.16518878936768,
      "epoch": 1.1465275158221215,
      "grad_norm": 0.21894307591906984,
      "kl": 0.09490966796875,
      "learning_rate": 4.994409522592008e-07,
      "loss": 0.0001,
      "reward": 1.7214286401867867,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7303571663796902,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6838
    },
    {
      "completion_length": 222.02679538726807,
      "epoch": 1.1468628190619892,
      "grad_norm": 0.24222570084093759,
      "kl": 0.08447265625,
      "learning_rate": 4.994401501940119e-07,
      "loss": 0.0001,
      "reward": 1.7000000849366188,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7089286111295223,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6840
    },
    {
      "completion_length": 233.45090579986572,
      "epoch": 1.1471981223018568,
      "grad_norm": 0.3205876539023307,
      "kl": 0.08941650390625,
      "learning_rate": 4.994393475545195e-07,
      "loss": 0.0001,
      "reward": 1.6843750774860382,
      "reward_std": 0.08775700209662318,
      "rewards/equation_reward_func": 0.7035714648663998,
      "rewards/format_reward_func": 0.980803582817316,
      "step": 6842
    },
    {
      "completion_length": 225.56250858306885,
      "epoch": 1.1475334255417242,
      "grad_norm": 0.23515632938836778,
      "kl": 0.0845794677734375,
      "learning_rate": 4.994385443407252e-07,
      "loss": 0.0001,
      "reward": 1.7696428894996643,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.7830357439815998,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6844
    },
    {
      "completion_length": 227.61161708831787,
      "epoch": 1.1478687287815919,
      "grad_norm": 0.2374903179827161,
      "kl": 0.1004638671875,
      "learning_rate": 4.994377405526308e-07,
      "loss": 0.0001,
      "reward": 1.7410715073347092,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7544643059372902,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6846
    },
    {
      "completion_length": 222.64733123779297,
      "epoch": 1.1482040320214595,
      "grad_norm": 0.40396699261744706,
      "kl": 0.49755859375,
      "learning_rate": 4.994369361902383e-07,
      "loss": 0.0005,
      "reward": 1.7196429371833801,
      "reward_std": 0.10354063380509615,
      "rewards/equation_reward_func": 0.733035746961832,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6848
    },
    {
      "completion_length": 222.08036613464355,
      "epoch": 1.148539335261327,
      "grad_norm": 0.31380228580436215,
      "kl": 0.088623046875,
      "learning_rate": 4.994361312535495e-07,
      "loss": 0.0001,
      "reward": 1.6897322162985802,
      "reward_std": 0.09154507238417864,
      "rewards/equation_reward_func": 0.7223214562982321,
      "rewards/format_reward_func": 0.9674107320606709,
      "step": 6850
    },
    {
      "completion_length": 211.91965293884277,
      "epoch": 1.1488746385011945,
      "grad_norm": 0.2694868807371034,
      "kl": 0.081146240234375,
      "learning_rate": 4.994353257425661e-07,
      "loss": 0.0001,
      "reward": 1.7839286103844643,
      "reward_std": 0.0833375845104456,
      "rewards/equation_reward_func": 0.7973214536905289,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6852
    },
    {
      "completion_length": 223.18304538726807,
      "epoch": 1.1492099417410622,
      "grad_norm": 0.24600340041907806,
      "kl": 0.798309326171875,
      "learning_rate": 4.994345196572903e-07,
      "loss": 0.0008,
      "reward": 1.7232143580913544,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7366071790456772,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6854
    },
    {
      "completion_length": 230.9419755935669,
      "epoch": 1.1495452449809296,
      "grad_norm": 0.15237309305910776,
      "kl": 0.0922698974609375,
      "learning_rate": 4.994337129977235e-07,
      "loss": 0.0001,
      "reward": 1.7254465147852898,
      "reward_std": 0.09154507424682379,
      "rewards/equation_reward_func": 0.7401786036789417,
      "rewards/format_reward_func": 0.9852678626775742,
      "step": 6856
    },
    {
      "completion_length": 228.14733219146729,
      "epoch": 1.1498805482207972,
      "grad_norm": 0.22325111909912654,
      "kl": 0.09490966796875,
      "learning_rate": 4.99432905763868e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7767857350409031,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6858
    },
    {
      "completion_length": 227.51340293884277,
      "epoch": 1.1502158514606649,
      "grad_norm": 0.2710958581186891,
      "kl": 0.7169189453125,
      "learning_rate": 4.994320979557255e-07,
      "loss": 0.0007,
      "reward": 1.7540179193019867,
      "reward_std": 0.11553619476035237,
      "rewards/equation_reward_func": 0.7687500305473804,
      "rewards/format_reward_func": 0.9852678664028645,
      "step": 6860
    },
    {
      "completion_length": 222.80804634094238,
      "epoch": 1.1505511547005323,
      "grad_norm": 0.26345604424103974,
      "kl": 0.079986572265625,
      "learning_rate": 4.994312895732978e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7839285917580128,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6862
    },
    {
      "completion_length": 227.67858123779297,
      "epoch": 1.1508864579404,
      "grad_norm": 0.26812911857919164,
      "kl": 0.0786590576171875,
      "learning_rate": 4.994304806165867e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.8017857484519482,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6864
    },
    {
      "completion_length": 235.8616189956665,
      "epoch": 1.1512217611802673,
      "grad_norm": 0.2522975419476787,
      "kl": 0.08685302734375,
      "learning_rate": 4.994296710855942e-07,
      "loss": 0.0001,
      "reward": 1.6357143446803093,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.6803571656346321,
      "rewards/format_reward_func": 0.955357164144516,
      "step": 6866
    },
    {
      "completion_length": 234.99554538726807,
      "epoch": 1.151557064420135,
      "grad_norm": 0.25997920012377895,
      "kl": 0.07647705078125,
      "learning_rate": 4.994288609803221e-07,
      "loss": 0.0001,
      "reward": 1.7142858058214188,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.7321428842842579,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 6868
    },
    {
      "completion_length": 234.01340579986572,
      "epoch": 1.1518923676600026,
      "grad_norm": 0.547764225858209,
      "kl": 0.92034912109375,
      "learning_rate": 4.994280503007723e-07,
      "loss": 0.0009,
      "reward": 1.7598215192556381,
      "reward_std": 0.11743023293092847,
      "rewards/equation_reward_func": 0.7848214469850063,
      "rewards/format_reward_func": 0.975000012665987,
      "step": 6870
    },
    {
      "completion_length": 223.48215293884277,
      "epoch": 1.15222767089987,
      "grad_norm": 0.004337909642485559,
      "kl": 0.0759124755859375,
      "learning_rate": 4.994272390469467e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 6872
    },
    {
      "completion_length": 227.67411994934082,
      "epoch": 1.1525629741397376,
      "grad_norm": 0.26786762702478173,
      "kl": 0.093414306640625,
      "learning_rate": 4.99426427218847e-07,
      "loss": 0.0001,
      "reward": 1.745089367032051,
      "reward_std": 0.0675539500080049,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 6874
    },
    {
      "completion_length": 223.37947368621826,
      "epoch": 1.1528982773796053,
      "grad_norm": 0.14117511452087392,
      "kl": 0.0843505859375,
      "learning_rate": 4.994256148164752e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7803571801632643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6876
    },
    {
      "completion_length": 227.93304634094238,
      "epoch": 1.1532335806194727,
      "grad_norm": 0.1684037263529093,
      "kl": 0.08544921875,
      "learning_rate": 4.994248018398332e-07,
      "loss": 0.0001,
      "reward": 1.8107143342494965,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.810714315623045,
      "rewards/format_reward_func": 1.0,
      "step": 6878
    },
    {
      "completion_length": 223.60268783569336,
      "epoch": 1.1535688838593403,
      "grad_norm": 0.23348290820565865,
      "kl": 0.0734100341796875,
      "learning_rate": 4.994239882889229e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 6880
    },
    {
      "completion_length": 229.4553680419922,
      "epoch": 1.153904187099208,
      "grad_norm": 0.21225142595897906,
      "kl": 0.092926025390625,
      "learning_rate": 4.99423174163746e-07,
      "loss": 0.0001,
      "reward": 1.8196429163217545,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8241071589291096,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6882
    },
    {
      "completion_length": 226.08036994934082,
      "epoch": 1.1542394903390754,
      "grad_norm": 0.3118696688418131,
      "kl": 0.1126251220703125,
      "learning_rate": 4.994223594643044e-07,
      "loss": 0.0001,
      "reward": 1.756250061094761,
      "reward_std": 0.06818529823794961,
      "rewards/equation_reward_func": 0.7723214607685804,
      "rewards/format_reward_func": 0.9839285798370838,
      "step": 6884
    },
    {
      "completion_length": 220.00000762939453,
      "epoch": 1.154574793578943,
      "grad_norm": 0.20714372614678384,
      "kl": 0.1049652099609375,
      "learning_rate": 4.994215441906003e-07,
      "loss": 0.0001,
      "reward": 1.7500000894069672,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 6886
    },
    {
      "completion_length": 227.08036994934082,
      "epoch": 1.1549100968188104,
      "grad_norm": 0.26264781214443184,
      "kl": 0.074127197265625,
      "learning_rate": 4.99420728342635e-07,
      "loss": 0.0001,
      "reward": 1.6964286342263222,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6964285932481289,
      "rewards/format_reward_func": 1.0,
      "step": 6888
    },
    {
      "completion_length": 221.78572273254395,
      "epoch": 1.155245400058678,
      "grad_norm": 0.18060283117654174,
      "kl": 0.096527099609375,
      "learning_rate": 4.994199119204109e-07,
      "loss": 0.0001,
      "reward": 1.6803572326898575,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.6848214734345675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6890
    },
    {
      "completion_length": 224.6384048461914,
      "epoch": 1.1555807032985457,
      "grad_norm": 0.31072415977608814,
      "kl": 0.2027130126953125,
      "learning_rate": 4.994190949239297e-07,
      "loss": 0.0002,
      "reward": 1.7375000640749931,
      "reward_std": 0.06818529404699802,
      "rewards/equation_reward_func": 0.74196432903409,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6892
    },
    {
      "completion_length": 223.92411708831787,
      "epoch": 1.155916006538413,
      "grad_norm": 0.42115750864099694,
      "kl": 0.1059722900390625,
      "learning_rate": 4.994182773531932e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7732143141329288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6894
    },
    {
      "completion_length": 223.92411613464355,
      "epoch": 1.1562513097782807,
      "grad_norm": 0.1894086306269926,
      "kl": 0.0964202880859375,
      "learning_rate": 4.994174592082034e-07,
      "loss": 0.0001,
      "reward": 1.7857143729925156,
      "reward_std": 0.0909137288108468,
      "rewards/equation_reward_func": 0.7946428842842579,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6896
    },
    {
      "completion_length": 226.8035831451416,
      "epoch": 1.1565866130181484,
      "grad_norm": 0.19003955983787937,
      "kl": 0.0852813720703125,
      "learning_rate": 4.994166404889621e-07,
      "loss": 0.0001,
      "reward": 1.7178572043776512,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7267857417464256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6898
    },
    {
      "completion_length": 226.31697368621826,
      "epoch": 1.1569219162580158,
      "grad_norm": 0.23749390714239918,
      "kl": 0.1407928466796875,
      "learning_rate": 4.994158211954712e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7910714484751225,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6900
    },
    {
      "completion_length": 213.43750762939453,
      "epoch": 1.1572572194978834,
      "grad_norm": 0.16679608761011747,
      "kl": 0.100677490234375,
      "learning_rate": 4.994150013277326e-07,
      "loss": 0.0001,
      "reward": 1.8392857536673546,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8392857387661934,
      "rewards/format_reward_func": 1.0,
      "step": 6902
    },
    {
      "completion_length": 216.508939743042,
      "epoch": 1.157592522737751,
      "grad_norm": 0.22082425907055644,
      "kl": 0.10552978515625,
      "learning_rate": 4.994141808857483e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 6904
    },
    {
      "completion_length": 220.37054538726807,
      "epoch": 1.1579278259776185,
      "grad_norm": 0.5021244800347835,
      "kl": 0.0793609619140625,
      "learning_rate": 4.9941335986952e-07,
      "loss": 0.0001,
      "reward": 1.7160715013742447,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7205357514321804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6906
    },
    {
      "completion_length": 217.09822368621826,
      "epoch": 1.158263129217486,
      "grad_norm": 0.20157959080101762,
      "kl": 0.08599853515625,
      "learning_rate": 4.994125382790496e-07,
      "loss": 0.0001,
      "reward": 1.7482143864035606,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6908
    },
    {
      "completion_length": 214.39733028411865,
      "epoch": 1.1585984324573535,
      "grad_norm": 0.17372353002627805,
      "kl": 0.075927734375,
      "learning_rate": 4.994117161143392e-07,
      "loss": 0.0001,
      "reward": 1.7250000685453415,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000424683094,
      "rewards/format_reward_func": 1.0,
      "step": 6910
    },
    {
      "completion_length": 217.69197177886963,
      "epoch": 1.1589337356972211,
      "grad_norm": 0.18357542006548058,
      "kl": 0.0815887451171875,
      "learning_rate": 4.994108933753905e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 6912
    },
    {
      "completion_length": 216.56250953674316,
      "epoch": 1.1592690389370888,
      "grad_norm": 0.14700847457999847,
      "kl": 0.0750274658203125,
      "learning_rate": 4.994100700622054e-07,
      "loss": 0.0001,
      "reward": 1.753571480512619,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714656114578,
      "rewards/format_reward_func": 1.0,
      "step": 6914
    },
    {
      "completion_length": 227.56251049041748,
      "epoch": 1.1596043421769564,
      "grad_norm": 0.24436108533912068,
      "kl": 0.0693511962890625,
      "learning_rate": 4.994092461747859e-07,
      "loss": 0.0001,
      "reward": 1.7303572371602058,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7348214611411095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6916
    },
    {
      "completion_length": 212.84375762939453,
      "epoch": 1.1599396454168238,
      "grad_norm": 0.24133830306304005,
      "kl": 0.0664825439453125,
      "learning_rate": 4.994084217131338e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 6918
    },
    {
      "completion_length": 212.49554443359375,
      "epoch": 1.1602749486566915,
      "grad_norm": 0.2615600831239175,
      "kl": 0.07427978515625,
      "learning_rate": 4.994075966772511e-07,
      "loss": 0.0001,
      "reward": 1.707142911851406,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7071428894996643,
      "rewards/format_reward_func": 1.0,
      "step": 6920
    },
    {
      "completion_length": 224.66965293884277,
      "epoch": 1.1606102518965589,
      "grad_norm": 0.15686356558490766,
      "kl": 0.4253997802734375,
      "learning_rate": 4.994067710671396e-07,
      "loss": 0.0004,
      "reward": 1.7267857640981674,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7312500439584255,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6922
    },
    {
      "completion_length": 215.59822273254395,
      "epoch": 1.1609455551364265,
      "grad_norm": 0.22075119949291844,
      "kl": 0.071746826171875,
      "learning_rate": 4.994059448828011e-07,
      "loss": 0.0001,
      "reward": 1.7678571939468384,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571790456772,
      "rewards/format_reward_func": 1.0,
      "step": 6924
    },
    {
      "completion_length": 224.16518878936768,
      "epoch": 1.1612808583762941,
      "grad_norm": 0.2866122016574234,
      "kl": 0.0772247314453125,
      "learning_rate": 4.994051181242379e-07,
      "loss": 0.0001,
      "reward": 1.7500000894069672,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 6926
    },
    {
      "completion_length": 199.56250858306885,
      "epoch": 1.1616161616161615,
      "grad_norm": 0.2835153067499442,
      "kl": 0.07659912109375,
      "learning_rate": 4.994042907914514e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 6928
    },
    {
      "completion_length": 220.72768878936768,
      "epoch": 1.1619514648560292,
      "grad_norm": 0.22777164421377802,
      "kl": 0.6842498779296875,
      "learning_rate": 4.99403462884444e-07,
      "loss": 0.0007,
      "reward": 1.7928571999073029,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571589291096,
      "rewards/format_reward_func": 1.0,
      "step": 6930
    },
    {
      "completion_length": 219.58482933044434,
      "epoch": 1.1622867680958968,
      "grad_norm": 0.17939756568933712,
      "kl": 0.073150634765625,
      "learning_rate": 4.994026344032172e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7696428932249546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6932
    },
    {
      "completion_length": 210.61608028411865,
      "epoch": 1.1626220713357642,
      "grad_norm": 0.3478303221101954,
      "kl": 0.0870361328125,
      "learning_rate": 4.994018053477731e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.760714303702116,
      "rewards/format_reward_func": 1.0,
      "step": 6934
    },
    {
      "completion_length": 210.92858219146729,
      "epoch": 1.1629573745756319,
      "grad_norm": 0.1651791347237627,
      "kl": 0.0748748779296875,
      "learning_rate": 4.994009757181136e-07,
      "loss": 0.0001,
      "reward": 1.758928656578064,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7633928805589676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6936
    },
    {
      "completion_length": 215.88393783569336,
      "epoch": 1.1632926778154995,
      "grad_norm": 0.3029528530372732,
      "kl": 0.0755157470703125,
      "learning_rate": 4.994001455142405e-07,
      "loss": 0.0001,
      "reward": 1.7017857730388641,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7062500566244125,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6938
    },
    {
      "completion_length": 211.4419755935669,
      "epoch": 1.163627981055367,
      "grad_norm": 0.26122991877460666,
      "kl": 0.087005615234375,
      "learning_rate": 4.993993147361558e-07,
      "loss": 0.0001,
      "reward": 1.7000000700354576,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7089286148548126,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6940
    },
    {
      "completion_length": 208.23661613464355,
      "epoch": 1.1639632842952345,
      "grad_norm": 0.2747568610792723,
      "kl": 0.0793609619140625,
      "learning_rate": 4.993984833838615e-07,
      "loss": 0.0001,
      "reward": 1.785714328289032,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143245637417,
      "rewards/format_reward_func": 1.0,
      "step": 6942
    },
    {
      "completion_length": 213.88393688201904,
      "epoch": 1.164298587535102,
      "grad_norm": 0.2901435145294046,
      "kl": 0.07421875,
      "learning_rate": 4.993976514573593e-07,
      "loss": 0.0001,
      "reward": 1.692857213318348,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.692857176065445,
      "rewards/format_reward_func": 1.0,
      "step": 6944
    },
    {
      "completion_length": 214.64286708831787,
      "epoch": 1.1646338907749696,
      "grad_norm": 0.2537096914951688,
      "kl": 0.07794189453125,
      "learning_rate": 4.993968189566513e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 6946
    },
    {
      "completion_length": 223.53572273254395,
      "epoch": 1.1649691940148372,
      "grad_norm": 0.2668001365446884,
      "kl": 0.0701446533203125,
      "learning_rate": 4.993959858817395e-07,
      "loss": 0.0001,
      "reward": 1.7732143253087997,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7776786088943481,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6948
    },
    {
      "completion_length": 218.05358123779297,
      "epoch": 1.1653044972547046,
      "grad_norm": 0.21744701461798605,
      "kl": 0.0718994140625,
      "learning_rate": 4.993951522326255e-07,
      "loss": 0.0001,
      "reward": 1.769642911851406,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.774107176810503,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6950
    },
    {
      "completion_length": 229.17858123779297,
      "epoch": 1.1656398004945723,
      "grad_norm": 0.21091420914175668,
      "kl": 0.073822021484375,
      "learning_rate": 4.993943180093114e-07,
      "loss": 0.0001,
      "reward": 1.7321429252624512,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321428880095482,
      "rewards/format_reward_func": 1.0,
      "step": 6952
    },
    {
      "completion_length": 229.37500953674316,
      "epoch": 1.16597510373444,
      "grad_norm": 0.28567924027709884,
      "kl": 0.0760650634765625,
      "learning_rate": 4.993934832117991e-07,
      "loss": 0.0001,
      "reward": 1.7125000730156898,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.725892897695303,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 6954
    },
    {
      "completion_length": 226.1384038925171,
      "epoch": 1.1663104069743073,
      "grad_norm": 0.30841935931803693,
      "kl": 0.0717315673828125,
      "learning_rate": 4.993926478400906e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.8107143193483353,
      "rewards/format_reward_func": 1.0,
      "step": 6956
    },
    {
      "completion_length": 235.94643783569336,
      "epoch": 1.166645710214175,
      "grad_norm": 0.23208670864060132,
      "kl": 0.3082733154296875,
      "learning_rate": 4.993918118941877e-07,
      "loss": 0.0003,
      "reward": 1.7410715222358704,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7455357424914837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6958
    },
    {
      "completion_length": 224.87054443359375,
      "epoch": 1.1669810134540426,
      "grad_norm": 0.14985084192465364,
      "kl": 0.0709381103515625,
      "learning_rate": 4.993909753740924e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 6960
    },
    {
      "completion_length": 235.30358123779297,
      "epoch": 1.16731631669391,
      "grad_norm": 0.24284178348934188,
      "kl": 0.543304443359375,
      "learning_rate": 4.993901382798067e-07,
      "loss": 0.0005,
      "reward": 1.766071505844593,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7705357372760773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6962
    },
    {
      "completion_length": 234.59822463989258,
      "epoch": 1.1676516199337776,
      "grad_norm": 0.15701413152998836,
      "kl": 0.3557891845703125,
      "learning_rate": 4.993893006113323e-07,
      "loss": 0.0004,
      "reward": 1.7339285984635353,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7562500312924385,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 6964
    },
    {
      "completion_length": 234.5134048461914,
      "epoch": 1.167986923173645,
      "grad_norm": 0.22042202431841593,
      "kl": 0.0718994140625,
      "learning_rate": 4.993884623686713e-07,
      "loss": 0.0001,
      "reward": 1.7642858028411865,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857320606709,
      "rewards/format_reward_func": 1.0,
      "step": 6966
    },
    {
      "completion_length": 234.33929443359375,
      "epoch": 1.1683222264135127,
      "grad_norm": 0.2321463814591959,
      "kl": 0.083343505859375,
      "learning_rate": 4.993876235518256e-07,
      "loss": 0.0001,
      "reward": 1.7160714715719223,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7205357551574707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6968
    },
    {
      "completion_length": 237.26340198516846,
      "epoch": 1.1686575296533803,
      "grad_norm": 0.23375655605151902,
      "kl": 0.0672149658203125,
      "learning_rate": 4.993867841607972e-07,
      "loss": 0.0001,
      "reward": 1.7535714656114578,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714842379093,
      "rewards/format_reward_func": 1.0,
      "step": 6970
    },
    {
      "completion_length": 236.7321538925171,
      "epoch": 1.1689928328932477,
      "grad_norm": 0.16916676139529008,
      "kl": 0.067230224609375,
      "learning_rate": 4.993859441955879e-07,
      "loss": 0.0001,
      "reward": 1.7303572073578835,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7348214723169804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6972
    },
    {
      "completion_length": 227.88393878936768,
      "epoch": 1.1693281361331154,
      "grad_norm": 0.21767153944629555,
      "kl": 0.068359375,
      "learning_rate": 4.993851036561996e-07,
      "loss": 0.0001,
      "reward": 1.7982143461704254,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8026785887777805,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6974
    },
    {
      "completion_length": 238.67858219146729,
      "epoch": 1.169663439372983,
      "grad_norm": 0.2871662999160085,
      "kl": 0.0770721435546875,
      "learning_rate": 4.993842625426344e-07,
      "loss": 0.0001,
      "reward": 1.755357213318348,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7598214577883482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6976
    },
    {
      "completion_length": 243.97322750091553,
      "epoch": 1.1699987426128504,
      "grad_norm": 0.2003109959117877,
      "kl": 0.0980682373046875,
      "learning_rate": 4.993834208548942e-07,
      "loss": 0.0001,
      "reward": 1.7178572341799736,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7267857436090708,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 6978
    },
    {
      "completion_length": 243.05358409881592,
      "epoch": 1.170334045852718,
      "grad_norm": 0.4412118535614136,
      "kl": 0.1660919189453125,
      "learning_rate": 4.993825785929808e-07,
      "loss": 0.0002,
      "reward": 1.7375000640749931,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7419643215835094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6980
    },
    {
      "completion_length": 238.2009038925171,
      "epoch": 1.1706693490925857,
      "grad_norm": 0.2531765682553049,
      "kl": 0.070648193359375,
      "learning_rate": 4.993817357568963e-07,
      "loss": 0.0001,
      "reward": 1.751785784959793,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7562500387430191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6982
    },
    {
      "completion_length": 237.69197463989258,
      "epoch": 1.171004652332453,
      "grad_norm": 0.1307315755554436,
      "kl": 0.07025146484375,
      "learning_rate": 4.993808923466427e-07,
      "loss": 0.0001,
      "reward": 1.8071429207921028,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071429021656513,
      "rewards/format_reward_func": 1.0,
      "step": 6984
    },
    {
      "completion_length": 229.9241180419922,
      "epoch": 1.1713399555723207,
      "grad_norm": 0.16282934602909968,
      "kl": 0.0719757080078125,
      "learning_rate": 4.993800483622217e-07,
      "loss": 0.0001,
      "reward": 1.7964286357164383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285872876644,
      "rewards/format_reward_func": 1.0,
      "step": 6986
    },
    {
      "completion_length": 235.06250858306885,
      "epoch": 1.1716752588121884,
      "grad_norm": 0.15396331685512085,
      "kl": 0.061370849609375,
      "learning_rate": 4.993792038036353e-07,
      "loss": 0.0001,
      "reward": 1.8785714581608772,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8785714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 6988
    },
    {
      "completion_length": 230.7321538925171,
      "epoch": 1.1720105620520558,
      "grad_norm": 0.20821104135962984,
      "kl": 0.1810455322265625,
      "learning_rate": 4.993783586708857e-07,
      "loss": 0.0002,
      "reward": 1.7785714864730835,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 6990
    },
    {
      "completion_length": 239.67411994934082,
      "epoch": 1.1723458652919234,
      "grad_norm": 0.13161959537442824,
      "kl": 0.0686798095703125,
      "learning_rate": 4.993775129639744e-07,
      "loss": 0.0001,
      "reward": 1.8071429133415222,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8071428798139095,
      "rewards/format_reward_func": 1.0,
      "step": 6992
    },
    {
      "completion_length": 240.89286994934082,
      "epoch": 1.172681168531791,
      "grad_norm": 0.20917233948066855,
      "kl": 0.2086334228515625,
      "learning_rate": 4.993766666829038e-07,
      "loss": 0.0002,
      "reward": 1.7125000655651093,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7169643230736256,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 6994
    },
    {
      "completion_length": 242.56697273254395,
      "epoch": 1.1730164717716585,
      "grad_norm": 0.14468665687786536,
      "kl": 0.064697265625,
      "learning_rate": 4.993758198276756e-07,
      "loss": 0.0001,
      "reward": 1.7678571939468384,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571529686451,
      "rewards/format_reward_func": 1.0,
      "step": 6996
    },
    {
      "completion_length": 246.53126049041748,
      "epoch": 1.173351775011526,
      "grad_norm": 0.24789853835463235,
      "kl": 0.076202392578125,
      "learning_rate": 4.993749723982918e-07,
      "loss": 0.0001,
      "reward": 1.8178571835160255,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.817857176065445,
      "rewards/format_reward_func": 1.0,
      "step": 6998
    },
    {
      "completion_length": 237.70983219146729,
      "epoch": 1.1736870782513935,
      "grad_norm": 0.22220896735230375,
      "kl": 0.0637969970703125,
      "learning_rate": 4.993741243947544e-07,
      "loss": 0.0001,
      "reward": 1.8357143327593803,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8357143066823483,
      "rewards/format_reward_func": 1.0,
      "step": 7000
    },
    {
      "completion_length": 243.8616180419922,
      "epoch": 1.1740223814912611,
      "grad_norm": 0.24457270120564578,
      "kl": 0.060516357421875,
      "learning_rate": 4.993732758170652e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 7002
    },
    {
      "completion_length": 234.6116189956665,
      "epoch": 1.1743576847311288,
      "grad_norm": 0.25798999402321,
      "kl": 0.06378173828125,
      "learning_rate": 4.993724266652263e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 7004
    },
    {
      "completion_length": 243.3660831451416,
      "epoch": 1.1746929879709962,
      "grad_norm": 0.12745495764158657,
      "kl": 0.0615386962890625,
      "learning_rate": 4.993715769392397e-07,
      "loss": 0.0001,
      "reward": 1.7053572088479996,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.709821468219161,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7006
    },
    {
      "completion_length": 241.02679443359375,
      "epoch": 1.1750282912108638,
      "grad_norm": 0.2753649872008471,
      "kl": 0.11676025390625,
      "learning_rate": 4.993707266391072e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7767857611179352,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7008
    },
    {
      "completion_length": 236.1473331451416,
      "epoch": 1.1753635944507315,
      "grad_norm": 0.21138871590711736,
      "kl": 0.42535400390625,
      "learning_rate": 4.993698757648308e-07,
      "loss": 0.0004,
      "reward": 1.8035714700818062,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 7010
    },
    {
      "completion_length": 236.2946538925171,
      "epoch": 1.1756988976905989,
      "grad_norm": 0.20702311310531854,
      "kl": 0.150177001953125,
      "learning_rate": 4.993690243164125e-07,
      "loss": 0.0002,
      "reward": 1.7178572192788124,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7178571633994579,
      "rewards/format_reward_func": 1.0,
      "step": 7012
    },
    {
      "completion_length": 246.5000123977661,
      "epoch": 1.1760342009304665,
      "grad_norm": 0.24995814190921872,
      "kl": 0.3637237548828125,
      "learning_rate": 4.993681722938542e-07,
      "loss": 0.0004,
      "reward": 1.7696429342031479,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.774107176810503,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7014
    },
    {
      "completion_length": 241.29912090301514,
      "epoch": 1.1763695041703341,
      "grad_norm": 0.1647114277394096,
      "kl": 0.104705810546875,
      "learning_rate": 4.993673196971581e-07,
      "loss": 0.0001,
      "reward": 1.739285796880722,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857521772385,
      "rewards/format_reward_func": 1.0,
      "step": 7016
    },
    {
      "completion_length": 244.20983409881592,
      "epoch": 1.1767048074102016,
      "grad_norm": 0.2431811099491955,
      "kl": 0.0694427490234375,
      "learning_rate": 4.993664665263258e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285928755999,
      "rewards/format_reward_func": 1.0,
      "step": 7018
    },
    {
      "completion_length": 250.6562623977661,
      "epoch": 1.1770401106500692,
      "grad_norm": 0.2823400285980501,
      "kl": 0.07000732421875,
      "learning_rate": 4.993656127813594e-07,
      "loss": 0.0001,
      "reward": 1.691071517765522,
      "reward_std": 0.09343910776078701,
      "rewards/equation_reward_func": 0.6955357547849417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7020
    },
    {
      "completion_length": 243.43304634094238,
      "epoch": 1.1773754138899366,
      "grad_norm": 0.2006690204806412,
      "kl": 0.0662994384765625,
      "learning_rate": 4.99364758462261e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 7022
    },
    {
      "completion_length": 239.68304538726807,
      "epoch": 1.1777107171298042,
      "grad_norm": 0.25763110318507637,
      "kl": 0.735565185546875,
      "learning_rate": 4.993639035690325e-07,
      "loss": 0.0007,
      "reward": 1.7750000804662704,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 7024
    },
    {
      "completion_length": 243.7634038925171,
      "epoch": 1.1780460203696719,
      "grad_norm": 0.2525095936361976,
      "kl": 0.0626220703125,
      "learning_rate": 4.993630481016758e-07,
      "loss": 0.0001,
      "reward": 1.821428619325161,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.8214286044239998,
      "rewards/format_reward_func": 1.0,
      "step": 7026
    },
    {
      "completion_length": 238.9330472946167,
      "epoch": 1.1783813236095393,
      "grad_norm": 0.32584782130502027,
      "kl": 0.08795166015625,
      "learning_rate": 4.993621920601928e-07,
      "loss": 0.0001,
      "reward": 1.7750001102685928,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 7028
    },
    {
      "completion_length": 238.77233219146729,
      "epoch": 1.178716626849407,
      "grad_norm": 0.18152868420415122,
      "kl": 0.1728973388671875,
      "learning_rate": 4.993613354445857e-07,
      "loss": 0.0002,
      "reward": 1.7464286237955093,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.755357176065445,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7030
    },
    {
      "completion_length": 234.59822463989258,
      "epoch": 1.1790519300892746,
      "grad_norm": 0.21606357065813922,
      "kl": 0.0672607421875,
      "learning_rate": 4.993604782548563e-07,
      "loss": 0.0001,
      "reward": 1.7410714998841286,
      "reward_std": 0.06313453521579504,
      "rewards/equation_reward_func": 0.7544643208384514,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7032
    },
    {
      "completion_length": 245.071439743042,
      "epoch": 1.179387233329142,
      "grad_norm": 0.11731433901397256,
      "kl": 0.0707244873046875,
      "learning_rate": 4.993596204910067e-07,
      "loss": 0.0001,
      "reward": 1.7125000804662704,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.716964315623045,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7034
    },
    {
      "completion_length": 234.6741189956665,
      "epoch": 1.1797225365690096,
      "grad_norm": 0.15904667048164609,
      "kl": 0.066925048828125,
      "learning_rate": 4.993587621530386e-07,
      "loss": 0.0001,
      "reward": 1.755357213318348,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7598214522004128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7036
    },
    {
      "completion_length": 244.46875953674316,
      "epoch": 1.1800578398088772,
      "grad_norm": 0.21918559336727544,
      "kl": 0.068328857421875,
      "learning_rate": 4.993579032409544e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 7038
    },
    {
      "completion_length": 241.32144165039062,
      "epoch": 1.1803931430487447,
      "grad_norm": 0.28183054748017666,
      "kl": 0.1956024169921875,
      "learning_rate": 4.993570437547558e-07,
      "loss": 0.0002,
      "reward": 1.8089286535978317,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.813392885029316,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7040
    },
    {
      "completion_length": 240.8571538925171,
      "epoch": 1.1807284462886123,
      "grad_norm": 0.17991330132898947,
      "kl": 0.0831451416015625,
      "learning_rate": 4.993561836944447e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 7042
    },
    {
      "completion_length": 244.83929920196533,
      "epoch": 1.1810637495284797,
      "grad_norm": 0.15519989350192553,
      "kl": 0.0687103271484375,
      "learning_rate": 4.993553230600233e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 7044
    },
    {
      "completion_length": 244.7812614440918,
      "epoch": 1.1813990527683473,
      "grad_norm": 0.2493715816928291,
      "kl": 0.0808258056640625,
      "learning_rate": 4.993544618514935e-07,
      "loss": 0.0001,
      "reward": 1.7040179297327995,
      "reward_std": 0.06502856919541955,
      "rewards/equation_reward_func": 0.7098214626312256,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 7046
    },
    {
      "completion_length": 250.14287090301514,
      "epoch": 1.181734356008215,
      "grad_norm": 0.22729720223898298,
      "kl": 0.0684661865234375,
      "learning_rate": 4.993536000688573e-07,
      "loss": 0.0001,
      "reward": 1.6785715147852898,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.6785714477300644,
      "rewards/format_reward_func": 1.0,
      "step": 7048
    },
    {
      "completion_length": 248.52679538726807,
      "epoch": 1.1820696592480826,
      "grad_norm": 0.2324995386469855,
      "kl": 0.06976318359375,
      "learning_rate": 4.993527377121166e-07,
      "loss": 0.0001,
      "reward": 1.7517857924103737,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7562500275671482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7050
    },
    {
      "completion_length": 248.4419765472412,
      "epoch": 1.18240496248795,
      "grad_norm": 0.18319454840783975,
      "kl": 0.0713653564453125,
      "learning_rate": 4.993518747812735e-07,
      "loss": 0.0001,
      "reward": 1.7107143551111221,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7276786006987095,
      "rewards/format_reward_func": 0.9830357283353806,
      "step": 7052
    },
    {
      "completion_length": 238.18751049041748,
      "epoch": 1.1827402657278177,
      "grad_norm": 0.23620609642510929,
      "kl": 0.100982666015625,
      "learning_rate": 4.993510112763299e-07,
      "loss": 0.0001,
      "reward": 1.751785784959793,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500201165676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7054
    },
    {
      "completion_length": 253.98661994934082,
      "epoch": 1.183075568967685,
      "grad_norm": 0.2541844366706011,
      "kl": 0.0661773681640625,
      "learning_rate": 4.993501471972879e-07,
      "loss": 0.0001,
      "reward": 1.7250000983476639,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7339286096394062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7056
    },
    {
      "completion_length": 243.8214406967163,
      "epoch": 1.1834108722075527,
      "grad_norm": 0.19942797787478084,
      "kl": 0.07354736328125,
      "learning_rate": 4.993492825441493e-07,
      "loss": 0.0001,
      "reward": 1.7642858028411865,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857320606709,
      "rewards/format_reward_func": 1.0,
      "step": 7058
    },
    {
      "completion_length": 236.7232265472412,
      "epoch": 1.1837461754474203,
      "grad_norm": 0.21271859218155764,
      "kl": 0.065704345703125,
      "learning_rate": 4.993484173169162e-07,
      "loss": 0.0001,
      "reward": 1.7285715118050575,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7285714596509933,
      "rewards/format_reward_func": 1.0,
      "step": 7060
    },
    {
      "completion_length": 239.19197368621826,
      "epoch": 1.1840814786872877,
      "grad_norm": 0.28199565232331336,
      "kl": 0.12799072265625,
      "learning_rate": 4.993475515155906e-07,
      "loss": 0.0001,
      "reward": 1.758928656578064,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928842842579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7062
    },
    {
      "completion_length": 246.8928680419922,
      "epoch": 1.1844167819271554,
      "grad_norm": 0.25315393129614683,
      "kl": 0.5302581787109375,
      "learning_rate": 4.993466851401745e-07,
      "loss": 0.0005,
      "reward": 1.7571429163217545,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7660714574158192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7064
    },
    {
      "completion_length": 235.8125123977661,
      "epoch": 1.184752085167023,
      "grad_norm": 0.2698316473301265,
      "kl": 0.0690765380859375,
      "learning_rate": 4.993458181906699e-07,
      "loss": 0.0001,
      "reward": 1.8089286237955093,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.813392885029316,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7066
    },
    {
      "completion_length": 244.77679538726807,
      "epoch": 1.1850873884068904,
      "grad_norm": 0.2320733825817402,
      "kl": 0.0703582763671875,
      "learning_rate": 4.993449506670788e-07,
      "loss": 0.0001,
      "reward": 1.7482143566012383,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526786103844643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7068
    },
    {
      "completion_length": 259.99554920196533,
      "epoch": 1.185422691646758,
      "grad_norm": 0.24226402232737604,
      "kl": 0.0719451904296875,
      "learning_rate": 4.99344082569403e-07,
      "loss": 0.0001,
      "reward": 1.7392857819795609,
      "reward_std": 0.08081220369786024,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 7070
    },
    {
      "completion_length": 247.50447940826416,
      "epoch": 1.1857579948866257,
      "grad_norm": 0.18774264494588802,
      "kl": 0.0692596435546875,
      "learning_rate": 4.993432138976448e-07,
      "loss": 0.0001,
      "reward": 1.8000000417232513,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000324100256,
      "rewards/format_reward_func": 1.0,
      "step": 7072
    },
    {
      "completion_length": 254.71876049041748,
      "epoch": 1.1860932981264931,
      "grad_norm": 0.2447953619963093,
      "kl": 0.2768707275390625,
      "learning_rate": 4.993423446518063e-07,
      "loss": 0.0003,
      "reward": 1.7392857894301414,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.748214315623045,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7074
    },
    {
      "completion_length": 252.3259038925171,
      "epoch": 1.1864286013663607,
      "grad_norm": 0.2295580947248307,
      "kl": 0.15301513671875,
      "learning_rate": 4.99341474831889e-07,
      "loss": 0.0002,
      "reward": 1.721428632736206,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7303571682423353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7076
    },
    {
      "completion_length": 255.95536422729492,
      "epoch": 1.1867639046062282,
      "grad_norm": 0.2032927882840729,
      "kl": 0.0631103515625,
      "learning_rate": 4.993406044378951e-07,
      "loss": 0.0001,
      "reward": 1.7196429371833801,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7241071779280901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7078
    },
    {
      "completion_length": 253.1696548461914,
      "epoch": 1.1870992078460958,
      "grad_norm": 0.2855426504789579,
      "kl": 0.1221771240234375,
      "learning_rate": 4.993397334698269e-07,
      "loss": 0.0001,
      "reward": 1.7464286461472511,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.746428593993187,
      "rewards/format_reward_func": 1.0,
      "step": 7080
    },
    {
      "completion_length": 259.46429538726807,
      "epoch": 1.1874345110859634,
      "grad_norm": 0.3224579946182638,
      "kl": 0.1277923583984375,
      "learning_rate": 4.993388619276861e-07,
      "loss": 0.0001,
      "reward": 1.7053572162985802,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7098214700818062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7082
    },
    {
      "completion_length": 247.4151906967163,
      "epoch": 1.1877698143258308,
      "grad_norm": 0.18945706403060317,
      "kl": 0.0734100341796875,
      "learning_rate": 4.993379898114748e-07,
      "loss": 0.0001,
      "reward": 1.7250000536441803,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7339286059141159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7084
    },
    {
      "completion_length": 253.5312614440918,
      "epoch": 1.1881051175656985,
      "grad_norm": 0.2508063849897521,
      "kl": 0.0714569091796875,
      "learning_rate": 4.99337117121195e-07,
      "loss": 0.0001,
      "reward": 1.7589286416769028,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7633928768336773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7086
    },
    {
      "completion_length": 242.6830472946167,
      "epoch": 1.1884404208055661,
      "grad_norm": 0.21219012626209383,
      "kl": 0.0686187744140625,
      "learning_rate": 4.993362438568487e-07,
      "loss": 0.0001,
      "reward": 1.8035714998841286,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8035714402794838,
      "rewards/format_reward_func": 1.0,
      "step": 7088
    },
    {
      "completion_length": 237.0357265472412,
      "epoch": 1.1887757240454335,
      "grad_norm": 0.19441894647892927,
      "kl": 0.092987060546875,
      "learning_rate": 4.993353700184379e-07,
      "loss": 0.0001,
      "reward": 1.7517857775092125,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500275671482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7090
    },
    {
      "completion_length": 257.4017972946167,
      "epoch": 1.1891110272853012,
      "grad_norm": 0.2658164508523097,
      "kl": 0.063018798828125,
      "learning_rate": 4.993344956059646e-07,
      "loss": 0.0001,
      "reward": 1.7125001028180122,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7258928902447224,
      "rewards/format_reward_func": 0.9866071455180645,
      "step": 7092
    },
    {
      "completion_length": 254.77679920196533,
      "epoch": 1.1894463305251688,
      "grad_norm": 0.13371165163675044,
      "kl": 0.062957763671875,
      "learning_rate": 4.99333620619431e-07,
      "loss": 0.0001,
      "reward": 1.769642896950245,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7741071879863739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7094
    },
    {
      "completion_length": 250.5982265472412,
      "epoch": 1.1897816337650362,
      "grad_norm": 0.24803401131551336,
      "kl": 0.0692291259765625,
      "learning_rate": 4.993327450588388e-07,
      "loss": 0.0001,
      "reward": 1.7214286252856255,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7214286141097546,
      "rewards/format_reward_func": 1.0,
      "step": 7096
    },
    {
      "completion_length": 250.93304538726807,
      "epoch": 1.1901169370049038,
      "grad_norm": 0.2041804890215718,
      "kl": 0.0629425048828125,
      "learning_rate": 4.993318689241902e-07,
      "loss": 0.0001,
      "reward": 1.8410714864730835,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.8455357365310192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7098
    },
    {
      "completion_length": 253.7589406967163,
      "epoch": 1.1904522402447713,
      "grad_norm": 0.27051242913768825,
      "kl": 0.101287841796875,
      "learning_rate": 4.993309922154872e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7732143141329288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7100
    },
    {
      "completion_length": 264.7232275009155,
      "epoch": 1.1907875434846389,
      "grad_norm": 0.22080903137693467,
      "kl": 0.06524658203125,
      "learning_rate": 4.993301149327319e-07,
      "loss": 0.0001,
      "reward": 1.6964286640286446,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.705357164144516,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7102
    },
    {
      "completion_length": 256.1785831451416,
      "epoch": 1.1911228467245065,
      "grad_norm": 0.125236473476325,
      "kl": 0.0610809326171875,
      "learning_rate": 4.993292370759261e-07,
      "loss": 0.0001,
      "reward": 1.7660714983940125,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7705357298254967,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7104
    },
    {
      "completion_length": 245.0491180419922,
      "epoch": 1.191458149964374,
      "grad_norm": 0.1692324014060084,
      "kl": 0.0619354248046875,
      "learning_rate": 4.99328358645072e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 7106
    },
    {
      "completion_length": 261.42858600616455,
      "epoch": 1.1917934532042416,
      "grad_norm": 0.3721622893705267,
      "kl": 0.1228485107421875,
      "learning_rate": 4.993274796401716e-07,
      "loss": 0.0001,
      "reward": 1.6660715118050575,
      "reward_std": 0.07828682195395231,
      "rewards/equation_reward_func": 0.6794643178582191,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7108
    },
    {
      "completion_length": 259.6696548461914,
      "epoch": 1.1921287564441092,
      "grad_norm": 0.0981046329887637,
      "kl": 0.069854736328125,
      "learning_rate": 4.99326600061227e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 7110
    },
    {
      "completion_length": 264.2544775009155,
      "epoch": 1.1924640596839766,
      "grad_norm": 0.17182040481196187,
      "kl": 0.0637969970703125,
      "learning_rate": 4.993257199082399e-07,
      "loss": 0.0001,
      "reward": 1.7232143729925156,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7276785876601934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7112
    },
    {
      "completion_length": 263.9687614440918,
      "epoch": 1.1927993629238443,
      "grad_norm": 0.20245448175755204,
      "kl": 0.0615386962890625,
      "learning_rate": 4.993248391812127e-07,
      "loss": 0.0001,
      "reward": 1.7946429029107094,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7991071715950966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7114
    },
    {
      "completion_length": 245.4375114440918,
      "epoch": 1.1931346661637119,
      "grad_norm": 0.18806544126057884,
      "kl": 0.0662994384765625,
      "learning_rate": 4.993239578801473e-07,
      "loss": 0.0001,
      "reward": 1.7428572252392769,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 7116
    },
    {
      "completion_length": 253.40179634094238,
      "epoch": 1.1934699694035793,
      "grad_norm": 0.16657846626696407,
      "kl": 0.1014556884765625,
      "learning_rate": 4.993230760050456e-07,
      "loss": 0.0001,
      "reward": 1.7571429088711739,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 7118
    },
    {
      "completion_length": 257.8794765472412,
      "epoch": 1.193805272643447,
      "grad_norm": 0.2076503664785633,
      "kl": 0.1319122314453125,
      "learning_rate": 4.993221935559098e-07,
      "loss": 0.0001,
      "reward": 1.800000049173832,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 7120
    },
    {
      "completion_length": 267.1250123977661,
      "epoch": 1.1941405758833143,
      "grad_norm": 0.27828751029224,
      "kl": 0.0604400634765625,
      "learning_rate": 4.993213105327418e-07,
      "loss": 0.0001,
      "reward": 1.7160715088248253,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7294643111526966,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7122
    },
    {
      "completion_length": 252.13840579986572,
      "epoch": 1.194475879123182,
      "grad_norm": 0.25794649504968276,
      "kl": 0.0631561279296875,
      "learning_rate": 4.993204269355438e-07,
      "loss": 0.0001,
      "reward": 1.7071429342031479,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7160714827477932,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7124
    },
    {
      "completion_length": 252.7142972946167,
      "epoch": 1.1948111823630496,
      "grad_norm": 0.13134844037671142,
      "kl": 0.0653228759765625,
      "learning_rate": 4.993195427643176e-07,
      "loss": 0.0001,
      "reward": 1.7392857819795609,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.7482143137603998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7126
    },
    {
      "completion_length": 248.602689743042,
      "epoch": 1.1951464856029173,
      "grad_norm": 0.2285108171615533,
      "kl": 0.06787109375,
      "learning_rate": 4.993186580190655e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7392857521772385,
      "rewards/format_reward_func": 1.0,
      "step": 7128
    },
    {
      "completion_length": 250.45537090301514,
      "epoch": 1.1954817888427847,
      "grad_norm": 0.1980039645739993,
      "kl": 0.06549072265625,
      "learning_rate": 4.993177726997894e-07,
      "loss": 0.0001,
      "reward": 1.7321429252624512,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321429066359997,
      "rewards/format_reward_func": 1.0,
      "step": 7130
    },
    {
      "completion_length": 260.33036518096924,
      "epoch": 1.1958170920826523,
      "grad_norm": 0.22274091348978325,
      "kl": 0.0615234375,
      "learning_rate": 4.993168868064913e-07,
      "loss": 0.0001,
      "reward": 1.751785784959793,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7562500238418579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7132
    },
    {
      "completion_length": 245.57144165039062,
      "epoch": 1.1961523953225197,
      "grad_norm": 0.19250171428754337,
      "kl": 0.0677642822265625,
      "learning_rate": 4.993160003391733e-07,
      "loss": 0.0001,
      "reward": 1.7553571909666061,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7687500268220901,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7134
    },
    {
      "completion_length": 248.99554538726807,
      "epoch": 1.1964876985623873,
      "grad_norm": 0.29356235790356494,
      "kl": 0.069305419921875,
      "learning_rate": 4.993151132978374e-07,
      "loss": 0.0001,
      "reward": 1.7285715118050575,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7285714633762836,
      "rewards/format_reward_func": 1.0,
      "step": 7136
    },
    {
      "completion_length": 253.4062614440918,
      "epoch": 1.196823001802255,
      "grad_norm": 0.19904524202169394,
      "kl": 0.114044189453125,
      "learning_rate": 4.993142256824857e-07,
      "loss": 0.0001,
      "reward": 1.719642922282219,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7241071835160255,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7138
    },
    {
      "completion_length": 251.6071538925171,
      "epoch": 1.1971583050421224,
      "grad_norm": 0.19771127345401926,
      "kl": 0.06500244140625,
      "learning_rate": 4.993133374931203e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714286185801029,
      "rewards/format_reward_func": 1.0,
      "step": 7140
    },
    {
      "completion_length": 246.4955472946167,
      "epoch": 1.19749360828199,
      "grad_norm": 0.29051666126122133,
      "kl": 0.1295928955078125,
      "learning_rate": 4.99312448729743e-07,
      "loss": 0.0001,
      "reward": 1.7625000849366188,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.766964316368103,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7142
    },
    {
      "completion_length": 242.80358505249023,
      "epoch": 1.1978289115218577,
      "grad_norm": 0.2147715894666388,
      "kl": 0.0679779052734375,
      "learning_rate": 4.993115593923561e-07,
      "loss": 0.0001,
      "reward": 1.7660714909434319,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7144
    },
    {
      "completion_length": 245.52679920196533,
      "epoch": 1.198164214761725,
      "grad_norm": 0.2806676529569814,
      "kl": 0.0613861083984375,
      "learning_rate": 4.993106694809615e-07,
      "loss": 0.0001,
      "reward": 1.7357143685221672,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7357143126428127,
      "rewards/format_reward_func": 1.0,
      "step": 7146
    },
    {
      "completion_length": 243.39733219146729,
      "epoch": 1.1984995180015927,
      "grad_norm": 0.1163241116677545,
      "kl": 0.06005859375,
      "learning_rate": 4.993097789955614e-07,
      "loss": 0.0001,
      "reward": 1.7750000730156898,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 7148
    },
    {
      "completion_length": 242.4241189956665,
      "epoch": 1.1988348212414603,
      "grad_norm": 0.2207702446446991,
      "kl": 0.064971923828125,
      "learning_rate": 4.993088879361576e-07,
      "loss": 0.0001,
      "reward": 1.7232143506407738,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7366071753203869,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7150
    },
    {
      "completion_length": 243.4285831451416,
      "epoch": 1.1991701244813278,
      "grad_norm": 0.17856647941258746,
      "kl": 0.06219482421875,
      "learning_rate": 4.993079963027525e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.782142873853445,
      "rewards/format_reward_func": 1.0,
      "step": 7152
    },
    {
      "completion_length": 231.9821538925171,
      "epoch": 1.1995054277211954,
      "grad_norm": 0.31929378908916767,
      "kl": 0.077178955078125,
      "learning_rate": 4.993071040953477e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143223285675,
      "rewards/format_reward_func": 1.0,
      "step": 7154
    },
    {
      "completion_length": 245.165189743042,
      "epoch": 1.1998407309610628,
      "grad_norm": 0.23651149317372253,
      "kl": 0.1234130859375,
      "learning_rate": 4.993062113139457e-07,
      "loss": 0.0001,
      "reward": 1.7482143715023994,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7526786103844643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7156
    },
    {
      "completion_length": 241.07590293884277,
      "epoch": 1.2001760342009304,
      "grad_norm": 0.18797884413854196,
      "kl": 0.0643157958984375,
      "learning_rate": 4.993053179585484e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714600235224,
      "rewards/format_reward_func": 1.0,
      "step": 7158
    },
    {
      "completion_length": 245.80804347991943,
      "epoch": 1.200511337440798,
      "grad_norm": 0.19977985178626378,
      "kl": 0.0759735107421875,
      "learning_rate": 4.993044240291576e-07,
      "loss": 0.0001,
      "reward": 1.7750000804662704,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 7160
    },
    {
      "completion_length": 244.26786994934082,
      "epoch": 1.2008466406806655,
      "grad_norm": 0.27002438185556893,
      "kl": 0.070404052734375,
      "learning_rate": 4.993035295257758e-07,
      "loss": 0.0001,
      "reward": 1.7232143506407738,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.727678619325161,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7162
    },
    {
      "completion_length": 246.47769165039062,
      "epoch": 1.2011819439205331,
      "grad_norm": 0.14532020676846388,
      "kl": 0.063720703125,
      "learning_rate": 4.993026344484047e-07,
      "loss": 0.0001,
      "reward": 1.7553572207689285,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.759821455925703,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7164
    },
    {
      "completion_length": 243.21429824829102,
      "epoch": 1.2015172471604008,
      "grad_norm": 0.32398441312856185,
      "kl": 0.1230926513671875,
      "learning_rate": 4.993017387970467e-07,
      "loss": 0.0001,
      "reward": 1.7071429565548897,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7071429006755352,
      "rewards/format_reward_func": 1.0,
      "step": 7166
    },
    {
      "completion_length": 249.39733600616455,
      "epoch": 1.2018525504002682,
      "grad_norm": 0.16493126908134675,
      "kl": 0.0857696533203125,
      "learning_rate": 4.993008425717034e-07,
      "loss": 0.0001,
      "reward": 1.7375000640749931,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.750892885029316,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7168
    },
    {
      "completion_length": 246.2009048461914,
      "epoch": 1.2021878536401358,
      "grad_norm": 0.17881583185915126,
      "kl": 0.06036376953125,
      "learning_rate": 4.992999457723773e-07,
      "loss": 0.0001,
      "reward": 1.8107143566012383,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8196428716182709,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7170
    },
    {
      "completion_length": 234.98661708831787,
      "epoch": 1.2025231568800034,
      "grad_norm": 0.20858932423195833,
      "kl": 0.067535400390625,
      "learning_rate": 4.992990483990702e-07,
      "loss": 0.0001,
      "reward": 1.7500000596046448,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7500000447034836,
      "rewards/format_reward_func": 1.0,
      "step": 7172
    },
    {
      "completion_length": 254.27233123779297,
      "epoch": 1.2028584601198709,
      "grad_norm": 0.3299755816245421,
      "kl": 0.110748291015625,
      "learning_rate": 4.992981504517843e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 7174
    },
    {
      "completion_length": 249.93305110931396,
      "epoch": 1.2031937633597385,
      "grad_norm": 0.1657871907998651,
      "kl": 0.1009368896484375,
      "learning_rate": 4.992972519305216e-07,
      "loss": 0.0001,
      "reward": 1.7625000774860382,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7176
    },
    {
      "completion_length": 254.86608219146729,
      "epoch": 1.203529066599606,
      "grad_norm": 0.19653879742103245,
      "kl": 0.065093994140625,
      "learning_rate": 4.992963528352843e-07,
      "loss": 0.0001,
      "reward": 1.7625000774860382,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7758928798139095,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7178
    },
    {
      "completion_length": 242.56251049041748,
      "epoch": 1.2038643698394735,
      "grad_norm": 0.22354122095251158,
      "kl": 0.1406707763671875,
      "learning_rate": 4.992954531660742e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143297791481,
      "rewards/format_reward_func": 1.0,
      "step": 7180
    },
    {
      "completion_length": 245.2544765472412,
      "epoch": 1.2041996730793412,
      "grad_norm": 0.2367578009193638,
      "kl": 0.0679779052734375,
      "learning_rate": 4.992945529228937e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321429029107094,
      "rewards/format_reward_func": 1.0,
      "step": 7182
    },
    {
      "completion_length": 257.9464416503906,
      "epoch": 1.2045349763192088,
      "grad_norm": 0.19217014082094047,
      "kl": 0.1060028076171875,
      "learning_rate": 4.992936521057446e-07,
      "loss": 0.0001,
      "reward": 1.7642858102917671,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857357859612,
      "rewards/format_reward_func": 1.0,
      "step": 7184
    },
    {
      "completion_length": 257.2276906967163,
      "epoch": 1.2048702795590762,
      "grad_norm": 0.2761550441553349,
      "kl": 0.1991729736328125,
      "learning_rate": 4.992927507146291e-07,
      "loss": 0.0002,
      "reward": 1.7785715088248253,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7785714659839869,
      "rewards/format_reward_func": 1.0,
      "step": 7186
    },
    {
      "completion_length": 251.54911708831787,
      "epoch": 1.2052055827989439,
      "grad_norm": 0.307690065103978,
      "kl": 0.085052490234375,
      "learning_rate": 4.992918487495492e-07,
      "loss": 0.0001,
      "reward": 1.801785759627819,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.8062500320374966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7188
    },
    {
      "completion_length": 255.63840579986572,
      "epoch": 1.2055408860388113,
      "grad_norm": 0.15497750976828262,
      "kl": 0.0872802734375,
      "learning_rate": 4.992909462105072e-07,
      "loss": 0.0001,
      "reward": 1.7375000640749931,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.74196432903409,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7190
    },
    {
      "completion_length": 256.4151916503906,
      "epoch": 1.205876189278679,
      "grad_norm": 0.22940595595717467,
      "kl": 0.0673065185546875,
      "learning_rate": 4.992900430975048e-07,
      "loss": 0.0001,
      "reward": 1.7553572058677673,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214633762836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7192
    },
    {
      "completion_length": 254.82590675354004,
      "epoch": 1.2062114925185465,
      "grad_norm": 0.2030428580858518,
      "kl": 0.0630340576171875,
      "learning_rate": 4.992891394105445e-07,
      "loss": 0.0001,
      "reward": 1.7357143685221672,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 7194
    },
    {
      "completion_length": 249.6160831451416,
      "epoch": 1.206546795758414,
      "grad_norm": 0.22482483113253907,
      "kl": 0.1019287109375,
      "learning_rate": 4.992882351496281e-07,
      "loss": 0.0001,
      "reward": 1.7732143327593803,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7776786275207996,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7196
    },
    {
      "completion_length": 254.1160831451416,
      "epoch": 1.2068820989982816,
      "grad_norm": 0.12312991086334027,
      "kl": 0.086456298828125,
      "learning_rate": 4.992873303147577e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.025253813713788986,
      "rewards/equation_reward_func": 0.8053571507334709,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7198
    },
    {
      "completion_length": 263.8794775009155,
      "epoch": 1.2072174022381492,
      "grad_norm": 0.10108203752975906,
      "kl": 0.229705810546875,
      "learning_rate": 4.992864249059354e-07,
      "loss": 0.0002,
      "reward": 1.7053572088479996,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7098214738070965,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7200
    },
    {
      "completion_length": 249.06697750091553,
      "epoch": 1.2075527054780166,
      "grad_norm": 0.305712170662913,
      "kl": 0.169677734375,
      "learning_rate": 4.992855189231634e-07,
      "loss": 0.0002,
      "reward": 1.7464286386966705,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 7202
    },
    {
      "completion_length": 260.4241199493408,
      "epoch": 1.2078880087178843,
      "grad_norm": 0.11947172731424033,
      "kl": 0.0626373291015625,
      "learning_rate": 4.992846123664437e-07,
      "loss": 0.0001,
      "reward": 1.8035714849829674,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 7204
    },
    {
      "completion_length": 260.3839445114136,
      "epoch": 1.208223311957752,
      "grad_norm": 0.17124817043019805,
      "kl": 0.0750732421875,
      "learning_rate": 4.992837052357783e-07,
      "loss": 0.0001,
      "reward": 1.7625000700354576,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7669643145054579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7206
    },
    {
      "completion_length": 258.6205463409424,
      "epoch": 1.2085586151976193,
      "grad_norm": 0.20993607262866407,
      "kl": 0.0743255615234375,
      "learning_rate": 4.992827975311695e-07,
      "loss": 0.0001,
      "reward": 1.7625000700354576,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643200933933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7208
    },
    {
      "completion_length": 249.5803689956665,
      "epoch": 1.208893918437487,
      "grad_norm": 0.18415055786807938,
      "kl": 0.0663604736328125,
      "learning_rate": 4.992818892526193e-07,
      "loss": 0.0001,
      "reward": 1.78035718947649,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214507102966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7210
    },
    {
      "completion_length": 252.821439743042,
      "epoch": 1.2092292216773544,
      "grad_norm": 0.1304870095299773,
      "kl": 0.0946197509765625,
      "learning_rate": 4.992809804001296e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8035714626312256,
      "rewards/format_reward_func": 1.0,
      "step": 7212
    },
    {
      "completion_length": 262.9776906967163,
      "epoch": 1.209564524917222,
      "grad_norm": 0.23406135216288343,
      "kl": 0.0801544189453125,
      "learning_rate": 4.992800709737029e-07,
      "loss": 0.0001,
      "reward": 1.7750000804662704,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 7214
    },
    {
      "completion_length": 260.1160840988159,
      "epoch": 1.2098998281570896,
      "grad_norm": 0.1897578285746444,
      "kl": 0.129669189453125,
      "learning_rate": 4.992791609733408e-07,
      "loss": 0.0001,
      "reward": 1.785714328289032,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7946428954601288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7216
    },
    {
      "completion_length": 256.3482275009155,
      "epoch": 1.210235131396957,
      "grad_norm": 0.23263252449293698,
      "kl": 0.1983489990234375,
      "learning_rate": 4.992782503990458e-07,
      "loss": 0.0002,
      "reward": 1.769642911851406,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7218
    },
    {
      "completion_length": 257.5759048461914,
      "epoch": 1.2105704346368247,
      "grad_norm": 0.20205381240686107,
      "kl": 0.0634307861328125,
      "learning_rate": 4.992773392508198e-07,
      "loss": 0.0001,
      "reward": 1.7410715073347092,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7455357499420643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7220
    },
    {
      "completion_length": 266.7098388671875,
      "epoch": 1.2109057378766923,
      "grad_norm": 0.2924996226040406,
      "kl": 0.0713043212890625,
      "learning_rate": 4.99276427528665e-07,
      "loss": 0.0001,
      "reward": 1.796428643167019,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 7222
    },
    {
      "completion_length": 254.8035831451416,
      "epoch": 1.2112410411165597,
      "grad_norm": 0.1814333949957524,
      "kl": 0.100006103515625,
      "learning_rate": 4.992755152325833e-07,
      "loss": 0.0001,
      "reward": 1.744642935693264,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071578115225,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7224
    },
    {
      "completion_length": 257.64287185668945,
      "epoch": 1.2115763443564274,
      "grad_norm": 0.24674960296846465,
      "kl": 0.1389617919921875,
      "learning_rate": 4.99274602362577e-07,
      "loss": 0.0001,
      "reward": 1.7660714909434319,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7705357596278191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7226
    },
    {
      "completion_length": 249.9107265472412,
      "epoch": 1.211911647596295,
      "grad_norm": 0.17770874745401008,
      "kl": 0.068023681640625,
      "learning_rate": 4.992736889186482e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857521772385,
      "rewards/format_reward_func": 1.0,
      "step": 7228
    },
    {
      "completion_length": 256.6250123977661,
      "epoch": 1.2122469508361624,
      "grad_norm": 0.2509594504708243,
      "kl": 0.1199951171875,
      "learning_rate": 4.992727749007988e-07,
      "loss": 0.0001,
      "reward": 1.7267857864499092,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7312500402331352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7230
    },
    {
      "completion_length": 267.089298248291,
      "epoch": 1.21258225407603,
      "grad_norm": 0.22356860342543525,
      "kl": 0.0696563720703125,
      "learning_rate": 4.992718603090312e-07,
      "loss": 0.0001,
      "reward": 1.7178572043776512,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7178571987897158,
      "rewards/format_reward_func": 1.0,
      "step": 7232
    },
    {
      "completion_length": 263.86608123779297,
      "epoch": 1.2129175573158975,
      "grad_norm": 0.4141292974356211,
      "kl": 0.0949554443359375,
      "learning_rate": 4.992709451433473e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.09091372694820166,
      "rewards/equation_reward_func": 0.801785733550787,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7234
    },
    {
      "completion_length": 259.0000123977661,
      "epoch": 1.213252860555765,
      "grad_norm": 0.2666180229355076,
      "kl": 0.1144866943359375,
      "learning_rate": 4.992700294037493e-07,
      "loss": 0.0001,
      "reward": 1.7214286625385284,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7214285954833031,
      "rewards/format_reward_func": 1.0,
      "step": 7236
    },
    {
      "completion_length": 252.33929538726807,
      "epoch": 1.2135881637956327,
      "grad_norm": 0.08668010444846727,
      "kl": 0.1309967041015625,
      "learning_rate": 4.992691130902392e-07,
      "loss": 0.0001,
      "reward": 1.7875000834465027,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7919643186032772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7238
    },
    {
      "completion_length": 255.44197368621826,
      "epoch": 1.2139234670355001,
      "grad_norm": 0.09888979720455138,
      "kl": 0.122528076171875,
      "learning_rate": 4.992681962028193e-07,
      "loss": 0.0001,
      "reward": 1.860714316368103,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8607143089175224,
      "rewards/format_reward_func": 1.0,
      "step": 7240
    },
    {
      "completion_length": 248.6830472946167,
      "epoch": 1.2142587702753678,
      "grad_norm": 0.18925414301509572,
      "kl": 0.1031341552734375,
      "learning_rate": 4.992672787414914e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714693367481,
      "rewards/format_reward_func": 1.0,
      "step": 7242
    },
    {
      "completion_length": 258.4464416503906,
      "epoch": 1.2145940735152354,
      "grad_norm": 0.2640904385696533,
      "kl": 0.074462890625,
      "learning_rate": 4.99266360706258e-07,
      "loss": 0.0001,
      "reward": 1.7892857939004898,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 7244
    },
    {
      "completion_length": 251.86162185668945,
      "epoch": 1.2149293767551028,
      "grad_norm": 0.31212532650075375,
      "kl": 0.06591796875,
      "learning_rate": 4.992654420971209e-07,
      "loss": 0.0001,
      "reward": 1.8071429133415222,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428798139095,
      "rewards/format_reward_func": 1.0,
      "step": 7246
    },
    {
      "completion_length": 261.6875114440918,
      "epoch": 1.2152646799949705,
      "grad_norm": 0.08009381221104957,
      "kl": 0.072265625,
      "learning_rate": 4.992645229140824e-07,
      "loss": 0.0001,
      "reward": 1.775000050663948,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 7248
    },
    {
      "completion_length": 255.63394260406494,
      "epoch": 1.215599983234838,
      "grad_norm": 0.16946388110095856,
      "kl": 0.06854248046875,
      "learning_rate": 4.992636031571444e-07,
      "loss": 0.0001,
      "reward": 1.6964286640286446,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.696428619325161,
      "rewards/format_reward_func": 1.0,
      "step": 7250
    },
    {
      "completion_length": 254.97769165039062,
      "epoch": 1.2159352864747055,
      "grad_norm": 0.23666184664199402,
      "kl": 0.078399658203125,
      "learning_rate": 4.992626828263093e-07,
      "loss": 0.0001,
      "reward": 1.7392857819795609,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 1.0,
      "step": 7252
    },
    {
      "completion_length": 253.17858028411865,
      "epoch": 1.2162705897145731,
      "grad_norm": 0.008383486510034467,
      "kl": 0.1063690185546875,
      "learning_rate": 4.992617619215791e-07,
      "loss": 0.0001,
      "reward": 1.753571480512619,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714600235224,
      "rewards/format_reward_func": 1.0,
      "step": 7254
    },
    {
      "completion_length": 262.36608505249023,
      "epoch": 1.2166058929544405,
      "grad_norm": 0.23928214815805107,
      "kl": 0.1030426025390625,
      "learning_rate": 4.992608404429558e-07,
      "loss": 0.0001,
      "reward": 1.7330357879400253,
      "reward_std": 0.044194172602146864,
      "rewards/equation_reward_func": 0.7348214592784643,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 7256
    },
    {
      "completion_length": 254.133939743042,
      "epoch": 1.2169411961943082,
      "grad_norm": 0.16791549556043,
      "kl": 0.0996551513671875,
      "learning_rate": 4.992599183904417e-07,
      "loss": 0.0001,
      "reward": 1.7232143580913544,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7276786006987095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7258
    },
    {
      "completion_length": 255.5178689956665,
      "epoch": 1.2172764994341758,
      "grad_norm": 0.25849978845251953,
      "kl": 0.084625244140625,
      "learning_rate": 4.992589957640388e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7517857402563095,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7260
    },
    {
      "completion_length": 254.0759048461914,
      "epoch": 1.2176118026740435,
      "grad_norm": 0.18565102559484725,
      "kl": 0.154052734375,
      "learning_rate": 4.992580725637494e-07,
      "loss": 0.0002,
      "reward": 1.7370536476373672,
      "reward_std": 0.0587151157669723,
      "rewards/equation_reward_func": 0.7446428835391998,
      "rewards/format_reward_func": 0.9924107193946838,
      "step": 7262
    },
    {
      "completion_length": 259.30358505249023,
      "epoch": 1.2179471059139109,
      "grad_norm": 0.21800854720790241,
      "kl": 0.1430206298828125,
      "learning_rate": 4.992571487895753e-07,
      "loss": 0.0001,
      "reward": 1.7232143431901932,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7276785969734192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7264
    },
    {
      "completion_length": 250.0759048461914,
      "epoch": 1.2182824091537785,
      "grad_norm": 0.1164286611566817,
      "kl": 0.0711212158203125,
      "learning_rate": 4.99256224441519e-07,
      "loss": 0.0001,
      "reward": 1.7392857521772385,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857633531094,
      "rewards/format_reward_func": 1.0,
      "step": 7266
    },
    {
      "completion_length": 264.60715198516846,
      "epoch": 1.218617712393646,
      "grad_norm": 0.24969613066518914,
      "kl": 0.062957763671875,
      "learning_rate": 4.992552995195825e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000279396772,
      "rewards/format_reward_func": 1.0,
      "step": 7268
    },
    {
      "completion_length": 255.99554920196533,
      "epoch": 1.2189530156335135,
      "grad_norm": 0.19538401377886458,
      "kl": 0.0625152587890625,
      "learning_rate": 4.992543740237677e-07,
      "loss": 0.0001,
      "reward": 1.8339286297559738,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.8383928760886192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7270
    },
    {
      "completion_length": 250.9642972946167,
      "epoch": 1.2192883188733812,
      "grad_norm": 0.23034703579469626,
      "kl": 0.0625762939453125,
      "learning_rate": 4.99253447954077e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714752972126,
      "rewards/format_reward_func": 1.0,
      "step": 7272
    },
    {
      "completion_length": 263.1696557998657,
      "epoch": 1.2196236221132486,
      "grad_norm": 0.19438148140689746,
      "kl": 0.109344482421875,
      "learning_rate": 4.992525213105124e-07,
      "loss": 0.0001,
      "reward": 1.7178572416305542,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7267857454717159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7274
    },
    {
      "completion_length": 266.7678680419922,
      "epoch": 1.2199589253531162,
      "grad_norm": 0.20735329597287194,
      "kl": 0.077850341796875,
      "learning_rate": 4.992515940930762e-07,
      "loss": 0.0001,
      "reward": 1.7357143461704254,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143238186836,
      "rewards/format_reward_func": 1.0,
      "step": 7276
    },
    {
      "completion_length": 258.3035840988159,
      "epoch": 1.2202942285929839,
      "grad_norm": 0.14736460091258483,
      "kl": 0.1506500244140625,
      "learning_rate": 4.992506663017702e-07,
      "loss": 0.0002,
      "reward": 1.742857240140438,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 7278
    },
    {
      "completion_length": 248.61608409881592,
      "epoch": 1.2206295318328513,
      "grad_norm": 0.2677453481686041,
      "kl": 0.126617431640625,
      "learning_rate": 4.99249737936597e-07,
      "loss": 0.0001,
      "reward": 1.762500062584877,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7669643051922321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7280
    },
    {
      "completion_length": 259.7901887893677,
      "epoch": 1.220964835072719,
      "grad_norm": 0.492668665701258,
      "kl": 0.084564208984375,
      "learning_rate": 4.992488089975583e-07,
      "loss": 0.0001,
      "reward": 1.7160715237259865,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7205357477068901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7282
    },
    {
      "completion_length": 264.9196548461914,
      "epoch": 1.2213001383125865,
      "grad_norm": 0.181095995568355,
      "kl": 0.087677001953125,
      "learning_rate": 4.992478794846565e-07,
      "loss": 0.0001,
      "reward": 1.723214328289032,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7366071715950966,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7284
    },
    {
      "completion_length": 267.1651906967163,
      "epoch": 1.221635441552454,
      "grad_norm": 0.2169639788945554,
      "kl": 0.0669708251953125,
      "learning_rate": 4.992469493978937e-07,
      "loss": 0.0001,
      "reward": 1.7946429178118706,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7991071790456772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7286
    },
    {
      "completion_length": 256.1875123977661,
      "epoch": 1.2219707447923216,
      "grad_norm": 0.6186112811288419,
      "kl": 0.09869384765625,
      "learning_rate": 4.992460187372719e-07,
      "loss": 0.0001,
      "reward": 1.7178572192788124,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7178571857511997,
      "rewards/format_reward_func": 1.0,
      "step": 7288
    },
    {
      "completion_length": 260.57590103149414,
      "epoch": 1.222306048032189,
      "grad_norm": 0.16121117399200693,
      "kl": 0.0945587158203125,
      "learning_rate": 4.992450875027935e-07,
      "loss": 0.0001,
      "reward": 1.7160715088248253,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7205357514321804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7290
    },
    {
      "completion_length": 260.7410879135132,
      "epoch": 1.2226413512720566,
      "grad_norm": 0.22727721872779744,
      "kl": 0.06622314453125,
      "learning_rate": 4.992441556944604e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143331319094,
      "rewards/format_reward_func": 1.0,
      "step": 7292
    },
    {
      "completion_length": 259.4107265472412,
      "epoch": 1.2229766545119243,
      "grad_norm": 0.24415005335576428,
      "kl": 0.1076812744140625,
      "learning_rate": 4.992432233122749e-07,
      "loss": 0.0001,
      "reward": 1.7607143744826317,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7607143204659224,
      "rewards/format_reward_func": 1.0,
      "step": 7294
    },
    {
      "completion_length": 257.56697845458984,
      "epoch": 1.2233119577517917,
      "grad_norm": 0.31739442791663675,
      "kl": 0.0947113037109375,
      "learning_rate": 4.99242290356239e-07,
      "loss": 0.0001,
      "reward": 1.7678571939468384,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 7296
    },
    {
      "completion_length": 270.28126430511475,
      "epoch": 1.2236472609916593,
      "grad_norm": 0.1385651283554177,
      "kl": 0.0988616943359375,
      "learning_rate": 4.99241356826355e-07,
      "loss": 0.0001,
      "reward": 1.7785715162754059,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714454948902,
      "rewards/format_reward_func": 1.0,
      "step": 7298
    },
    {
      "completion_length": 266.92857933044434,
      "epoch": 1.223982564231527,
      "grad_norm": 0.2780453894697533,
      "kl": 0.0828094482421875,
      "learning_rate": 4.992404227226249e-07,
      "loss": 0.0001,
      "reward": 1.7553571984171867,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.759821455925703,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7300
    },
    {
      "completion_length": 266.64733123779297,
      "epoch": 1.2243178674713944,
      "grad_norm": 0.1710191122109286,
      "kl": 0.0935516357421875,
      "learning_rate": 4.99239488045051e-07,
      "loss": 0.0001,
      "reward": 1.7446429505944252,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7302
    },
    {
      "completion_length": 269.4509057998657,
      "epoch": 1.224653170711262,
      "grad_norm": 0.24567936420836164,
      "kl": 0.071990966796875,
      "learning_rate": 4.992385527936354e-07,
      "loss": 0.0001,
      "reward": 1.7714286521077156,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7803571820259094,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7304
    },
    {
      "completion_length": 267.95090770721436,
      "epoch": 1.2249884739511296,
      "grad_norm": 0.2317099030341503,
      "kl": 0.098785400390625,
      "learning_rate": 4.992376169683803e-07,
      "loss": 0.0001,
      "reward": 1.8071429207921028,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428872644901,
      "rewards/format_reward_func": 1.0,
      "step": 7306
    },
    {
      "completion_length": 262.07590675354004,
      "epoch": 1.225323777190997,
      "grad_norm": 0.21304988210294232,
      "kl": 0.065704345703125,
      "learning_rate": 4.992366805692877e-07,
      "loss": 0.0001,
      "reward": 1.7696429193019867,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7308
    },
    {
      "completion_length": 267.8169775009155,
      "epoch": 1.2256590804308647,
      "grad_norm": 0.226278049448281,
      "kl": 0.0792388916015625,
      "learning_rate": 4.992357435963599e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7696428820490837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7310
    },
    {
      "completion_length": 261.1964454650879,
      "epoch": 1.225994383670732,
      "grad_norm": 0.24893292964449734,
      "kl": 0.0957183837890625,
      "learning_rate": 4.992348060495989e-07,
      "loss": 0.0001,
      "reward": 1.783928632736206,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7883928716182709,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7312
    },
    {
      "completion_length": 266.98662185668945,
      "epoch": 1.2263296869105997,
      "grad_norm": 0.24435019680281253,
      "kl": 0.075714111328125,
      "learning_rate": 4.99233867929007e-07,
      "loss": 0.0001,
      "reward": 1.7250000834465027,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7250000312924385,
      "rewards/format_reward_func": 1.0,
      "step": 7314
    },
    {
      "completion_length": 245.55358219146729,
      "epoch": 1.2266649901504674,
      "grad_norm": 0.174986762818187,
      "kl": 0.0735015869140625,
      "learning_rate": 4.992329292345865e-07,
      "loss": 0.0001,
      "reward": 1.7428572252392769,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7428571656346321,
      "rewards/format_reward_func": 1.0,
      "step": 7316
    },
    {
      "completion_length": 262.0759029388428,
      "epoch": 1.227000293390335,
      "grad_norm": 0.05003726187541759,
      "kl": 0.06793212890625,
      "learning_rate": 4.992319899663391e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.05555839091539383,
      "rewards/equation_reward_func": 0.7839285992085934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7318
    },
    {
      "completion_length": 249.6071538925171,
      "epoch": 1.2273355966302024,
      "grad_norm": 0.0020244057203651937,
      "kl": 0.06622314453125,
      "learning_rate": 4.992310501242676e-07,
      "loss": 0.0001,
      "reward": 1.8142857551574707,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.8142857402563095,
      "rewards/format_reward_func": 1.0,
      "step": 7320
    },
    {
      "completion_length": 260.71430110931396,
      "epoch": 1.22767089987007,
      "grad_norm": 0.20053090389361175,
      "kl": 0.091400146484375,
      "learning_rate": 4.992301097083735e-07,
      "loss": 0.0001,
      "reward": 1.732142947614193,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321428842842579,
      "rewards/format_reward_func": 1.0,
      "step": 7322
    },
    {
      "completion_length": 261.8348331451416,
      "epoch": 1.2280062031099375,
      "grad_norm": 0.2737092773881806,
      "kl": 0.0717315673828125,
      "learning_rate": 4.992291687186595e-07,
      "loss": 0.0001,
      "reward": 1.7285715192556381,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7285714671015739,
      "rewards/format_reward_func": 1.0,
      "step": 7324
    },
    {
      "completion_length": 256.0134057998657,
      "epoch": 1.228341506349805,
      "grad_norm": 0.33311744361146584,
      "kl": 0.130706787109375,
      "learning_rate": 4.992282271551274e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 7326
    },
    {
      "completion_length": 249.45536994934082,
      "epoch": 1.2286768095896727,
      "grad_norm": 0.3027524607523363,
      "kl": 0.1055755615234375,
      "learning_rate": 4.992272850177795e-07,
      "loss": 0.0001,
      "reward": 1.8089286237955093,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8133928813040257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7328
    },
    {
      "completion_length": 257.183048248291,
      "epoch": 1.2290121128295401,
      "grad_norm": 0.2134971880773326,
      "kl": 0.068511962890625,
      "learning_rate": 4.992263423066182e-07,
      "loss": 0.0001,
      "reward": 1.7410715073347092,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7455357573926449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7330
    },
    {
      "completion_length": 265.4241189956665,
      "epoch": 1.2293474160694078,
      "grad_norm": 0.09134287624357415,
      "kl": 0.0912628173828125,
      "learning_rate": 4.992253990216453e-07,
      "loss": 0.0001,
      "reward": 1.7839286550879478,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7883928641676903,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7332
    },
    {
      "completion_length": 255.87054634094238,
      "epoch": 1.2296827193092754,
      "grad_norm": 0.2729861269599726,
      "kl": 0.0726165771484375,
      "learning_rate": 4.992244551628631e-07,
      "loss": 0.0001,
      "reward": 1.7142857909202576,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7142857499420643,
      "rewards/format_reward_func": 1.0,
      "step": 7334
    },
    {
      "completion_length": 243.59375858306885,
      "epoch": 1.2300180225491428,
      "grad_norm": 0.12997354741489184,
      "kl": 0.0819549560546875,
      "learning_rate": 4.992235107302738e-07,
      "loss": 0.0001,
      "reward": 1.7071429044008255,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7071428969502449,
      "rewards/format_reward_func": 1.0,
      "step": 7336
    },
    {
      "completion_length": 253.1562623977661,
      "epoch": 1.2303533257890105,
      "grad_norm": 0.13434226070764227,
      "kl": 0.089324951171875,
      "learning_rate": 4.992225657238797e-07,
      "loss": 0.0001,
      "reward": 1.7000000923871994,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7000000346451998,
      "rewards/format_reward_func": 1.0,
      "step": 7338
    },
    {
      "completion_length": 242.4419765472412,
      "epoch": 1.230688629028878,
      "grad_norm": 0.1688100720296217,
      "kl": 0.079833984375,
      "learning_rate": 4.992216201436827e-07,
      "loss": 0.0001,
      "reward": 1.739285796880722,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857428640127,
      "rewards/format_reward_func": 1.0,
      "step": 7340
    },
    {
      "completion_length": 240.30358123779297,
      "epoch": 1.2310239322687455,
      "grad_norm": 0.22757167938869166,
      "kl": 0.0682830810546875,
      "learning_rate": 4.992206739896851e-07,
      "loss": 0.0001,
      "reward": 1.8214286416769028,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8214285932481289,
      "rewards/format_reward_func": 1.0,
      "step": 7342
    },
    {
      "completion_length": 256.2857265472412,
      "epoch": 1.2313592355086131,
      "grad_norm": 0.24103032576905367,
      "kl": 0.0616302490234375,
      "learning_rate": 4.992197272618893e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 7344
    },
    {
      "completion_length": 249.7500114440918,
      "epoch": 1.2316945387484806,
      "grad_norm": 0.22890032159343376,
      "kl": 0.084991455078125,
      "learning_rate": 4.992187799602972e-07,
      "loss": 0.0001,
      "reward": 1.6910715252161026,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.6955357491970062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7346
    },
    {
      "completion_length": 239.84376335144043,
      "epoch": 1.2320298419883482,
      "grad_norm": 0.11993653416671786,
      "kl": 0.0696258544921875,
      "learning_rate": 4.992178320849109e-07,
      "loss": 0.0001,
      "reward": 1.714285783469677,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7142857611179352,
      "rewards/format_reward_func": 1.0,
      "step": 7348
    },
    {
      "completion_length": 229.56697463989258,
      "epoch": 1.2323651452282158,
      "grad_norm": 0.283880895706985,
      "kl": 0.1734161376953125,
      "learning_rate": 4.992168836357329e-07,
      "loss": 0.0002,
      "reward": 1.8250000551342964,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8250000290572643,
      "rewards/format_reward_func": 1.0,
      "step": 7350
    },
    {
      "completion_length": 243.64733219146729,
      "epoch": 1.2327004484680832,
      "grad_norm": 0.2535059072968625,
      "kl": 0.0676727294921875,
      "learning_rate": 4.992159346127652e-07,
      "loss": 0.0001,
      "reward": 1.7464286461472511,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 7352
    },
    {
      "completion_length": 237.9464406967163,
      "epoch": 1.2330357517079509,
      "grad_norm": 0.3366521986910057,
      "kl": 0.0709991455078125,
      "learning_rate": 4.9921498501601e-07,
      "loss": 0.0001,
      "reward": 1.7035715356469154,
      "reward_std": 0.0858629634603858,
      "rewards/equation_reward_func": 0.7035714499652386,
      "rewards/format_reward_func": 1.0,
      "step": 7354
    },
    {
      "completion_length": 230.95983219146729,
      "epoch": 1.2333710549478185,
      "grad_norm": 0.3199696882096981,
      "kl": 0.098846435546875,
      "learning_rate": 4.992140348454695e-07,
      "loss": 0.0001,
      "reward": 1.728571504354477,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7285714503377676,
      "rewards/format_reward_func": 1.0,
      "step": 7356
    },
    {
      "completion_length": 232.62054824829102,
      "epoch": 1.233706358187686,
      "grad_norm": 0.20618256158180728,
      "kl": 0.0677337646484375,
      "learning_rate": 4.992130841011461e-07,
      "loss": 0.0001,
      "reward": 1.725000075995922,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000257045031,
      "rewards/format_reward_func": 1.0,
      "step": 7358
    },
    {
      "completion_length": 226.79018878936768,
      "epoch": 1.2340416614275536,
      "grad_norm": 0.24268664351870567,
      "kl": 0.0743560791015625,
      "learning_rate": 4.992121327830415e-07,
      "loss": 0.0001,
      "reward": 1.8071429133415222,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428760886192,
      "rewards/format_reward_func": 1.0,
      "step": 7360
    },
    {
      "completion_length": 229.5535831451416,
      "epoch": 1.2343769646674212,
      "grad_norm": 0.21594958900650493,
      "kl": 0.066436767578125,
      "learning_rate": 4.992111808911583e-07,
      "loss": 0.0001,
      "reward": 1.775000050663948,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000543892384,
      "rewards/format_reward_func": 1.0,
      "step": 7362
    },
    {
      "completion_length": 237.1830472946167,
      "epoch": 1.2347122679072886,
      "grad_norm": 0.2437006081039398,
      "kl": 0.0667266845703125,
      "learning_rate": 4.992102284254985e-07,
      "loss": 0.0001,
      "reward": 1.717857226729393,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7178571671247482,
      "rewards/format_reward_func": 1.0,
      "step": 7364
    },
    {
      "completion_length": 232.06697463989258,
      "epoch": 1.2350475711471562,
      "grad_norm": 0.21956220909973093,
      "kl": 0.06585693359375,
      "learning_rate": 4.992092753860644e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8107143044471741,
      "rewards/format_reward_func": 1.0,
      "step": 7366
    },
    {
      "completion_length": 231.67858123779297,
      "epoch": 1.2353828743870237,
      "grad_norm": 0.22599447075235282,
      "kl": 0.0912017822265625,
      "learning_rate": 4.992083217728581e-07,
      "loss": 0.0001,
      "reward": 1.775000087916851,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 7368
    },
    {
      "completion_length": 221.7634038925171,
      "epoch": 1.2357181776268913,
      "grad_norm": 0.2089522879086932,
      "kl": 0.0640411376953125,
      "learning_rate": 4.992073675858818e-07,
      "loss": 0.0001,
      "reward": 1.7678572311997414,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571604192257,
      "rewards/format_reward_func": 1.0,
      "step": 7370
    },
    {
      "completion_length": 225.31251049041748,
      "epoch": 1.236053480866759,
      "grad_norm": 0.27154299086942546,
      "kl": 0.0687255859375,
      "learning_rate": 4.992064128251379e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7392857633531094,
      "rewards/format_reward_func": 1.0,
      "step": 7372
    },
    {
      "completion_length": 231.61608219146729,
      "epoch": 1.2363887841066263,
      "grad_norm": 0.2652996298189875,
      "kl": 0.061309814453125,
      "learning_rate": 4.992054574906284e-07,
      "loss": 0.0001,
      "reward": 1.7482143491506577,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7374
    },
    {
      "completion_length": 231.45983409881592,
      "epoch": 1.236724087346494,
      "grad_norm": 0.1997123223499637,
      "kl": 0.067901611328125,
      "learning_rate": 4.992045015823555e-07,
      "loss": 0.0001,
      "reward": 1.735714353621006,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7357143312692642,
      "rewards/format_reward_func": 1.0,
      "step": 7376
    },
    {
      "completion_length": 226.44643878936768,
      "epoch": 1.2370593905863616,
      "grad_norm": 0.22936699167594957,
      "kl": 0.0925445556640625,
      "learning_rate": 4.992035451003214e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 7378
    },
    {
      "completion_length": 226.43304538726807,
      "epoch": 1.237394693826229,
      "grad_norm": 0.3561236632263674,
      "kl": 0.072845458984375,
      "learning_rate": 4.992025880445284e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.782142873853445,
      "rewards/format_reward_func": 1.0,
      "step": 7380
    },
    {
      "completion_length": 225.39733123779297,
      "epoch": 1.2377299970660967,
      "grad_norm": 0.13552781415865434,
      "kl": 0.06146240234375,
      "learning_rate": 4.992016304149786e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 7382
    },
    {
      "completion_length": 225.352689743042,
      "epoch": 1.2380653003059643,
      "grad_norm": 0.16034323470596654,
      "kl": 0.0637054443359375,
      "learning_rate": 4.992006722116743e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.771428594365716,
      "rewards/format_reward_func": 1.0,
      "step": 7384
    },
    {
      "completion_length": 234.2009038925171,
      "epoch": 1.2384006035458317,
      "grad_norm": 0.11643361950936694,
      "kl": 0.06671142578125,
      "learning_rate": 4.991997134346176e-07,
      "loss": 0.0001,
      "reward": 1.725000061094761,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7250000424683094,
      "rewards/format_reward_func": 1.0,
      "step": 7386
    },
    {
      "completion_length": 237.43304538726807,
      "epoch": 1.2387359067856993,
      "grad_norm": 0.34066513169603546,
      "kl": 0.0608978271484375,
      "learning_rate": 4.991987540838108e-07,
      "loss": 0.0001,
      "reward": 1.7464286610484123,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 7388
    },
    {
      "completion_length": 227.40626049041748,
      "epoch": 1.2390712100255667,
      "grad_norm": 0.14493342284575672,
      "kl": 0.064849853515625,
      "learning_rate": 4.99197794159256e-07,
      "loss": 0.0001,
      "reward": 1.8142857551574707,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857439815998,
      "rewards/format_reward_func": 1.0,
      "step": 7390
    },
    {
      "completion_length": 232.2991189956665,
      "epoch": 1.2394065132654344,
      "grad_norm": 0.27793165057230773,
      "kl": 0.067626953125,
      "learning_rate": 4.991968336609556e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 7392
    },
    {
      "completion_length": 225.19643878936768,
      "epoch": 1.239741816505302,
      "grad_norm": 0.2876380698092266,
      "kl": 0.07135009765625,
      "learning_rate": 4.991958725889116e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 7394
    },
    {
      "completion_length": 229.30358123779297,
      "epoch": 1.2400771197451697,
      "grad_norm": 0.21288203957645194,
      "kl": 0.064697265625,
      "learning_rate": 4.991949109431264e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7357143089175224,
      "rewards/format_reward_func": 1.0,
      "step": 7396
    },
    {
      "completion_length": 221.82143783569336,
      "epoch": 1.240412422985037,
      "grad_norm": 0.10024101777540191,
      "kl": 0.0673828125,
      "learning_rate": 4.99193948723602e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7571428883820772,
      "rewards/format_reward_func": 1.0,
      "step": 7398
    },
    {
      "completion_length": 226.58036708831787,
      "epoch": 1.2407477262249047,
      "grad_norm": 0.22284009957725762,
      "kl": 0.0872955322265625,
      "learning_rate": 4.991929859303408e-07,
      "loss": 0.0001,
      "reward": 1.710714377462864,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7107143141329288,
      "rewards/format_reward_func": 1.0,
      "step": 7400
    },
    {
      "completion_length": 223.60268783569336,
      "epoch": 1.2410830294647721,
      "grad_norm": 0.2600750417552333,
      "kl": 0.06585693359375,
      "learning_rate": 4.99192022563345e-07,
      "loss": 0.0001,
      "reward": 1.79642865806818,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 7402
    },
    {
      "completion_length": 227.6919755935669,
      "epoch": 1.2414183327046397,
      "grad_norm": 0.26431094508941017,
      "kl": 0.0618896484375,
      "learning_rate": 4.991910586226166e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857447266579,
      "rewards/format_reward_func": 1.0,
      "step": 7404
    },
    {
      "completion_length": 221.63840198516846,
      "epoch": 1.2417536359445074,
      "grad_norm": 0.34389640598822085,
      "kl": 0.0604095458984375,
      "learning_rate": 4.991900941081583e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 7406
    },
    {
      "completion_length": 225.4419755935669,
      "epoch": 1.2420889391843748,
      "grad_norm": 0.38325653075172783,
      "kl": 0.096160888671875,
      "learning_rate": 4.991891290199717e-07,
      "loss": 0.0001,
      "reward": 1.8071429058909416,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8071428909897804,
      "rewards/format_reward_func": 1.0,
      "step": 7408
    },
    {
      "completion_length": 223.94197463989258,
      "epoch": 1.2424242424242424,
      "grad_norm": 0.16090653148401493,
      "kl": 0.065399169921875,
      "learning_rate": 4.991881633580594e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 7410
    },
    {
      "completion_length": 228.5000123977661,
      "epoch": 1.24275954566411,
      "grad_norm": 0.28773662489496515,
      "kl": 0.074371337890625,
      "learning_rate": 4.991871971224237e-07,
      "loss": 0.0001,
      "reward": 1.7665179297327995,
      "reward_std": 0.0776554741896689,
      "rewards/equation_reward_func": 0.7678571846336126,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 7412
    },
    {
      "completion_length": 230.33929634094238,
      "epoch": 1.2430948489039775,
      "grad_norm": 0.16326829694279263,
      "kl": 0.0712127685546875,
      "learning_rate": 4.991862303130666e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 7414
    },
    {
      "completion_length": 225.54018783569336,
      "epoch": 1.2434301521438451,
      "grad_norm": 0.14390607940358557,
      "kl": 0.0926513671875,
      "learning_rate": 4.991852629299904e-07,
      "loss": 0.0001,
      "reward": 1.800000049173832,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 7416
    },
    {
      "completion_length": 222.42858123779297,
      "epoch": 1.2437654553837127,
      "grad_norm": 0.22509298933939004,
      "kl": 0.0763702392578125,
      "learning_rate": 4.991842949731974e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 7418
    },
    {
      "completion_length": 221.15179538726807,
      "epoch": 1.2441007586235802,
      "grad_norm": 0.28116774174842096,
      "kl": 0.0690460205078125,
      "learning_rate": 4.991833264426896e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 7420
    },
    {
      "completion_length": 218.59376049041748,
      "epoch": 1.2444360618634478,
      "grad_norm": 0.2843754473417763,
      "kl": 0.06597900390625,
      "learning_rate": 4.991823573384695e-07,
      "loss": 0.0001,
      "reward": 1.8178572058677673,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8178571686148643,
      "rewards/format_reward_func": 1.0,
      "step": 7422
    },
    {
      "completion_length": 222.20983219146729,
      "epoch": 1.2447713651033152,
      "grad_norm": 0.20690636300290868,
      "kl": 0.0728607177734375,
      "learning_rate": 4.991813876605392e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 7424
    },
    {
      "completion_length": 218.0178680419922,
      "epoch": 1.2451066683431828,
      "grad_norm": 0.13231542882177258,
      "kl": 0.06787109375,
      "learning_rate": 4.99180417408901e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 7426
    },
    {
      "completion_length": 221.59375667572021,
      "epoch": 1.2454419715830505,
      "grad_norm": 0.33948420671007135,
      "kl": 0.0762939453125,
      "learning_rate": 4.99179446583557e-07,
      "loss": 0.0001,
      "reward": 1.807142935693264,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.8071428760886192,
      "rewards/format_reward_func": 1.0,
      "step": 7428
    },
    {
      "completion_length": 222.44197463989258,
      "epoch": 1.245777274822918,
      "grad_norm": 0.18968566266064457,
      "kl": 0.0654754638671875,
      "learning_rate": 4.991784751845096e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964286208152771,
      "rewards/format_reward_func": 1.0,
      "step": 7430
    },
    {
      "completion_length": 221.1741180419922,
      "epoch": 1.2461125780627855,
      "grad_norm": 0.13362021561969276,
      "kl": 0.0767059326171875,
      "learning_rate": 4.99177503211761e-07,
      "loss": 0.0001,
      "reward": 1.82857146859169,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8285714574158192,
      "rewards/format_reward_func": 1.0,
      "step": 7432
    },
    {
      "completion_length": 220.6696538925171,
      "epoch": 1.2464478813026532,
      "grad_norm": 0.40983168744085197,
      "kl": 0.0979766845703125,
      "learning_rate": 4.991765306653133e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 7434
    },
    {
      "completion_length": 217.53572273254395,
      "epoch": 1.2467831845425206,
      "grad_norm": 0.24499121073236454,
      "kl": 0.073272705078125,
      "learning_rate": 4.991755575451689e-07,
      "loss": 0.0001,
      "reward": 1.7750000432133675,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 7436
    },
    {
      "completion_length": 215.0267972946167,
      "epoch": 1.2471184877823882,
      "grad_norm": 0.246875958035262,
      "kl": 0.086761474609375,
      "learning_rate": 4.991745838513299e-07,
      "loss": 0.0001,
      "reward": 1.8250000476837158,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.8250000178813934,
      "rewards/format_reward_func": 1.0,
      "step": 7438
    },
    {
      "completion_length": 212.84822463989258,
      "epoch": 1.2474537910222558,
      "grad_norm": 0.20762540243109817,
      "kl": 0.0724029541015625,
      "learning_rate": 4.991736095837987e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 7440
    },
    {
      "completion_length": 221.09822463989258,
      "epoch": 1.2477890942621233,
      "grad_norm": 0.1869219659646194,
      "kl": 0.0802001953125,
      "learning_rate": 4.991726347425774e-07,
      "loss": 0.0001,
      "reward": 1.7392857521772385,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857559025288,
      "rewards/format_reward_func": 1.0,
      "step": 7442
    },
    {
      "completion_length": 232.01340293884277,
      "epoch": 1.248124397501991,
      "grad_norm": 0.27557571789481994,
      "kl": 0.072235107421875,
      "learning_rate": 4.991716593276684e-07,
      "loss": 0.0001,
      "reward": 1.7839286252856255,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7883928865194321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7444
    },
    {
      "completion_length": 220.5937614440918,
      "epoch": 1.2484597007418583,
      "grad_norm": 0.18519122568768503,
      "kl": 0.07379150390625,
      "learning_rate": 4.991706833390738e-07,
      "loss": 0.0001,
      "reward": 1.8392857536673546,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.839285746216774,
      "rewards/format_reward_func": 1.0,
      "step": 7446
    },
    {
      "completion_length": 232.02679443359375,
      "epoch": 1.248795003981726,
      "grad_norm": 0.24044948180768533,
      "kl": 0.0777587890625,
      "learning_rate": 4.99169706776796e-07,
      "loss": 0.0001,
      "reward": 1.7482143566012383,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7526786103844643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7448
    },
    {
      "completion_length": 228.03125858306885,
      "epoch": 1.2491303072215936,
      "grad_norm": 0.11412891327814846,
      "kl": 0.0772857666015625,
      "learning_rate": 4.991687296408371e-07,
      "loss": 0.0001,
      "reward": 1.7678571939468384,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.767857164144516,
      "rewards/format_reward_func": 1.0,
      "step": 7450
    },
    {
      "completion_length": 232.45090293884277,
      "epoch": 1.2494656104614612,
      "grad_norm": 0.22571443973421887,
      "kl": 0.069427490234375,
      "learning_rate": 4.991677519311993e-07,
      "loss": 0.0001,
      "reward": 1.7857143580913544,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 7452
    },
    {
      "completion_length": 234.48215293884277,
      "epoch": 1.2498009137013286,
      "grad_norm": 0.21476238687167007,
      "kl": 0.0712890625,
      "learning_rate": 4.99166773647885e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 7454
    },
    {
      "completion_length": 230.04018783569336,
      "epoch": 1.2501362169411963,
      "grad_norm": 0.25465075300402357,
      "kl": 0.0692291259765625,
      "learning_rate": 4.991657947908964e-07,
      "loss": 0.0001,
      "reward": 1.7357143387198448,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7357143238186836,
      "rewards/format_reward_func": 1.0,
      "step": 7456
    },
    {
      "completion_length": 229.64286613464355,
      "epoch": 1.2504715201810637,
      "grad_norm": 0.6685778048694985,
      "kl": 0.0740203857421875,
      "learning_rate": 4.991648153602358e-07,
      "loss": 0.0001,
      "reward": 1.748214341700077,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7526786141097546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7458
    },
    {
      "completion_length": 238.12054920196533,
      "epoch": 1.2508068234209313,
      "grad_norm": 0.27757295188385095,
      "kl": 0.0650787353515625,
      "learning_rate": 4.991638353559054e-07,
      "loss": 0.0001,
      "reward": 1.785714365541935,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 7460
    },
    {
      "completion_length": 231.2500114440918,
      "epoch": 1.251142126660799,
      "grad_norm": 0.25223749992442296,
      "kl": 0.0772247314453125,
      "learning_rate": 4.991628547779074e-07,
      "loss": 0.0001,
      "reward": 1.775000087916851,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 7462
    },
    {
      "completion_length": 232.54465579986572,
      "epoch": 1.2514774299006663,
      "grad_norm": 0.20840119266884935,
      "kl": 0.0738983154296875,
      "learning_rate": 4.991618736262442e-07,
      "loss": 0.0001,
      "reward": 1.7285714894533157,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7285714596509933,
      "rewards/format_reward_func": 1.0,
      "step": 7464
    },
    {
      "completion_length": 235.50447845458984,
      "epoch": 1.251812733140534,
      "grad_norm": 0.24398954293589886,
      "kl": 0.07470703125,
      "learning_rate": 4.991608919009179e-07,
      "loss": 0.0001,
      "reward": 1.7678571939468384,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 7466
    },
    {
      "completion_length": 242.83929443359375,
      "epoch": 1.2521480363804014,
      "grad_norm": 0.19045043384066485,
      "kl": 0.0765533447265625,
      "learning_rate": 4.991599096019309e-07,
      "loss": 0.0001,
      "reward": 1.6910714954137802,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.6955357603728771,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7468
    },
    {
      "completion_length": 229.02233219146729,
      "epoch": 1.252483339620269,
      "grad_norm": 0.2739781670065189,
      "kl": 0.090850830078125,
      "learning_rate": 4.991589267292854e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143145054579,
      "rewards/format_reward_func": 1.0,
      "step": 7470
    },
    {
      "completion_length": 234.63393688201904,
      "epoch": 1.2528186428601367,
      "grad_norm": 0.5638522629412447,
      "kl": 0.0961456298828125,
      "learning_rate": 4.991579432829837e-07,
      "loss": 0.0001,
      "reward": 1.8160714879631996,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8205357380211353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7472
    },
    {
      "completion_length": 227.69644165039062,
      "epoch": 1.2531539461000043,
      "grad_norm": 0.24793558166991733,
      "kl": 0.0794677734375,
      "learning_rate": 4.99156959263028e-07,
      "loss": 0.0001,
      "reward": 1.789285771548748,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.789285734295845,
      "rewards/format_reward_func": 1.0,
      "step": 7474
    },
    {
      "completion_length": 231.8973331451416,
      "epoch": 1.2534892493398717,
      "grad_norm": 0.10141429069354206,
      "kl": 0.0692138671875,
      "learning_rate": 4.991559746694206e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.796428594738245,
      "rewards/format_reward_func": 1.0,
      "step": 7476
    },
    {
      "completion_length": 231.5446538925171,
      "epoch": 1.2538245525797393,
      "grad_norm": 0.2815565446267239,
      "kl": 0.0777587890625,
      "learning_rate": 4.991549895021638e-07,
      "loss": 0.0001,
      "reward": 1.721428669989109,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7214285954833031,
      "rewards/format_reward_func": 1.0,
      "step": 7478
    },
    {
      "completion_length": 230.41518878936768,
      "epoch": 1.2541598558196068,
      "grad_norm": 0.19892908493645406,
      "kl": 0.0809478759765625,
      "learning_rate": 4.991540037612598e-07,
      "loss": 0.0001,
      "reward": 1.8107143491506577,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 7480
    },
    {
      "completion_length": 237.64733028411865,
      "epoch": 1.2544951590594744,
      "grad_norm": 0.21884502968045427,
      "kl": 0.0768890380859375,
      "learning_rate": 4.991530174467109e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 7482
    },
    {
      "completion_length": 236.6919755935669,
      "epoch": 1.254830462299342,
      "grad_norm": 0.16027566375000496,
      "kl": 0.0768585205078125,
      "learning_rate": 4.991520305585194e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 7484
    },
    {
      "completion_length": 251.01340293884277,
      "epoch": 1.2551657655392094,
      "grad_norm": 0.2896072683566364,
      "kl": 0.082977294921875,
      "learning_rate": 4.991510430966875e-07,
      "loss": 0.0001,
      "reward": 1.7232143804430962,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.727678619325161,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7486
    },
    {
      "completion_length": 236.20537090301514,
      "epoch": 1.255501068779077,
      "grad_norm": 0.24092087177097757,
      "kl": 0.0662078857421875,
      "learning_rate": 4.991500550612176e-07,
      "loss": 0.0001,
      "reward": 1.814285784959793,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.8142857551574707,
      "rewards/format_reward_func": 1.0,
      "step": 7488
    },
    {
      "completion_length": 239.6919765472412,
      "epoch": 1.2558363720189447,
      "grad_norm": 0.27121315571398896,
      "kl": 0.0826873779296875,
      "learning_rate": 4.991490664521119e-07,
      "loss": 0.0001,
      "reward": 1.7178572118282318,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7178571783006191,
      "rewards/format_reward_func": 1.0,
      "step": 7490
    },
    {
      "completion_length": 239.20983028411865,
      "epoch": 1.2561716752588121,
      "grad_norm": 0.16039280148087817,
      "kl": 0.0776519775390625,
      "learning_rate": 4.991480772693726e-07,
      "loss": 0.0001,
      "reward": 1.760714367032051,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 7492
    },
    {
      "completion_length": 231.55358219146729,
      "epoch": 1.2565069784986798,
      "grad_norm": 0.25401639410075905,
      "kl": 0.0705413818359375,
      "learning_rate": 4.99147087513002e-07,
      "loss": 0.0001,
      "reward": 1.7553571984171867,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7598214522004128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7494
    },
    {
      "completion_length": 236.26340293884277,
      "epoch": 1.2568422817385474,
      "grad_norm": 0.31600402330566846,
      "kl": 0.071044921875,
      "learning_rate": 4.991460971830026e-07,
      "loss": 0.0001,
      "reward": 1.7303571999073029,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7348214536905289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7496
    },
    {
      "completion_length": 236.80358123779297,
      "epoch": 1.2571775849784148,
      "grad_norm": 0.38116439183130013,
      "kl": 0.0737762451171875,
      "learning_rate": 4.991451062793763e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714286148548126,
      "rewards/format_reward_func": 1.0,
      "step": 7498
    },
    {
      "completion_length": 237.85715198516846,
      "epoch": 1.2575128882182824,
      "grad_norm": 0.2152823007031557,
      "kl": 0.0875701904296875,
      "learning_rate": 4.991441148021258e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143055647612,
      "rewards/format_reward_func": 1.0,
      "step": 7500
    },
    {
      "completion_length": 239.27233219146729,
      "epoch": 1.2578481914581499,
      "grad_norm": 0.23772054865269418,
      "kl": 0.0738372802734375,
      "learning_rate": 4.991431227512531e-07,
      "loss": 0.0001,
      "reward": 1.7857143133878708,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143320143223,
      "rewards/format_reward_func": 1.0,
      "step": 7502
    },
    {
      "completion_length": 233.86607933044434,
      "epoch": 1.2581834946980175,
      "grad_norm": 0.16101616603817884,
      "kl": 0.11566162109375,
      "learning_rate": 4.991421301267604e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 7504
    },
    {
      "completion_length": 234.1071548461914,
      "epoch": 1.2585187979378851,
      "grad_norm": 0.2618933580358117,
      "kl": 0.07000732421875,
      "learning_rate": 4.991411369286503e-07,
      "loss": 0.0001,
      "reward": 1.7642858028411865,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857544124126,
      "rewards/format_reward_func": 1.0,
      "step": 7506
    },
    {
      "completion_length": 238.22768783569336,
      "epoch": 1.2588541011777528,
      "grad_norm": 0.2759809096611727,
      "kl": 0.073516845703125,
      "learning_rate": 4.99140143156925e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 7508
    },
    {
      "completion_length": 229.99107933044434,
      "epoch": 1.2591894044176202,
      "grad_norm": 0.13394452526505557,
      "kl": 0.0980377197265625,
      "learning_rate": 4.991391488115866e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 7510
    },
    {
      "completion_length": 234.29018688201904,
      "epoch": 1.2595247076574878,
      "grad_norm": 0.6323557898045636,
      "kl": 0.08587646484375,
      "learning_rate": 4.991381538926374e-07,
      "loss": 0.0001,
      "reward": 1.7589286118745804,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7633928898721933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7512
    },
    {
      "completion_length": 229.91965293884277,
      "epoch": 1.2598600108973552,
      "grad_norm": 0.40135163456317896,
      "kl": 0.07708740234375,
      "learning_rate": 4.991371584000799e-07,
      "loss": 0.0001,
      "reward": 1.7660714834928513,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7705357354134321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7514
    },
    {
      "completion_length": 227.17858028411865,
      "epoch": 1.2601953141372229,
      "grad_norm": 0.9782226270408743,
      "kl": 0.07330322265625,
      "learning_rate": 4.991361623339164e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 7516
    },
    {
      "completion_length": 232.57143878936768,
      "epoch": 1.2605306173770905,
      "grad_norm": 0.2562746459472891,
      "kl": 0.0767974853515625,
      "learning_rate": 4.991351656941489e-07,
      "loss": 0.0001,
      "reward": 1.8000000640749931,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000193715096,
      "rewards/format_reward_func": 1.0,
      "step": 7518
    },
    {
      "completion_length": 241.55358219146729,
      "epoch": 1.260865920616958,
      "grad_norm": 0.2472584451345171,
      "kl": 0.0744781494140625,
      "learning_rate": 4.9913416848078e-07,
      "loss": 0.0001,
      "reward": 1.800000049173832,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 7520
    },
    {
      "completion_length": 234.4107255935669,
      "epoch": 1.2612012238568255,
      "grad_norm": 0.36177741112164086,
      "kl": 0.0704498291015625,
      "learning_rate": 4.991331706938118e-07,
      "loss": 0.0001,
      "reward": 1.7607143446803093,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 7522
    },
    {
      "completion_length": 242.14286708831787,
      "epoch": 1.261536527096693,
      "grad_norm": 0.2512713998995826,
      "kl": 0.073394775390625,
      "learning_rate": 4.991321723332467e-07,
      "loss": 0.0001,
      "reward": 1.73392865806818,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7383928820490837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7524
    },
    {
      "completion_length": 238.54911994934082,
      "epoch": 1.2618718303365606,
      "grad_norm": 0.23311703994054409,
      "kl": 0.072662353515625,
      "learning_rate": 4.99131173399087e-07,
      "loss": 0.0001,
      "reward": 1.7553572058677673,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7598214596509933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7526
    },
    {
      "completion_length": 240.94197463989258,
      "epoch": 1.2622071335764282,
      "grad_norm": 0.23393040111609653,
      "kl": 0.07373046875,
      "learning_rate": 4.99130173891335e-07,
      "loss": 0.0001,
      "reward": 1.7928571850061417,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571626543999,
      "rewards/format_reward_func": 1.0,
      "step": 7528
    },
    {
      "completion_length": 238.2544755935669,
      "epoch": 1.2625424368162959,
      "grad_norm": 0.2596660312043446,
      "kl": 0.07232666015625,
      "learning_rate": 4.991291738099929e-07,
      "loss": 0.0001,
      "reward": 1.733928643167019,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7383928932249546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7530
    },
    {
      "completion_length": 227.68304538726807,
      "epoch": 1.2628777400561633,
      "grad_norm": 0.1490256935867386,
      "kl": 0.0784912109375,
      "learning_rate": 4.991281731550631e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 7532
    },
    {
      "completion_length": 226.21876049041748,
      "epoch": 1.263213043296031,
      "grad_norm": 0.33889375367672325,
      "kl": 0.081512451171875,
      "learning_rate": 4.991271719265477e-07,
      "loss": 0.0001,
      "reward": 1.7500000894069672,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000204890966,
      "rewards/format_reward_func": 1.0,
      "step": 7534
    },
    {
      "completion_length": 229.75893592834473,
      "epoch": 1.2635483465358983,
      "grad_norm": 0.19253892513658385,
      "kl": 0.07342529296875,
      "learning_rate": 4.991261701244494e-07,
      "loss": 0.0001,
      "reward": 1.764285795390606,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 7536
    },
    {
      "completion_length": 232.67411994934082,
      "epoch": 1.263883649775766,
      "grad_norm": 0.1851524664373545,
      "kl": 0.0771484375,
      "learning_rate": 4.991251677487702e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143335044384,
      "rewards/format_reward_func": 1.0,
      "step": 7538
    },
    {
      "completion_length": 228.49108123779297,
      "epoch": 1.2642189530156336,
      "grad_norm": 0.3447227639574531,
      "kl": 0.0768890380859375,
      "learning_rate": 4.991241647995126e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428809314966,
      "rewards/format_reward_func": 1.0,
      "step": 7540
    },
    {
      "completion_length": 223.60715293884277,
      "epoch": 1.264554256255501,
      "grad_norm": 0.21475977305995408,
      "kl": 0.079498291015625,
      "learning_rate": 4.991231612766786e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 7542
    },
    {
      "completion_length": 228.71429538726807,
      "epoch": 1.2648895594953686,
      "grad_norm": 0.26691773585830647,
      "kl": 0.1363372802734375,
      "learning_rate": 4.99122157180271e-07,
      "loss": 0.0001,
      "reward": 1.792857214808464,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571626543999,
      "rewards/format_reward_func": 1.0,
      "step": 7544
    },
    {
      "completion_length": 220.34822368621826,
      "epoch": 1.265224862735236,
      "grad_norm": 0.20614781368504556,
      "kl": 0.094970703125,
      "learning_rate": 4.991211525102916e-07,
      "loss": 0.0001,
      "reward": 1.7821429371833801,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 7546
    },
    {
      "completion_length": 224.1696538925171,
      "epoch": 1.2655601659751037,
      "grad_norm": 0.23147580973292636,
      "kl": 0.0751190185546875,
      "learning_rate": 4.99120147266743e-07,
      "loss": 0.0001,
      "reward": 1.7125000730156898,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7169643249362707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7548
    },
    {
      "completion_length": 221.75447273254395,
      "epoch": 1.2658954692149713,
      "grad_norm": 0.6172895430411022,
      "kl": 0.0866851806640625,
      "learning_rate": 4.991191414496274e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 7550
    },
    {
      "completion_length": 218.9553680419922,
      "epoch": 1.266230772454839,
      "grad_norm": 0.24688172383365634,
      "kl": 0.11077880859375,
      "learning_rate": 4.991181350589473e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857529222965,
      "rewards/format_reward_func": 1.0,
      "step": 7552
    },
    {
      "completion_length": 233.47768878936768,
      "epoch": 1.2665660756947064,
      "grad_norm": 0.36008920665302907,
      "kl": 0.0836181640625,
      "learning_rate": 4.991171280947047e-07,
      "loss": 0.0001,
      "reward": 1.7160715013742447,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7205357477068901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7554
    },
    {
      "completion_length": 221.57590103149414,
      "epoch": 1.266901378934574,
      "grad_norm": 0.2618971345873311,
      "kl": 0.087188720703125,
      "learning_rate": 4.991161205569023e-07,
      "loss": 0.0001,
      "reward": 1.7375000715255737,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7419643122702837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7556
    },
    {
      "completion_length": 224.26340293884277,
      "epoch": 1.2672366821744414,
      "grad_norm": 0.331582401421873,
      "kl": 0.1146087646484375,
      "learning_rate": 4.991151124455422e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 7558
    },
    {
      "completion_length": 220.55358219146729,
      "epoch": 1.267571985414309,
      "grad_norm": 0.25120591195005054,
      "kl": 0.08917236328125,
      "learning_rate": 4.991141037606266e-07,
      "loss": 0.0001,
      "reward": 1.7375000938773155,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7419643141329288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7560
    },
    {
      "completion_length": 212.33483123779297,
      "epoch": 1.2679072886541767,
      "grad_norm": 0.40357051105146413,
      "kl": 0.08935546875,
      "learning_rate": 4.991130945021581e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286126196384,
      "rewards/format_reward_func": 1.0,
      "step": 7562
    },
    {
      "completion_length": 219.60268878936768,
      "epoch": 1.2682425918940443,
      "grad_norm": 0.40444888759563763,
      "kl": 0.0786590576171875,
      "learning_rate": 4.991120846701388e-07,
      "loss": 0.0001,
      "reward": 1.7696429342031479,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7741071581840515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7564
    },
    {
      "completion_length": 226.77679538726807,
      "epoch": 1.2685778951339117,
      "grad_norm": 0.5012331277544384,
      "kl": 0.18408203125,
      "learning_rate": 4.991110742645712e-07,
      "loss": 0.0002,
      "reward": 1.75178574770689,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7562500275671482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7566
    },
    {
      "completion_length": 227.01786708831787,
      "epoch": 1.2689131983737794,
      "grad_norm": 0.41045920652982976,
      "kl": 0.0800323486328125,
      "learning_rate": 4.991100632854575e-07,
      "loss": 0.0001,
      "reward": 1.7125000804662704,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7169643137603998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7568
    },
    {
      "completion_length": 228.75000953674316,
      "epoch": 1.2692485016136468,
      "grad_norm": 0.41643280522481385,
      "kl": 0.0793914794921875,
      "learning_rate": 4.991090517328001e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 7570
    },
    {
      "completion_length": 214.71875953674316,
      "epoch": 1.2695838048535144,
      "grad_norm": 1.6521458727305902,
      "kl": 0.091827392578125,
      "learning_rate": 4.991080396066013e-07,
      "loss": 0.0001,
      "reward": 1.7375001013278961,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7419643104076385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7572
    },
    {
      "completion_length": 211.85715103149414,
      "epoch": 1.269919108093382,
      "grad_norm": 0.7937714569865104,
      "kl": 0.311981201171875,
      "learning_rate": 4.991070269068633e-07,
      "loss": 0.0003,
      "reward": 1.7678572088479996,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 7574
    },
    {
      "completion_length": 220.1384038925171,
      "epoch": 1.2702544113332495,
      "grad_norm": 2.7509181354526198,
      "kl": 0.102203369140625,
      "learning_rate": 4.991060136335887e-07,
      "loss": 0.0001,
      "reward": 1.7821428999304771,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 7576
    },
    {
      "completion_length": 223.99554443359375,
      "epoch": 1.270589714573117,
      "grad_norm": 0.6848668616946066,
      "kl": 0.173553466796875,
      "learning_rate": 4.991049997867797e-07,
      "loss": 0.0002,
      "reward": 1.7571429312229156,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 1.0,
      "step": 7578
    },
    {
      "completion_length": 222.53572368621826,
      "epoch": 1.2709250178129845,
      "grad_norm": 0.6383890615748599,
      "kl": 0.078521728515625,
      "learning_rate": 4.991039853664386e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 7580
    },
    {
      "completion_length": 222.75447463989258,
      "epoch": 1.2712603210528521,
      "grad_norm": 0.6796629798388669,
      "kl": 0.0955047607421875,
      "learning_rate": 4.991029703725677e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 7582
    },
    {
      "completion_length": 224.05358219146729,
      "epoch": 1.2715956242927198,
      "grad_norm": 0.6087448935727924,
      "kl": 0.0820770263671875,
      "learning_rate": 4.991019548051696e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 7584
    },
    {
      "completion_length": 222.65179347991943,
      "epoch": 1.2719309275325874,
      "grad_norm": 0.2702490964335112,
      "kl": 0.08392333984375,
      "learning_rate": 4.991009386642463e-07,
      "loss": 0.0001,
      "reward": 1.74642863124609,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.746428593993187,
      "rewards/format_reward_func": 1.0,
      "step": 7586
    },
    {
      "completion_length": 220.91072463989258,
      "epoch": 1.2722662307724548,
      "grad_norm": 0.9077446820456878,
      "kl": 0.0844268798828125,
      "learning_rate": 4.990999219498002e-07,
      "loss": 0.0001,
      "reward": 1.7571429163217545,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 7588
    },
    {
      "completion_length": 222.08929347991943,
      "epoch": 1.2726015340123225,
      "grad_norm": 0.39540866250941975,
      "kl": 0.0833282470703125,
      "learning_rate": 4.990989046618338e-07,
      "loss": 0.0001,
      "reward": 1.803571455180645,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714663565159,
      "rewards/format_reward_func": 1.0,
      "step": 7590
    },
    {
      "completion_length": 220.80804538726807,
      "epoch": 1.2729368372521899,
      "grad_norm": 1.953295584845841,
      "kl": 0.0895843505859375,
      "learning_rate": 4.990978868003494e-07,
      "loss": 0.0001,
      "reward": 1.7303572073578835,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7348214611411095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7592
    },
    {
      "completion_length": 219.47768878936768,
      "epoch": 1.2732721404920575,
      "grad_norm": 0.25660438766797417,
      "kl": 0.0877685546875,
      "learning_rate": 4.990968683653492e-07,
      "loss": 0.0001,
      "reward": 1.7678572311997414,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7678571790456772,
      "rewards/format_reward_func": 1.0,
      "step": 7594
    },
    {
      "completion_length": 221.82590293884277,
      "epoch": 1.2736074437319251,
      "grad_norm": 0.4354199758652535,
      "kl": 0.088165283203125,
      "learning_rate": 4.990958493568358e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 7596
    },
    {
      "completion_length": 211.41518783569336,
      "epoch": 1.2739427469717926,
      "grad_norm": 0.20289326916956163,
      "kl": 0.0782318115234375,
      "learning_rate": 4.990948297748113e-07,
      "loss": 0.0001,
      "reward": 1.8071429133415222,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428798139095,
      "rewards/format_reward_func": 1.0,
      "step": 7598
    },
    {
      "completion_length": 218.65179538726807,
      "epoch": 1.2742780502116602,
      "grad_norm": 0.48082271814419353,
      "kl": 0.08978271484375,
      "learning_rate": 4.990938096192782e-07,
      "loss": 0.0001,
      "reward": 1.7165179327130318,
      "reward_std": 0.06755394907668233,
      "rewards/equation_reward_func": 0.7178571857511997,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 7600
    },
    {
      "completion_length": 213.29018688201904,
      "epoch": 1.2746133534515276,
      "grad_norm": 0.44731461791586646,
      "kl": 0.09454345703125,
      "learning_rate": 4.990927888902387e-07,
      "loss": 0.0001,
      "reward": 1.767410784959793,
      "reward_std": 0.0460882093757391,
      "rewards/equation_reward_func": 0.7705357484519482,
      "rewards/format_reward_func": 0.9968750029802322,
      "step": 7602
    },
    {
      "completion_length": 218.48661708831787,
      "epoch": 1.2749486566913952,
      "grad_norm": 0.4131093057482022,
      "kl": 0.0911865234375,
      "learning_rate": 4.990917675876953e-07,
      "loss": 0.0001,
      "reward": 1.6982143893837929,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7026785910129547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7604
    },
    {
      "completion_length": 209.18750762939453,
      "epoch": 1.2752839599312629,
      "grad_norm": 0.5033205739515128,
      "kl": 0.077178955078125,
      "learning_rate": 4.990907457116503e-07,
      "loss": 0.0001,
      "reward": 1.7285715341567993,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7285714522004128,
      "rewards/format_reward_func": 1.0,
      "step": 7606
    },
    {
      "completion_length": 212.22768783569336,
      "epoch": 1.2756192631711305,
      "grad_norm": 0.4563452985721969,
      "kl": 0.07989501953125,
      "learning_rate": 4.990897232621061e-07,
      "loss": 0.0001,
      "reward": 1.8178571984171867,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8178571537137032,
      "rewards/format_reward_func": 1.0,
      "step": 7608
    },
    {
      "completion_length": 213.19643878936768,
      "epoch": 1.275954566410998,
      "grad_norm": 0.5585952840788717,
      "kl": 0.07147216796875,
      "learning_rate": 4.990887002390649e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 7610
    },
    {
      "completion_length": 213.9285831451416,
      "epoch": 1.2762898696508655,
      "grad_norm": 0.32565894244767235,
      "kl": 0.0990753173828125,
      "learning_rate": 4.990876766425292e-07,
      "loss": 0.0001,
      "reward": 1.7732143625617027,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7776786014437675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7612
    },
    {
      "completion_length": 216.44197368621826,
      "epoch": 1.276625172890733,
      "grad_norm": 0.2877972666853063,
      "kl": 0.0857696533203125,
      "learning_rate": 4.990866524725013e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714286148548126,
      "rewards/format_reward_func": 1.0,
      "step": 7614
    },
    {
      "completion_length": 208.59822273254395,
      "epoch": 1.2769604761306006,
      "grad_norm": 3.039888470742808,
      "kl": 0.1195831298828125,
      "learning_rate": 4.990856277289836e-07,
      "loss": 0.0001,
      "reward": 1.787500038743019,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7919643148779869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7616
    },
    {
      "completion_length": 198.33483028411865,
      "epoch": 1.2772957793704682,
      "grad_norm": 0.004197014084676649,
      "kl": 0.0860595703125,
      "learning_rate": 4.990846024119785e-07,
      "loss": 0.0001,
      "reward": 1.8000000789761543,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 7618
    },
    {
      "completion_length": 197.69643878936768,
      "epoch": 1.2776310826103356,
      "grad_norm": 0.23318182594139808,
      "kl": 0.092559814453125,
      "learning_rate": 4.990835765214882e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 7620
    },
    {
      "completion_length": 202.15625953674316,
      "epoch": 1.2779663858502033,
      "grad_norm": 1.2937995948418413,
      "kl": 0.087005615234375,
      "learning_rate": 4.990825500575152e-07,
      "loss": 0.0001,
      "reward": 1.7785715088248253,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 1.0,
      "step": 7622
    },
    {
      "completion_length": 196.40626049041748,
      "epoch": 1.278301689090071,
      "grad_norm": 0.3271859286057177,
      "kl": 0.100067138671875,
      "learning_rate": 4.990815230200618e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 7624
    },
    {
      "completion_length": 192.56697368621826,
      "epoch": 1.2786369923299383,
      "grad_norm": 0.3072899399621595,
      "kl": 0.098114013671875,
      "learning_rate": 4.990804954091302e-07,
      "loss": 0.0001,
      "reward": 1.792857214808464,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 7626
    },
    {
      "completion_length": 194.65625858306885,
      "epoch": 1.278972295569806,
      "grad_norm": 0.315932345161278,
      "kl": 0.0919036865234375,
      "learning_rate": 4.990794672247232e-07,
      "loss": 0.0001,
      "reward": 1.7392857819795609,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857596278191,
      "rewards/format_reward_func": 1.0,
      "step": 7628
    },
    {
      "completion_length": 198.43750953674316,
      "epoch": 1.2793075988096736,
      "grad_norm": 0.0880141059924044,
      "kl": 0.0974578857421875,
      "learning_rate": 4.990784384668428e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 7630
    },
    {
      "completion_length": 188.29465293884277,
      "epoch": 1.279642902049541,
      "grad_norm": 0.25524708052962836,
      "kl": 0.106048583984375,
      "learning_rate": 4.990774091354915e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714286055415869,
      "rewards/format_reward_func": 1.0,
      "step": 7632
    },
    {
      "completion_length": 187.6160774230957,
      "epoch": 1.2799782052894086,
      "grad_norm": 0.27881454543754314,
      "kl": 0.1068115234375,
      "learning_rate": 4.990763792306717e-07,
      "loss": 0.0001,
      "reward": 1.753571517765522,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714525729418,
      "rewards/format_reward_func": 1.0,
      "step": 7634
    },
    {
      "completion_length": 188.61161518096924,
      "epoch": 1.280313508529276,
      "grad_norm": 0.31344919963315515,
      "kl": 0.11016845703125,
      "learning_rate": 4.990753487523857e-07,
      "loss": 0.0001,
      "reward": 1.760714367032051,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 7636
    },
    {
      "completion_length": 193.87054538726807,
      "epoch": 1.2806488117691437,
      "grad_norm": 0.2240318405017893,
      "kl": 0.0923614501953125,
      "learning_rate": 4.990743177006358e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714454948902,
      "rewards/format_reward_func": 1.0,
      "step": 7638
    },
    {
      "completion_length": 194.4017915725708,
      "epoch": 1.2809841150090113,
      "grad_norm": 0.22470951217114793,
      "kl": 0.105865478515625,
      "learning_rate": 4.990732860754246e-07,
      "loss": 0.0001,
      "reward": 1.7785714715719223,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714697092772,
      "rewards/format_reward_func": 1.0,
      "step": 7640
    },
    {
      "completion_length": 188.56250953674316,
      "epoch": 1.281319418248879,
      "grad_norm": 0.37580742577286996,
      "kl": 0.088165283203125,
      "learning_rate": 4.990722538767543e-07,
      "loss": 0.0001,
      "reward": 1.7571429461240768,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 7642
    },
    {
      "completion_length": 188.35715103149414,
      "epoch": 1.2816547214887464,
      "grad_norm": 0.1599668296641991,
      "kl": 0.10662841796875,
      "learning_rate": 4.990712211046273e-07,
      "loss": 0.0001,
      "reward": 1.7285715118050575,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714633762836,
      "rewards/format_reward_func": 1.0,
      "step": 7644
    },
    {
      "completion_length": 192.56697368621826,
      "epoch": 1.281990024728614,
      "grad_norm": 0.31148905908911567,
      "kl": 0.083526611328125,
      "learning_rate": 4.990701877590461e-07,
      "loss": 0.0001,
      "reward": 1.7285714969038963,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714633762836,
      "rewards/format_reward_func": 1.0,
      "step": 7646
    },
    {
      "completion_length": 194.48215293884277,
      "epoch": 1.2823253279684814,
      "grad_norm": 0.20670813668367558,
      "kl": 0.090240478515625,
      "learning_rate": 4.990691538400128e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 7648
    },
    {
      "completion_length": 188.84822177886963,
      "epoch": 1.282660631208349,
      "grad_norm": 0.14393064848231346,
      "kl": 0.095794677734375,
      "learning_rate": 4.990681193475301e-07,
      "loss": 0.0001,
      "reward": 1.792857214808464,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 7650
    },
    {
      "completion_length": 190.66072177886963,
      "epoch": 1.2829959344482167,
      "grad_norm": 0.17232390997388092,
      "kl": 0.104400634765625,
      "learning_rate": 4.990670842816002e-07,
      "loss": 0.0001,
      "reward": 1.7486608028411865,
      "reward_std": 0.03219861118122935,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 7652
    },
    {
      "completion_length": 192.68304443359375,
      "epoch": 1.283331237688084,
      "grad_norm": 0.2523200705211646,
      "kl": 0.0912628173828125,
      "learning_rate": 4.990660486422256e-07,
      "loss": 0.0001,
      "reward": 1.8214286118745804,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8214286006987095,
      "rewards/format_reward_func": 1.0,
      "step": 7654
    },
    {
      "completion_length": 192.41965007781982,
      "epoch": 1.2836665409279517,
      "grad_norm": 0.09130844397646869,
      "kl": 0.083648681640625,
      "learning_rate": 4.990650124294085e-07,
      "loss": 0.0001,
      "reward": 1.7321429178118706,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7321428917348385,
      "rewards/format_reward_func": 1.0,
      "step": 7656
    },
    {
      "completion_length": 198.56250953674316,
      "epoch": 1.2840018441678192,
      "grad_norm": 0.2058536185574342,
      "kl": 0.0894012451171875,
      "learning_rate": 4.990639756431514e-07,
      "loss": 0.0001,
      "reward": 1.8035714849829674,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714663565159,
      "rewards/format_reward_func": 1.0,
      "step": 7658
    },
    {
      "completion_length": 192.67411613464355,
      "epoch": 1.2843371474076868,
      "grad_norm": 0.4357492926407639,
      "kl": 0.0915069580078125,
      "learning_rate": 4.990629382834567e-07,
      "loss": 0.0001,
      "reward": 1.7357143685221672,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7357143182307482,
      "rewards/format_reward_func": 1.0,
      "step": 7660
    },
    {
      "completion_length": 199.83482837677002,
      "epoch": 1.2846724506475544,
      "grad_norm": 0.31513147134974884,
      "kl": 0.083404541015625,
      "learning_rate": 4.990619003503268e-07,
      "loss": 0.0001,
      "reward": 1.7250000908970833,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000312924385,
      "rewards/format_reward_func": 1.0,
      "step": 7662
    },
    {
      "completion_length": 194.86161613464355,
      "epoch": 1.285007753887422,
      "grad_norm": 0.19215721495269542,
      "kl": 0.096099853515625,
      "learning_rate": 4.990608618437641e-07,
      "loss": 0.0001,
      "reward": 1.7714286148548126,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 7664
    },
    {
      "completion_length": 198.66072273254395,
      "epoch": 1.2853430571272895,
      "grad_norm": 0.2876697309370327,
      "kl": 0.0840911865234375,
      "learning_rate": 4.990598227637708e-07,
      "loss": 0.0001,
      "reward": 1.6950893551111221,
      "reward_std": 0.02714784862473607,
      "rewards/equation_reward_func": 0.6964286249130964,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 7666
    },
    {
      "completion_length": 206.60268688201904,
      "epoch": 1.285678360367157,
      "grad_norm": 0.2944065370264894,
      "kl": 0.0862579345703125,
      "learning_rate": 4.990587831103495e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 7668
    },
    {
      "completion_length": 210.10715293884277,
      "epoch": 1.2860136636070245,
      "grad_norm": 0.23531668996266744,
      "kl": 0.0872650146484375,
      "learning_rate": 4.990577428835026e-07,
      "loss": 0.0001,
      "reward": 1.7821429371833801,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 7670
    },
    {
      "completion_length": 209.81697177886963,
      "epoch": 1.2863489668468922,
      "grad_norm": 0.31108048300835284,
      "kl": 0.0871734619140625,
      "learning_rate": 4.990567020832325e-07,
      "loss": 0.0001,
      "reward": 1.7035715132951736,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7035714574158192,
      "rewards/format_reward_func": 1.0,
      "step": 7672
    },
    {
      "completion_length": 200.08482933044434,
      "epoch": 1.2866842700867598,
      "grad_norm": 0.30616690330063095,
      "kl": 0.07989501953125,
      "learning_rate": 4.990556607095414e-07,
      "loss": 0.0001,
      "reward": 1.8142858073115349,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8142857365310192,
      "rewards/format_reward_func": 1.0,
      "step": 7674
    },
    {
      "completion_length": 200.36161518096924,
      "epoch": 1.2870195733266272,
      "grad_norm": 0.31553270795768634,
      "kl": 0.0896759033203125,
      "learning_rate": 4.990546187624319e-07,
      "loss": 0.0001,
      "reward": 1.7821429371833801,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428701281548,
      "rewards/format_reward_func": 1.0,
      "step": 7676
    },
    {
      "completion_length": 207.80357837677002,
      "epoch": 1.2873548765664948,
      "grad_norm": 0.4448366525953187,
      "kl": 0.09698486328125,
      "learning_rate": 4.990535762419062e-07,
      "loss": 0.0001,
      "reward": 1.7553572207689285,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7598214596509933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7678
    },
    {
      "completion_length": 207.01340198516846,
      "epoch": 1.2876901798063622,
      "grad_norm": 0.3822823166526979,
      "kl": 0.075531005859375,
      "learning_rate": 4.990525331479669e-07,
      "loss": 0.0001,
      "reward": 1.7714286148548126,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714286092668772,
      "rewards/format_reward_func": 1.0,
      "step": 7680
    },
    {
      "completion_length": 213.92858123779297,
      "epoch": 1.2880254830462299,
      "grad_norm": 0.1842616328582102,
      "kl": 0.0823974609375,
      "learning_rate": 4.990514894806164e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 7682
    },
    {
      "completion_length": 213.35268783569336,
      "epoch": 1.2883607862860975,
      "grad_norm": 0.3047119390670931,
      "kl": 0.0763092041015625,
      "learning_rate": 4.99050445239857e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 7684
    },
    {
      "completion_length": 201.67858123779297,
      "epoch": 1.2886960895259651,
      "grad_norm": 0.24882400891042786,
      "kl": 0.0864715576171875,
      "learning_rate": 4.990494004256911e-07,
      "loss": 0.0001,
      "reward": 1.7750000730156898,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 7686
    },
    {
      "completion_length": 210.66518783569336,
      "epoch": 1.2890313927658326,
      "grad_norm": 0.23940484799827028,
      "kl": 0.0796661376953125,
      "learning_rate": 4.990483550381211e-07,
      "loss": 0.0001,
      "reward": 1.764285795390606,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 7688
    },
    {
      "completion_length": 215.89733123779297,
      "epoch": 1.2893666960057002,
      "grad_norm": 0.30294980556103956,
      "kl": 0.0738983154296875,
      "learning_rate": 4.990473090771494e-07,
      "loss": 0.0001,
      "reward": 1.814285770058632,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8142857439815998,
      "rewards/format_reward_func": 1.0,
      "step": 7690
    },
    {
      "completion_length": 215.60715103149414,
      "epoch": 1.2897019992455676,
      "grad_norm": 0.07984567628802022,
      "kl": 0.0729522705078125,
      "learning_rate": 4.990462625427786e-07,
      "loss": 0.0001,
      "reward": 1.7392857670783997,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857521772385,
      "rewards/format_reward_func": 1.0,
      "step": 7692
    },
    {
      "completion_length": 209.93304538726807,
      "epoch": 1.2900373024854352,
      "grad_norm": 0.2513657048792789,
      "kl": 0.0680999755859375,
      "learning_rate": 4.990452154350109e-07,
      "loss": 0.0001,
      "reward": 1.7843750640749931,
      "reward_std": 0.05240166140720248,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 7694
    },
    {
      "completion_length": 220.35268878936768,
      "epoch": 1.2903726057253029,
      "grad_norm": 0.23234477772818157,
      "kl": 0.0713348388671875,
      "learning_rate": 4.990441677538489e-07,
      "loss": 0.0001,
      "reward": 1.7250000685453415,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7250000275671482,
      "rewards/format_reward_func": 1.0,
      "step": 7696
    },
    {
      "completion_length": 209.14733028411865,
      "epoch": 1.2907079089651705,
      "grad_norm": 0.2776459759810098,
      "kl": 0.0708160400390625,
      "learning_rate": 4.990431194992946e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 7698
    },
    {
      "completion_length": 220.24554443359375,
      "epoch": 1.291043212205038,
      "grad_norm": 0.3385624563063057,
      "kl": 0.0813751220703125,
      "learning_rate": 4.99042070671351e-07,
      "loss": 0.0001,
      "reward": 1.692857250571251,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.6928571835160255,
      "rewards/format_reward_func": 1.0,
      "step": 7700
    },
    {
      "completion_length": 218.83483028411865,
      "epoch": 1.2913785154449056,
      "grad_norm": 0.22457947375299586,
      "kl": 0.0742645263671875,
      "learning_rate": 4.9904102127002e-07,
      "loss": 0.0001,
      "reward": 1.8357143551111221,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.835714302957058,
      "rewards/format_reward_func": 1.0,
      "step": 7702
    },
    {
      "completion_length": 218.10268783569336,
      "epoch": 1.291713818684773,
      "grad_norm": 0.4024637053056976,
      "kl": 0.079010009765625,
      "learning_rate": 4.990399712953044e-07,
      "loss": 0.0001,
      "reward": 1.7357143461704254,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7357143349945545,
      "rewards/format_reward_func": 1.0,
      "step": 7704
    },
    {
      "completion_length": 217.05804824829102,
      "epoch": 1.2920491219246406,
      "grad_norm": 0.15498067836171614,
      "kl": 0.0782928466796875,
      "learning_rate": 4.990389207472064e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714286111295223,
      "rewards/format_reward_func": 1.0,
      "step": 7706
    },
    {
      "completion_length": 222.46875953674316,
      "epoch": 1.2923844251645082,
      "grad_norm": 0.3098561962327767,
      "kl": 0.0785369873046875,
      "learning_rate": 4.990378696257284e-07,
      "loss": 0.0001,
      "reward": 1.7700893506407738,
      "reward_std": 0.06250318721868098,
      "rewards/equation_reward_func": 0.7714285925030708,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 7708
    },
    {
      "completion_length": 226.8259038925171,
      "epoch": 1.2927197284043757,
      "grad_norm": 0.23701639769112526,
      "kl": 0.0764312744140625,
      "learning_rate": 4.990368179308728e-07,
      "loss": 0.0001,
      "reward": 1.764285795390606,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 7710
    },
    {
      "completion_length": 217.90625858306885,
      "epoch": 1.2930550316442433,
      "grad_norm": 0.1698667963121498,
      "kl": 0.0675811767578125,
      "learning_rate": 4.990357656626423e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 7712
    },
    {
      "completion_length": 224.6205472946167,
      "epoch": 1.2933903348841107,
      "grad_norm": 0.239622530564761,
      "kl": 0.0841827392578125,
      "learning_rate": 4.990347128210391e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8107143193483353,
      "rewards/format_reward_func": 1.0,
      "step": 7714
    },
    {
      "completion_length": 233.29911518096924,
      "epoch": 1.2937256381239783,
      "grad_norm": 0.2877922547255133,
      "kl": 0.076263427734375,
      "learning_rate": 4.990336594060656e-07,
      "loss": 0.0001,
      "reward": 1.7160715088248253,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7205357383936644,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7716
    },
    {
      "completion_length": 237.74108219146729,
      "epoch": 1.294060941363846,
      "grad_norm": 0.17369064662715267,
      "kl": 0.07440185546875,
      "learning_rate": 4.990326054177243e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464286126196384,
      "rewards/format_reward_func": 1.0,
      "step": 7718
    },
    {
      "completion_length": 234.40625953674316,
      "epoch": 1.2943962446037136,
      "grad_norm": 0.1456112782303269,
      "kl": 0.0776824951171875,
      "learning_rate": 4.990315508560176e-07,
      "loss": 0.0001,
      "reward": 1.696428656578064,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6964286025613546,
      "rewards/format_reward_func": 1.0,
      "step": 7720
    },
    {
      "completion_length": 227.63840293884277,
      "epoch": 1.294731547843581,
      "grad_norm": 0.2851940854186735,
      "kl": 0.07769775390625,
      "learning_rate": 4.99030495720948e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 7722
    },
    {
      "completion_length": 225.8571548461914,
      "epoch": 1.2950668510834487,
      "grad_norm": 0.16598813182764588,
      "kl": 0.081024169921875,
      "learning_rate": 4.990294400125177e-07,
      "loss": 0.0001,
      "reward": 1.7553571909666061,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7598214577883482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7724
    },
    {
      "completion_length": 238.36608219146729,
      "epoch": 1.295402154323316,
      "grad_norm": 0.29466927347255356,
      "kl": 0.079254150390625,
      "learning_rate": 4.990283837307294e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 7726
    },
    {
      "completion_length": 237.75447463989258,
      "epoch": 1.2957374575631837,
      "grad_norm": 0.2723778547154022,
      "kl": 0.09173583984375,
      "learning_rate": 4.990273268755855e-07,
      "loss": 0.0001,
      "reward": 1.7214286550879478,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7214286159723997,
      "rewards/format_reward_func": 1.0,
      "step": 7728
    },
    {
      "completion_length": 229.79465293884277,
      "epoch": 1.2960727608030513,
      "grad_norm": 0.23247535891431542,
      "kl": 0.0806121826171875,
      "learning_rate": 4.990262694470882e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 7730
    },
    {
      "completion_length": 233.72768878936768,
      "epoch": 1.2964080640429188,
      "grad_norm": 0.23696849547676943,
      "kl": 0.0730438232421875,
      "learning_rate": 4.990252114452403e-07,
      "loss": 0.0001,
      "reward": 1.7857143580913544,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7857143059372902,
      "rewards/format_reward_func": 1.0,
      "step": 7732
    },
    {
      "completion_length": 232.67411708831787,
      "epoch": 1.2967433672827864,
      "grad_norm": 0.26712344526497833,
      "kl": 0.0766448974609375,
      "learning_rate": 4.990241528700439e-07,
      "loss": 0.0001,
      "reward": 1.8000000640749931,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000193715096,
      "rewards/format_reward_func": 1.0,
      "step": 7734
    },
    {
      "completion_length": 239.3169755935669,
      "epoch": 1.2970786705226538,
      "grad_norm": 0.3596598026506631,
      "kl": 0.077606201171875,
      "learning_rate": 4.990230937215016e-07,
      "loss": 0.0001,
      "reward": 1.6946429312229156,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.6991071812808514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7736
    },
    {
      "completion_length": 228.24554634094238,
      "epoch": 1.2974139737625214,
      "grad_norm": 0.20678090185850923,
      "kl": 0.0703277587890625,
      "learning_rate": 4.990220339996158e-07,
      "loss": 0.0001,
      "reward": 1.7392858043313026,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857428640127,
      "rewards/format_reward_func": 1.0,
      "step": 7738
    },
    {
      "completion_length": 234.41965293884277,
      "epoch": 1.297749277002389,
      "grad_norm": 0.15463451116909518,
      "kl": 0.075775146484375,
      "learning_rate": 4.99020973704389e-07,
      "loss": 0.0001,
      "reward": 1.7410715073347092,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7455357480794191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7740
    },
    {
      "completion_length": 231.96429634094238,
      "epoch": 1.2980845802422567,
      "grad_norm": 0.2654924669106703,
      "kl": 0.0730743408203125,
      "learning_rate": 4.990199128358234e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714693367481,
      "rewards/format_reward_func": 1.0,
      "step": 7742
    },
    {
      "completion_length": 229.4196538925171,
      "epoch": 1.2984198834821241,
      "grad_norm": 0.20582707966686384,
      "kl": 0.081146240234375,
      "learning_rate": 4.990188513939219e-07,
      "loss": 0.0001,
      "reward": 1.707142911851406,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7071428894996643,
      "rewards/format_reward_func": 1.0,
      "step": 7744
    },
    {
      "completion_length": 219.16518878936768,
      "epoch": 1.2987551867219918,
      "grad_norm": 0.13035483487149235,
      "kl": 0.0706787109375,
      "learning_rate": 4.990177893786865e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 7746
    },
    {
      "completion_length": 227.17858123779297,
      "epoch": 1.2990904899618592,
      "grad_norm": 0.2269454033186533,
      "kl": 0.0691375732421875,
      "learning_rate": 4.990167267901199e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143059372902,
      "rewards/format_reward_func": 1.0,
      "step": 7748
    },
    {
      "completion_length": 229.66965198516846,
      "epoch": 1.2994257932017268,
      "grad_norm": 0.30071007329818733,
      "kl": 0.086517333984375,
      "learning_rate": 4.990156636282243e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 7750
    },
    {
      "completion_length": 231.19643878936768,
      "epoch": 1.2997610964415944,
      "grad_norm": 0.2517132251564651,
      "kl": 0.1706085205078125,
      "learning_rate": 4.990145998930025e-07,
      "loss": 0.0002,
      "reward": 1.7607143595814705,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 7752
    },
    {
      "completion_length": 223.51786708831787,
      "epoch": 1.3000963996814618,
      "grad_norm": 0.11299277167844926,
      "kl": 0.0755767822265625,
      "learning_rate": 4.990135355844567e-07,
      "loss": 0.0001,
      "reward": 1.8285714760422707,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8285714499652386,
      "rewards/format_reward_func": 1.0,
      "step": 7754
    },
    {
      "completion_length": 225.43304443359375,
      "epoch": 1.3004317029213295,
      "grad_norm": 0.2308531874026641,
      "kl": 0.0781707763671875,
      "learning_rate": 4.990124707025894e-07,
      "loss": 0.0001,
      "reward": 1.7500000521540642,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 7756
    },
    {
      "completion_length": 229.23661613464355,
      "epoch": 1.3007670061611971,
      "grad_norm": 0.21235278465773202,
      "kl": 0.084625244140625,
      "learning_rate": 4.990114052474031e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 7758
    },
    {
      "completion_length": 230.1116189956665,
      "epoch": 1.3011023094010645,
      "grad_norm": 0.21202217971475054,
      "kl": 0.0723724365234375,
      "learning_rate": 4.990103392189002e-07,
      "loss": 0.0001,
      "reward": 1.807142935693264,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428798139095,
      "rewards/format_reward_func": 1.0,
      "step": 7760
    },
    {
      "completion_length": 226.2678689956665,
      "epoch": 1.3014376126409322,
      "grad_norm": 0.3171202168947194,
      "kl": 0.081817626953125,
      "learning_rate": 4.990092726170832e-07,
      "loss": 0.0001,
      "reward": 1.728571504354477,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7285714671015739,
      "rewards/format_reward_func": 1.0,
      "step": 7762
    },
    {
      "completion_length": 231.81697463989258,
      "epoch": 1.3017729158807998,
      "grad_norm": 0.07547602806092736,
      "kl": 0.079559326171875,
      "learning_rate": 4.990082054419544e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.07071067858487368,
      "rewards/equation_reward_func": 0.7875000312924385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7764
    },
    {
      "completion_length": 221.40179538726807,
      "epoch": 1.3021082191206672,
      "grad_norm": 0.19411017636005276,
      "kl": 0.075775146484375,
      "learning_rate": 4.990071376935165e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 7766
    },
    {
      "completion_length": 222.508939743042,
      "epoch": 1.3024435223605348,
      "grad_norm": 0.2174186617837793,
      "kl": 0.0889739990234375,
      "learning_rate": 4.99006069371772e-07,
      "loss": 0.0001,
      "reward": 1.708928644657135,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7133928798139095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7768
    },
    {
      "completion_length": 218.52233123779297,
      "epoch": 1.3027788256004023,
      "grad_norm": 0.30772843925684945,
      "kl": 0.07672119140625,
      "learning_rate": 4.990050004767229e-07,
      "loss": 0.0001,
      "reward": 1.792857214808464,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 7770
    },
    {
      "completion_length": 224.69197463989258,
      "epoch": 1.30311412884027,
      "grad_norm": 0.3108382816493444,
      "kl": 0.084442138671875,
      "learning_rate": 4.990039310083722e-07,
      "loss": 0.0001,
      "reward": 1.728571504354477,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7285714615136385,
      "rewards/format_reward_func": 1.0,
      "step": 7772
    },
    {
      "completion_length": 228.41965103149414,
      "epoch": 1.3034494320801375,
      "grad_norm": 0.14404724983058562,
      "kl": 0.0753021240234375,
      "learning_rate": 4.99002860966722e-07,
      "loss": 0.0001,
      "reward": 1.8178571984171867,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.817857164889574,
      "rewards/format_reward_func": 1.0,
      "step": 7774
    },
    {
      "completion_length": 216.98661518096924,
      "epoch": 1.3037847353200052,
      "grad_norm": 0.2621768758533317,
      "kl": 0.0835113525390625,
      "learning_rate": 4.990017903517748e-07,
      "loss": 0.0001,
      "reward": 1.746428668498993,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 7776
    },
    {
      "completion_length": 212.20090103149414,
      "epoch": 1.3041200385598726,
      "grad_norm": 0.3576231235033641,
      "kl": 0.077362060546875,
      "learning_rate": 4.990007191635333e-07,
      "loss": 0.0001,
      "reward": 1.7821428999304771,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 7778
    },
    {
      "completion_length": 217.42858028411865,
      "epoch": 1.3044553417997402,
      "grad_norm": 0.33123491228549057,
      "kl": 0.0748291015625,
      "learning_rate": 4.989996474019998e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 7780
    },
    {
      "completion_length": 223.04018878936768,
      "epoch": 1.3047906450396076,
      "grad_norm": 0.2896579109829537,
      "kl": 0.0826416015625,
      "learning_rate": 4.989985750671768e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000469386578,
      "rewards/format_reward_func": 1.0,
      "step": 7782
    },
    {
      "completion_length": 224.85268878936768,
      "epoch": 1.3051259482794753,
      "grad_norm": 0.38177411239092823,
      "kl": 0.0991973876953125,
      "learning_rate": 4.989975021590668e-07,
      "loss": 0.0001,
      "reward": 1.739285796880722,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7482143193483353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7784
    },
    {
      "completion_length": 217.80358123779297,
      "epoch": 1.305461251519343,
      "grad_norm": 0.391786171910137,
      "kl": 0.09320068359375,
      "learning_rate": 4.989964286776721e-07,
      "loss": 0.0001,
      "reward": 1.7486607655882835,
      "reward_std": 0.06250318652018905,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 7786
    },
    {
      "completion_length": 219.24108219146729,
      "epoch": 1.3057965547592103,
      "grad_norm": 0.29750057352726367,
      "kl": 0.0730133056640625,
      "learning_rate": 4.989953546229954e-07,
      "loss": 0.0001,
      "reward": 1.7660714909434319,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7705357372760773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7788
    },
    {
      "completion_length": 228.39733123779297,
      "epoch": 1.306131857999078,
      "grad_norm": 0.339059597344952,
      "kl": 0.06829833984375,
      "learning_rate": 4.98994279995039e-07,
      "loss": 0.0001,
      "reward": 1.8035715073347092,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 7790
    },
    {
      "completion_length": 224.32143783569336,
      "epoch": 1.3064671612389454,
      "grad_norm": 0.17052533501531209,
      "kl": 0.0645294189453125,
      "learning_rate": 4.989932047938055e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 7792
    },
    {
      "completion_length": 232.91965579986572,
      "epoch": 1.306802464478813,
      "grad_norm": 0.3939521238618866,
      "kl": 0.0809478759765625,
      "learning_rate": 4.989921290192974e-07,
      "loss": 0.0001,
      "reward": 1.7982143312692642,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8026786036789417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7794
    },
    {
      "completion_length": 225.34375953674316,
      "epoch": 1.3071377677186806,
      "grad_norm": 0.14989733463861954,
      "kl": 0.069671630859375,
      "learning_rate": 4.98991052671517e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 7796
    },
    {
      "completion_length": 224.74107933044434,
      "epoch": 1.3074730709585483,
      "grad_norm": 0.22708778689946796,
      "kl": 0.081268310546875,
      "learning_rate": 4.989899757504669e-07,
      "loss": 0.0001,
      "reward": 1.739285796880722,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857372760773,
      "rewards/format_reward_func": 1.0,
      "step": 7798
    },
    {
      "completion_length": 237.43751049041748,
      "epoch": 1.3078083741984157,
      "grad_norm": 0.2108386163132425,
      "kl": 0.1372833251953125,
      "learning_rate": 4.989888982561495e-07,
      "loss": 0.0001,
      "reward": 1.7464286163449287,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7553571779280901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7800
    },
    {
      "completion_length": 234.86161708831787,
      "epoch": 1.3081436774382833,
      "grad_norm": 0.1809852379315868,
      "kl": 0.1102294921875,
      "learning_rate": 4.989878201885674e-07,
      "loss": 0.0001,
      "reward": 1.7285714969038963,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7375000342726707,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7802
    },
    {
      "completion_length": 234.46875858306885,
      "epoch": 1.3084789806781507,
      "grad_norm": 0.2641798629492874,
      "kl": 0.0806427001953125,
      "learning_rate": 4.98986741547723e-07,
      "loss": 0.0001,
      "reward": 1.7504464983940125,
      "reward_std": 0.09028238197788596,
      "rewards/equation_reward_func": 0.7562500443309546,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 7804
    },
    {
      "completion_length": 229.58483219146729,
      "epoch": 1.3088142839180184,
      "grad_norm": 0.1348543022563862,
      "kl": 0.0857391357421875,
      "learning_rate": 4.989856623336188e-07,
      "loss": 0.0001,
      "reward": 1.7464286461472511,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 7806
    },
    {
      "completion_length": 236.95090103149414,
      "epoch": 1.309149587157886,
      "grad_norm": 0.2442230067538243,
      "kl": 0.08929443359375,
      "learning_rate": 4.989845825462574e-07,
      "loss": 0.0001,
      "reward": 1.7410714998841286,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7455357387661934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7808
    },
    {
      "completion_length": 239.0134048461914,
      "epoch": 1.3094848903977534,
      "grad_norm": 0.11169604253691609,
      "kl": 0.089385986328125,
      "learning_rate": 4.98983502185641e-07,
      "loss": 0.0001,
      "reward": 1.7964286357164383,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964285872876644,
      "rewards/format_reward_func": 1.0,
      "step": 7810
    },
    {
      "completion_length": 237.43304634094238,
      "epoch": 1.309820193637621,
      "grad_norm": 0.33155649606336196,
      "kl": 0.0821075439453125,
      "learning_rate": 4.989824212517724e-07,
      "loss": 0.0001,
      "reward": 1.7446429133415222,
      "reward_std": 0.0883883461356163,
      "rewards/equation_reward_func": 0.7580357417464256,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7812
    },
    {
      "completion_length": 242.7500123977661,
      "epoch": 1.3101554968774884,
      "grad_norm": 0.2148534984579552,
      "kl": 0.0830078125,
      "learning_rate": 4.989813397446538e-07,
      "loss": 0.0001,
      "reward": 1.782589353621006,
      "reward_std": 0.039774756878614426,
      "rewards/equation_reward_func": 0.797321442514658,
      "rewards/format_reward_func": 0.9852678626775742,
      "step": 7814
    },
    {
      "completion_length": 236.7142972946167,
      "epoch": 1.310490800117356,
      "grad_norm": 0.2801228926143477,
      "kl": 0.090118408203125,
      "learning_rate": 4.98980257664288e-07,
      "loss": 0.0001,
      "reward": 1.79464291036129,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7991071753203869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7816
    },
    {
      "completion_length": 237.13394165039062,
      "epoch": 1.3108261033572237,
      "grad_norm": 0.21059345642327687,
      "kl": 0.1118316650390625,
      "learning_rate": 4.989791750106772e-07,
      "loss": 0.0001,
      "reward": 1.7325893640518188,
      "reward_std": 0.05492704175412655,
      "rewards/equation_reward_func": 0.7383928783237934,
      "rewards/format_reward_func": 0.994196429848671,
      "step": 7818
    },
    {
      "completion_length": 240.52679634094238,
      "epoch": 1.3111614065970914,
      "grad_norm": 0.22267002073074174,
      "kl": 0.0712890625,
      "learning_rate": 4.989780917838241e-07,
      "loss": 0.0001,
      "reward": 1.7732143327593803,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7776786126196384,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7820
    },
    {
      "completion_length": 231.9375114440918,
      "epoch": 1.3114967098369588,
      "grad_norm": 0.29483440744225925,
      "kl": 0.0724029541015625,
      "learning_rate": 4.989770079837311e-07,
      "loss": 0.0001,
      "reward": 1.7767857536673546,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7812500223517418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7822
    },
    {
      "completion_length": 228.0134048461914,
      "epoch": 1.3118320130768264,
      "grad_norm": 0.18827096899708526,
      "kl": 0.111663818359375,
      "learning_rate": 4.989759236104007e-07,
      "loss": 0.0001,
      "reward": 1.7785714715719223,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 7824
    },
    {
      "completion_length": 228.86608219146729,
      "epoch": 1.3121673163166938,
      "grad_norm": 0.2393960966488759,
      "kl": 0.08990478515625,
      "learning_rate": 4.989748386638354e-07,
      "loss": 0.0001,
      "reward": 1.7339286357164383,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7383928969502449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7826
    },
    {
      "completion_length": 234.2901906967163,
      "epoch": 1.3125026195565614,
      "grad_norm": 0.3691026151200785,
      "kl": 0.083038330078125,
      "learning_rate": 4.989737531440378e-07,
      "loss": 0.0001,
      "reward": 1.7732143625617027,
      "reward_std": 0.138895976357162,
      "rewards/equation_reward_func": 0.8044643178582191,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 7828
    },
    {
      "completion_length": 225.30804538726807,
      "epoch": 1.312837922796429,
      "grad_norm": 0.17235853000671505,
      "kl": 0.159576416015625,
      "learning_rate": 4.989726670510102e-07,
      "loss": 0.0002,
      "reward": 1.7660714834928513,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7705357447266579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7830
    },
    {
      "completion_length": 223.16518783569336,
      "epoch": 1.3131732260362965,
      "grad_norm": 0.20867123219482434,
      "kl": 0.0762176513671875,
      "learning_rate": 4.989715803847553e-07,
      "loss": 0.0001,
      "reward": 1.828571505844593,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8285714462399483,
      "rewards/format_reward_func": 1.0,
      "step": 7832
    },
    {
      "completion_length": 233.1919755935669,
      "epoch": 1.3135085292761641,
      "grad_norm": 0.24279714655234397,
      "kl": 0.09161376953125,
      "learning_rate": 4.989704931452754e-07,
      "loss": 0.0001,
      "reward": 1.8071429058909416,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071429058909416,
      "rewards/format_reward_func": 1.0,
      "step": 7834
    },
    {
      "completion_length": 231.41072368621826,
      "epoch": 1.3138438325160318,
      "grad_norm": 0.16989311883932803,
      "kl": 0.1334381103515625,
      "learning_rate": 4.989694053325732e-07,
      "loss": 0.0001,
      "reward": 1.7285715192556381,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7285714540630579,
      "rewards/format_reward_func": 1.0,
      "step": 7836
    },
    {
      "completion_length": 230.12054634094238,
      "epoch": 1.3141791357558992,
      "grad_norm": 0.26715548521941174,
      "kl": 0.0696868896484375,
      "learning_rate": 4.989683169466511e-07,
      "loss": 0.0001,
      "reward": 1.7982143387198448,
      "reward_std": 0.06313453428447247,
      "rewards/equation_reward_func": 0.8116071671247482,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7838
    },
    {
      "completion_length": 231.8928689956665,
      "epoch": 1.3145144389957668,
      "grad_norm": 0.269292185061531,
      "kl": 0.0734405517578125,
      "learning_rate": 4.989672279875116e-07,
      "loss": 0.0001,
      "reward": 1.7732143551111221,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7866071686148643,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7840
    },
    {
      "completion_length": 235.4910831451416,
      "epoch": 1.3148497422356344,
      "grad_norm": 0.20359318194734094,
      "kl": 0.06927490234375,
      "learning_rate": 4.989661384551573e-07,
      "loss": 0.0001,
      "reward": 1.7500000596046448,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 7842
    },
    {
      "completion_length": 236.55358505249023,
      "epoch": 1.3151850454755019,
      "grad_norm": 0.5545695180093116,
      "kl": 0.2422637939453125,
      "learning_rate": 4.989650483495906e-07,
      "loss": 0.0002,
      "reward": 1.7660715132951736,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7844
    },
    {
      "completion_length": 223.20983123779297,
      "epoch": 1.3155203487153695,
      "grad_norm": 0.5333929645774035,
      "kl": 0.2826690673828125,
      "learning_rate": 4.98963957670814e-07,
      "loss": 0.0003,
      "reward": 1.751785770058632,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500424683094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7846
    },
    {
      "completion_length": 225.40625953674316,
      "epoch": 1.315855651955237,
      "grad_norm": 0.20162221324994153,
      "kl": 0.5379486083984375,
      "learning_rate": 4.9896286641883e-07,
      "loss": 0.0005,
      "reward": 1.7571429386734962,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7660714499652386,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7848
    },
    {
      "completion_length": 240.8705472946167,
      "epoch": 1.3161909551951045,
      "grad_norm": 0.5158776214210989,
      "kl": 0.2147674560546875,
      "learning_rate": 4.989617745936413e-07,
      "loss": 0.0002,
      "reward": 1.6625000685453415,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.6848214585334063,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 7850
    },
    {
      "completion_length": 229.63840293884277,
      "epoch": 1.3165262584349722,
      "grad_norm": 0.3106276145668767,
      "kl": 0.5753936767578125,
      "learning_rate": 4.989606821952503e-07,
      "loss": 0.0006,
      "reward": 1.7892857566475868,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.798214316368103,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7852
    },
    {
      "completion_length": 231.4866180419922,
      "epoch": 1.3168615616748398,
      "grad_norm": 0.20367040132891345,
      "kl": 0.349151611328125,
      "learning_rate": 4.989595892236594e-07,
      "loss": 0.0003,
      "reward": 1.7500000670552254,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7589286006987095,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7854
    },
    {
      "completion_length": 246.17411518096924,
      "epoch": 1.3171968649147072,
      "grad_norm": 0.36064056384011967,
      "kl": 0.149871826171875,
      "learning_rate": 4.989584956788713e-07,
      "loss": 0.0001,
      "reward": 1.6415179520845413,
      "reward_std": 0.10796005232259631,
      "rewards/equation_reward_func": 0.6785714626312256,
      "rewards/format_reward_func": 0.9629464484751225,
      "step": 7856
    },
    {
      "completion_length": 232.4509038925171,
      "epoch": 1.3175321681545749,
      "grad_norm": 0.27023069215816137,
      "kl": 0.454803466796875,
      "learning_rate": 4.989574015608883e-07,
      "loss": 0.0005,
      "reward": 1.6946429386734962,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7169643137603998,
      "rewards/format_reward_func": 0.9776785783469677,
      "step": 7858
    },
    {
      "completion_length": 233.8303680419922,
      "epoch": 1.3178674713944423,
      "grad_norm": 0.3203622502088371,
      "kl": 0.268585205078125,
      "learning_rate": 4.989563068697133e-07,
      "loss": 0.0003,
      "reward": 1.7535715028643608,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 7860
    },
    {
      "completion_length": 230.17411613464355,
      "epoch": 1.31820277463431,
      "grad_norm": 0.29306740052637414,
      "kl": 0.418975830078125,
      "learning_rate": 4.989552116053485e-07,
      "loss": 0.0004,
      "reward": 1.7821429297327995,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7910714522004128,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7862
    },
    {
      "completion_length": 230.42858123779297,
      "epoch": 1.3185380778741775,
      "grad_norm": 0.2370092201280241,
      "kl": 0.0861053466796875,
      "learning_rate": 4.989541157677964e-07,
      "loss": 0.0001,
      "reward": 1.7589286416769028,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7723214514553547,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7864
    },
    {
      "completion_length": 227.3750114440918,
      "epoch": 1.318873381114045,
      "grad_norm": 0.26860458742969134,
      "kl": 0.0887451171875,
      "learning_rate": 4.989530193570596e-07,
      "loss": 0.0001,
      "reward": 1.7214286774396896,
      "reward_std": 0.1010152529925108,
      "rewards/equation_reward_func": 0.7303571738302708,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7866
    },
    {
      "completion_length": 216.43304634094238,
      "epoch": 1.3192086843539126,
      "grad_norm": 0.2723609670431003,
      "kl": 0.082366943359375,
      "learning_rate": 4.989519223731408e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714454948902,
      "rewards/format_reward_func": 1.0,
      "step": 7868
    },
    {
      "completion_length": 218.71876049041748,
      "epoch": 1.31954398759378,
      "grad_norm": 0.38728502112057167,
      "kl": 0.2922515869140625,
      "learning_rate": 4.989508248160423e-07,
      "loss": 0.0003,
      "reward": 1.78035718947649,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7870
    },
    {
      "completion_length": 219.24108123779297,
      "epoch": 1.3198792908336476,
      "grad_norm": 0.27127451546841486,
      "kl": 0.170257568359375,
      "learning_rate": 4.989497266857666e-07,
      "loss": 0.0002,
      "reward": 1.730357214808464,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7348214574158192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7872
    },
    {
      "completion_length": 219.01340198516846,
      "epoch": 1.3202145940735153,
      "grad_norm": 0.283198873620819,
      "kl": 0.092987060546875,
      "learning_rate": 4.989486279823164e-07,
      "loss": 0.0001,
      "reward": 1.7946429327130318,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7991071678698063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7874
    },
    {
      "completion_length": 222.977689743042,
      "epoch": 1.320549897313383,
      "grad_norm": 0.29053446930371934,
      "kl": 0.107940673828125,
      "learning_rate": 4.989475287056941e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 7876
    },
    {
      "completion_length": 221.5089406967163,
      "epoch": 1.3208852005532503,
      "grad_norm": 0.31560908109020885,
      "kl": 0.1349945068359375,
      "learning_rate": 4.989464288559024e-07,
      "loss": 0.0001,
      "reward": 1.7357143685221672,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143238186836,
      "rewards/format_reward_func": 1.0,
      "step": 7878
    },
    {
      "completion_length": 208.75000953674316,
      "epoch": 1.321220503793118,
      "grad_norm": 0.23600630732348715,
      "kl": 0.0809326171875,
      "learning_rate": 4.989453284329436e-07,
      "loss": 0.0001,
      "reward": 1.7379464954137802,
      "reward_std": 0.03724937466904521,
      "rewards/equation_reward_func": 0.7392857316881418,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 7880
    },
    {
      "completion_length": 216.7232265472412,
      "epoch": 1.3215558070329854,
      "grad_norm": 0.3341347334253455,
      "kl": 0.1131591796875,
      "learning_rate": 4.989442274368204e-07,
      "loss": 0.0001,
      "reward": 1.7803571969270706,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214693367481,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7882
    },
    {
      "completion_length": 215.28125953674316,
      "epoch": 1.321891110272853,
      "grad_norm": 0.1194486987207903,
      "kl": 0.1649322509765625,
      "learning_rate": 4.989431258675353e-07,
      "loss": 0.0002,
      "reward": 1.7571429088711739,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 1.0,
      "step": 7884
    },
    {
      "completion_length": 213.87054538726807,
      "epoch": 1.3222264135127206,
      "grad_norm": 0.16692747952837317,
      "kl": 0.9911956787109375,
      "learning_rate": 4.989420237250909e-07,
      "loss": 0.001,
      "reward": 1.7892857789993286,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857398837805,
      "rewards/format_reward_func": 1.0,
      "step": 7886
    },
    {
      "completion_length": 208.29911613464355,
      "epoch": 1.322561716752588,
      "grad_norm": 0.43172098055364655,
      "kl": 0.154571533203125,
      "learning_rate": 4.989409210094895e-07,
      "loss": 0.0002,
      "reward": 1.798214353621006,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8026785925030708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7888
    },
    {
      "completion_length": 211.33036518096924,
      "epoch": 1.3228970199924557,
      "grad_norm": 0.33694847774059483,
      "kl": 0.096923828125,
      "learning_rate": 4.989398177207339e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7321428749710321,
      "rewards/format_reward_func": 1.0,
      "step": 7890
    },
    {
      "completion_length": 211.00893783569336,
      "epoch": 1.3232323232323233,
      "grad_norm": 0.29598731413356466,
      "kl": 0.2599945068359375,
      "learning_rate": 4.989387138588265e-07,
      "loss": 0.0003,
      "reward": 1.8017857521772385,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.806250024586916,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7892
    },
    {
      "completion_length": 220.1741180419922,
      "epoch": 1.3235676264721907,
      "grad_norm": 0.2904979857112134,
      "kl": 0.0965576171875,
      "learning_rate": 4.989376094237699e-07,
      "loss": 0.0001,
      "reward": 1.717857226729393,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7267857305705547,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7894
    },
    {
      "completion_length": 220.26786613464355,
      "epoch": 1.3239029297120584,
      "grad_norm": 0.27770757884590075,
      "kl": 0.145721435546875,
      "learning_rate": 4.989365044155667e-07,
      "loss": 0.0001,
      "reward": 1.6714286506175995,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.6803571786731482,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7896
    },
    {
      "completion_length": 217.25447368621826,
      "epoch": 1.324238232951926,
      "grad_norm": 0.32922269503566337,
      "kl": 0.196502685546875,
      "learning_rate": 4.989353988342192e-07,
      "loss": 0.0002,
      "reward": 1.7392857894301414,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.748214315623045,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7898
    },
    {
      "completion_length": 223.9107265472412,
      "epoch": 1.3245735361917934,
      "grad_norm": 0.998183898998436,
      "kl": 0.1063232421875,
      "learning_rate": 4.989342926797303e-07,
      "loss": 0.0001,
      "reward": 1.7660714834928513,
      "reward_std": 0.0883883461356163,
      "rewards/equation_reward_func": 0.7794643230736256,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7900
    },
    {
      "completion_length": 226.12054538726807,
      "epoch": 1.324908839431661,
      "grad_norm": 0.1907288726707059,
      "kl": 0.097930908203125,
      "learning_rate": 4.989331859521022e-07,
      "loss": 0.0001,
      "reward": 1.7178572043776512,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7267857529222965,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7902
    },
    {
      "completion_length": 224.27233219146729,
      "epoch": 1.3252441426715285,
      "grad_norm": 2.421709001433735,
      "kl": 0.112579345703125,
      "learning_rate": 4.989320786513376e-07,
      "loss": 0.0001,
      "reward": 1.7267857864499092,
      "reward_std": 0.09343910869210958,
      "rewards/equation_reward_func": 0.740178607404232,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7904
    },
    {
      "completion_length": 226.09822463989258,
      "epoch": 1.325579445911396,
      "grad_norm": 1.1546552639652499,
      "kl": 0.12115478515625,
      "learning_rate": 4.989309707774391e-07,
      "loss": 0.0001,
      "reward": 1.769642896950245,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.7830357477068901,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7906
    },
    {
      "completion_length": 221.54018878936768,
      "epoch": 1.3259147491512637,
      "grad_norm": 0.2796337690765062,
      "kl": 0.14959716796875,
      "learning_rate": 4.989298623304094e-07,
      "loss": 0.0001,
      "reward": 1.7964286357164383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 7908
    },
    {
      "completion_length": 210.87501049041748,
      "epoch": 1.3262500523911314,
      "grad_norm": 0.3019066657629454,
      "kl": 0.1559600830078125,
      "learning_rate": 4.989287533102506e-07,
      "loss": 0.0002,
      "reward": 1.8178572058677673,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8178571611642838,
      "rewards/format_reward_func": 1.0,
      "step": 7910
    },
    {
      "completion_length": 220.26340103149414,
      "epoch": 1.3265853556309988,
      "grad_norm": 0.34559487640089015,
      "kl": 0.151519775390625,
      "learning_rate": 4.989276437169656e-07,
      "loss": 0.0002,
      "reward": 1.7535714954137802,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 7912
    },
    {
      "completion_length": 213.41518688201904,
      "epoch": 1.3269206588708664,
      "grad_norm": 0.292567252173144,
      "kl": 0.08819580078125,
      "learning_rate": 4.989265335505569e-07,
      "loss": 0.0001,
      "reward": 1.760714367032051,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143074274063,
      "rewards/format_reward_func": 1.0,
      "step": 7914
    },
    {
      "completion_length": 208.04465103149414,
      "epoch": 1.3272559621107338,
      "grad_norm": 0.3393569470550142,
      "kl": 0.0819091796875,
      "learning_rate": 4.989254228110271e-07,
      "loss": 0.0001,
      "reward": 1.7714286595582962,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 7916
    },
    {
      "completion_length": 214.92858028411865,
      "epoch": 1.3275912653506015,
      "grad_norm": 0.0855442589689345,
      "kl": 0.0947113037109375,
      "learning_rate": 4.989243114983785e-07,
      "loss": 0.0001,
      "reward": 1.8142857775092125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857551574707,
      "rewards/format_reward_func": 1.0,
      "step": 7918
    },
    {
      "completion_length": 214.98661708831787,
      "epoch": 1.327926568590469,
      "grad_norm": 0.23092207006843854,
      "kl": 0.0896759033203125,
      "learning_rate": 4.98923199612614e-07,
      "loss": 0.0001,
      "reward": 1.7267857939004898,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7401785831898451,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 7920
    },
    {
      "completion_length": 211.44197273254395,
      "epoch": 1.3282618718303365,
      "grad_norm": 0.2068555217674292,
      "kl": 0.0815582275390625,
      "learning_rate": 4.989220871537359e-07,
      "loss": 0.0001,
      "reward": 1.7750000730156898,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 7922
    },
    {
      "completion_length": 212.82590198516846,
      "epoch": 1.3285971750702041,
      "grad_norm": 0.28652744944235065,
      "kl": 0.0796966552734375,
      "learning_rate": 4.98920974121747e-07,
      "loss": 0.0001,
      "reward": 1.8321429044008255,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8321428708732128,
      "rewards/format_reward_func": 1.0,
      "step": 7924
    },
    {
      "completion_length": 206.47768688201904,
      "epoch": 1.3289324783100716,
      "grad_norm": 0.29701924863495033,
      "kl": 0.09295654296875,
      "learning_rate": 4.989198605166495e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 7926
    },
    {
      "completion_length": 210.23215103149414,
      "epoch": 1.3292677815499392,
      "grad_norm": 0.39862120982234434,
      "kl": 0.09503173828125,
      "learning_rate": 4.989187463384463e-07,
      "loss": 0.0001,
      "reward": 1.7250000834465027,
      "reward_std": 0.0858629634603858,
      "rewards/equation_reward_func": 0.7250000350177288,
      "rewards/format_reward_func": 1.0,
      "step": 7928
    },
    {
      "completion_length": 223.2232255935669,
      "epoch": 1.3296030847898068,
      "grad_norm": 0.3389688200507532,
      "kl": 0.0911407470703125,
      "learning_rate": 4.989176315871398e-07,
      "loss": 0.0001,
      "reward": 1.74821437895298,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7930
    },
    {
      "completion_length": 211.71429538726807,
      "epoch": 1.3299383880296745,
      "grad_norm": 0.30581841187049746,
      "kl": 0.0901336669921875,
      "learning_rate": 4.989165162627328e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.767857164144516,
      "rewards/format_reward_func": 1.0,
      "step": 7932
    },
    {
      "completion_length": 223.56251049041748,
      "epoch": 1.3302736912695419,
      "grad_norm": 0.3911307845060985,
      "kl": 0.0928497314453125,
      "learning_rate": 4.989154003652275e-07,
      "loss": 0.0001,
      "reward": 1.7928572222590446,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 7934
    },
    {
      "completion_length": 214.97322463989258,
      "epoch": 1.3306089945094095,
      "grad_norm": 0.3465375626222913,
      "kl": 0.102874755859375,
      "learning_rate": 4.989142838946267e-07,
      "loss": 0.0001,
      "reward": 1.7428572252392769,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571619093418,
      "rewards/format_reward_func": 1.0,
      "step": 7936
    },
    {
      "completion_length": 215.17411708831787,
      "epoch": 1.330944297749277,
      "grad_norm": 0.2738688404966579,
      "kl": 0.0842132568359375,
      "learning_rate": 4.98913166850933e-07,
      "loss": 0.0001,
      "reward": 1.792857214808464,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571589291096,
      "rewards/format_reward_func": 1.0,
      "step": 7938
    },
    {
      "completion_length": 220.07590293884277,
      "epoch": 1.3312796009891446,
      "grad_norm": 0.20492354543887023,
      "kl": 0.379638671875,
      "learning_rate": 4.989120492341489e-07,
      "loss": 0.0004,
      "reward": 1.76071435213089,
      "reward_std": 0.05555838719010353,
      "rewards/equation_reward_func": 0.7607142999768257,
      "rewards/format_reward_func": 1.0,
      "step": 7940
    },
    {
      "completion_length": 215.39733123779297,
      "epoch": 1.3316149042290122,
      "grad_norm": 0.30510096492084604,
      "kl": 0.0955047607421875,
      "learning_rate": 4.989109310442769e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 7942
    },
    {
      "completion_length": 218.67858028411865,
      "epoch": 1.3319502074688796,
      "grad_norm": 0.34981667176896825,
      "kl": 6.00408935546875,
      "learning_rate": 4.989098122813197e-07,
      "loss": 0.006,
      "reward": 1.7892857864499092,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 7944
    },
    {
      "completion_length": 206.76786708831787,
      "epoch": 1.3322855107087472,
      "grad_norm": 0.5514828094624111,
      "kl": 0.338348388671875,
      "learning_rate": 4.989086929452797e-07,
      "loss": 0.0003,
      "reward": 1.7678572088479996,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.767857177183032,
      "rewards/format_reward_func": 1.0,
      "step": 7946
    },
    {
      "completion_length": 218.44197273254395,
      "epoch": 1.3326208139486146,
      "grad_norm": 0.21252407205890636,
      "kl": 0.2454071044921875,
      "learning_rate": 4.989075730361598e-07,
      "loss": 0.0002,
      "reward": 1.7785714715719223,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 7948
    },
    {
      "completion_length": 212.97768688201904,
      "epoch": 1.3329561171884823,
      "grad_norm": 0.24461102951428326,
      "kl": 0.1788330078125,
      "learning_rate": 4.989064525539622e-07,
      "loss": 0.0002,
      "reward": 1.7928571850061417,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571719676256,
      "rewards/format_reward_func": 1.0,
      "step": 7950
    },
    {
      "completion_length": 213.28126049041748,
      "epoch": 1.33329142042835,
      "grad_norm": 0.25165262044320724,
      "kl": 0.157684326171875,
      "learning_rate": 4.989053314986898e-07,
      "loss": 0.0002,
      "reward": 1.7964286357164383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286059141159,
      "rewards/format_reward_func": 1.0,
      "step": 7952
    },
    {
      "completion_length": 218.63840007781982,
      "epoch": 1.3336267236682176,
      "grad_norm": 0.3567360154219791,
      "kl": 0.103912353515625,
      "learning_rate": 4.989042098703449e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 7954
    },
    {
      "completion_length": 222.22768783569336,
      "epoch": 1.333962026908085,
      "grad_norm": 0.251345285427576,
      "kl": 0.21942138671875,
      "learning_rate": 4.989030876689303e-07,
      "loss": 0.0002,
      "reward": 1.7892857864499092,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 7956
    },
    {
      "completion_length": 233.54465293884277,
      "epoch": 1.3342973301479526,
      "grad_norm": 0.3240709202095466,
      "kl": 0.26336669921875,
      "learning_rate": 4.989019648944486e-07,
      "loss": 0.0003,
      "reward": 1.739285796880722,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 1.0,
      "step": 7958
    },
    {
      "completion_length": 227.4107255935669,
      "epoch": 1.33463263338782,
      "grad_norm": 0.3387491696363639,
      "kl": 0.09735107421875,
      "learning_rate": 4.989008415469022e-07,
      "loss": 0.0001,
      "reward": 1.8178572058677673,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8178571630269289,
      "rewards/format_reward_func": 1.0,
      "step": 7960
    },
    {
      "completion_length": 239.0803689956665,
      "epoch": 1.3349679366276876,
      "grad_norm": 0.3606421949515841,
      "kl": 0.183502197265625,
      "learning_rate": 4.988997176262937e-07,
      "loss": 0.0002,
      "reward": 1.7642858028411865,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857320606709,
      "rewards/format_reward_func": 1.0,
      "step": 7962
    },
    {
      "completion_length": 229.78126049041748,
      "epoch": 1.3353032398675553,
      "grad_norm": 0.16454299451021592,
      "kl": 0.39007568359375,
      "learning_rate": 4.988985931326259e-07,
      "loss": 0.0004,
      "reward": 1.7392857894301414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857372760773,
      "rewards/format_reward_func": 1.0,
      "step": 7964
    },
    {
      "completion_length": 231.1250114440918,
      "epoch": 1.3356385431074227,
      "grad_norm": 0.4889025325932433,
      "kl": 0.591400146484375,
      "learning_rate": 4.988974680659012e-07,
      "loss": 0.0006,
      "reward": 1.7964286357164383,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 7966
    },
    {
      "completion_length": 232.08483409881592,
      "epoch": 1.3359738463472903,
      "grad_norm": 0.40254854105558957,
      "kl": 0.818389892578125,
      "learning_rate": 4.988963424261221e-07,
      "loss": 0.0008,
      "reward": 1.772321529686451,
      "reward_std": 0.06944798585027456,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 0.9973214343190193,
      "step": 7968
    },
    {
      "completion_length": 230.89733600616455,
      "epoch": 1.336309149587158,
      "grad_norm": 0.43390298939426425,
      "kl": 1.7117919921875,
      "learning_rate": 4.988952162132916e-07,
      "loss": 0.0017,
      "reward": 1.779464341700077,
      "reward_std": 0.06944798701442778,
      "rewards/equation_reward_func": 0.7821428962051868,
      "rewards/format_reward_func": 0.9973214343190193,
      "step": 7970
    },
    {
      "completion_length": 222.60715293884277,
      "epoch": 1.3366444528270254,
      "grad_norm": 0.2671454665293245,
      "kl": 0.3524322509765625,
      "learning_rate": 4.988940894274118e-07,
      "loss": 0.0004,
      "reward": 1.81428574770689,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857365310192,
      "rewards/format_reward_func": 1.0,
      "step": 7972
    },
    {
      "completion_length": 231.09375953674316,
      "epoch": 1.336979756066893,
      "grad_norm": 0.20695948130143144,
      "kl": 0.834014892578125,
      "learning_rate": 4.988929620684857e-07,
      "loss": 0.0008,
      "reward": 1.7558036223053932,
      "reward_std": 0.032198611879721284,
      "rewards/equation_reward_func": 0.757142897695303,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 7974
    },
    {
      "completion_length": 231.2187623977661,
      "epoch": 1.3373150593067606,
      "grad_norm": 0.25623959963115334,
      "kl": 0.179534912109375,
      "learning_rate": 4.988918341365156e-07,
      "loss": 0.0002,
      "reward": 1.7464286461472511,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7553571574389935,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 7976
    },
    {
      "completion_length": 232.86161708831787,
      "epoch": 1.337650362546628,
      "grad_norm": 0.07860704228809769,
      "kl": 0.182830810546875,
      "learning_rate": 4.988907056315043e-07,
      "loss": 0.0002,
      "reward": 1.7272322252392769,
      "reward_std": 0.03219861118122935,
      "rewards/equation_reward_func": 0.7285714615136385,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 7978
    },
    {
      "completion_length": 241.88394165039062,
      "epoch": 1.3379856657864957,
      "grad_norm": 0.3020735349351519,
      "kl": 0.331207275390625,
      "learning_rate": 4.988895765534542e-07,
      "loss": 0.0003,
      "reward": 1.7017857879400253,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7062500528991222,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7980
    },
    {
      "completion_length": 235.10268878936768,
      "epoch": 1.338320969026363,
      "grad_norm": 1.859077809209031,
      "kl": 0.191925048828125,
      "learning_rate": 4.98888446902368e-07,
      "loss": 0.0002,
      "reward": 1.7428572252392769,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571693599224,
      "rewards/format_reward_func": 1.0,
      "step": 7982
    },
    {
      "completion_length": 242.65180015563965,
      "epoch": 1.3386562722662307,
      "grad_norm": 0.3395116480501293,
      "kl": 0.2183990478515625,
      "learning_rate": 4.988873166782484e-07,
      "loss": 0.0002,
      "reward": 1.7700893580913544,
      "reward_std": 0.07260471256449819,
      "rewards/equation_reward_func": 0.7714285887777805,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 7984
    },
    {
      "completion_length": 244.06697463989258,
      "epoch": 1.3389915755060984,
      "grad_norm": 0.30610268249495964,
      "kl": 0.224273681640625,
      "learning_rate": 4.98886185881098e-07,
      "loss": 0.0002,
      "reward": 1.7392858117818832,
      "reward_std": 0.09596448857337236,
      "rewards/equation_reward_func": 0.7392857372760773,
      "rewards/format_reward_func": 1.0,
      "step": 7986
    },
    {
      "completion_length": 248.25001049041748,
      "epoch": 1.339326878745966,
      "grad_norm": 0.23843987852462686,
      "kl": 0.1561737060546875,
      "learning_rate": 4.988850545109193e-07,
      "loss": 0.0002,
      "reward": 1.7357143461704254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.735714316368103,
      "rewards/format_reward_func": 1.0,
      "step": 7988
    },
    {
      "completion_length": 247.65180015563965,
      "epoch": 1.3396621819858334,
      "grad_norm": 0.2668127447982816,
      "kl": 0.113250732421875,
      "learning_rate": 4.988839225677147e-07,
      "loss": 0.0001,
      "reward": 1.7589286416769028,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7633928917348385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7990
    },
    {
      "completion_length": 243.83483123779297,
      "epoch": 1.339997485225701,
      "grad_norm": 0.17925726284968613,
      "kl": 0.08819580078125,
      "learning_rate": 4.988827900514873e-07,
      "loss": 0.0001,
      "reward": 1.7482143566012383,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.752678606659174,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7992
    },
    {
      "completion_length": 238.2321548461914,
      "epoch": 1.3403327884655685,
      "grad_norm": 0.17101341252068045,
      "kl": 0.1177520751953125,
      "learning_rate": 4.988816569622392e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 7994
    },
    {
      "completion_length": 232.06697463989258,
      "epoch": 1.340668091705436,
      "grad_norm": 0.22431370698058403,
      "kl": 0.083343505859375,
      "learning_rate": 4.988805232999735e-07,
      "loss": 0.0001,
      "reward": 1.8160714730620384,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8205357417464256,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 7996
    },
    {
      "completion_length": 240.4375123977661,
      "epoch": 1.3410033949453037,
      "grad_norm": 0.2647862614895852,
      "kl": 0.082122802734375,
      "learning_rate": 4.988793890646924e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857640981674,
      "rewards/format_reward_func": 1.0,
      "step": 7998
    },
    {
      "completion_length": 235.17858123779297,
      "epoch": 1.3413386981851712,
      "grad_norm": 0.18667197312641198,
      "kl": 0.0843658447265625,
      "learning_rate": 4.988782542563988e-07,
      "loss": 0.0001,
      "reward": 1.7392857819795609,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857447266579,
      "rewards/format_reward_func": 1.0,
      "step": 8000
    },
    {
      "completion_length": 243.3928689956665,
      "epoch": 1.3416740014250388,
      "grad_norm": 0.30499551404800335,
      "kl": 0.139678955078125,
      "learning_rate": 4.988771188750949e-07,
      "loss": 0.0001,
      "reward": 1.835714377462864,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8357142992317677,
      "rewards/format_reward_func": 1.0,
      "step": 8002
    },
    {
      "completion_length": 248.0000123977661,
      "epoch": 1.3420093046649062,
      "grad_norm": 0.2519452088794451,
      "kl": 0.1127777099609375,
      "learning_rate": 4.988759829207839e-07,
      "loss": 0.0001,
      "reward": 1.7821429371833801,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 8004
    },
    {
      "completion_length": 234.8169755935669,
      "epoch": 1.3423446079047738,
      "grad_norm": 0.2282132101143147,
      "kl": 0.1110076904296875,
      "learning_rate": 4.98874846393468e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571790456772,
      "rewards/format_reward_func": 1.0,
      "step": 8006
    },
    {
      "completion_length": 240.383939743042,
      "epoch": 1.3426799111446415,
      "grad_norm": 0.3600678021417168,
      "kl": 0.17425537109375,
      "learning_rate": 4.988737092931499e-07,
      "loss": 0.0002,
      "reward": 1.7232143729925156,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7276786044239998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8008
    },
    {
      "completion_length": 247.48662090301514,
      "epoch": 1.343015214384509,
      "grad_norm": 0.3601856091328208,
      "kl": 0.108642578125,
      "learning_rate": 4.988725716198322e-07,
      "loss": 0.0001,
      "reward": 1.7714286521077156,
      "reward_std": 0.09091372601687908,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 8010
    },
    {
      "completion_length": 234.50447368621826,
      "epoch": 1.3433505176243765,
      "grad_norm": 0.30609978360331364,
      "kl": 0.09759521484375,
      "learning_rate": 4.988714333735176e-07,
      "loss": 0.0001,
      "reward": 1.8035714849829674,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 8012
    },
    {
      "completion_length": 229.9419765472412,
      "epoch": 1.3436858208642442,
      "grad_norm": 0.12760156152712052,
      "kl": 0.085052490234375,
      "learning_rate": 4.988702945542088e-07,
      "loss": 0.0001,
      "reward": 1.8321429193019867,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8321428783237934,
      "rewards/format_reward_func": 1.0,
      "step": 8014
    },
    {
      "completion_length": 243.6428680419922,
      "epoch": 1.3440211241041116,
      "grad_norm": 0.39674180978928336,
      "kl": 0.156829833984375,
      "learning_rate": 4.988691551619081e-07,
      "loss": 0.0002,
      "reward": 1.7517857775092125,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7562500461935997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8016
    },
    {
      "completion_length": 233.2544755935669,
      "epoch": 1.3443564273439792,
      "grad_norm": 0.16498579602363317,
      "kl": 0.084625244140625,
      "learning_rate": 4.988680151966186e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 8018
    },
    {
      "completion_length": 240.07590293884277,
      "epoch": 1.3446917305838468,
      "grad_norm": 0.26198383378203854,
      "kl": 0.0823822021484375,
      "learning_rate": 4.988668746583424e-07,
      "loss": 0.0001,
      "reward": 1.7517857626080513,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500350177288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8020
    },
    {
      "completion_length": 243.4553680419922,
      "epoch": 1.3450270338237142,
      "grad_norm": 0.27333683560494554,
      "kl": 0.0854644775390625,
      "learning_rate": 4.988657335470826e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 8022
    },
    {
      "completion_length": 246.2901906967163,
      "epoch": 1.3453623370635819,
      "grad_norm": 0.43851686807388973,
      "kl": 0.092041015625,
      "learning_rate": 4.988645918628414e-07,
      "loss": 0.0001,
      "reward": 1.789285771548748,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 8024
    },
    {
      "completion_length": 235.3928680419922,
      "epoch": 1.3456976403034493,
      "grad_norm": 0.30842295436356104,
      "kl": 0.08282470703125,
      "learning_rate": 4.988634496056218e-07,
      "loss": 0.0001,
      "reward": 1.7982143387198448,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.8026785999536514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8026
    },
    {
      "completion_length": 238.07590293884277,
      "epoch": 1.346032943543317,
      "grad_norm": 0.12391568944034573,
      "kl": 0.0708465576171875,
      "learning_rate": 4.988623067754262e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 8028
    },
    {
      "completion_length": 243.37947273254395,
      "epoch": 1.3463682467831846,
      "grad_norm": 0.18952970799683763,
      "kl": 0.0840911865234375,
      "learning_rate": 4.988611633722573e-07,
      "loss": 0.0001,
      "reward": 1.7357143834233284,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.735714316368103,
      "rewards/format_reward_func": 1.0,
      "step": 8030
    },
    {
      "completion_length": 247.5401906967163,
      "epoch": 1.3467035500230522,
      "grad_norm": 0.22351549042268146,
      "kl": 0.08148193359375,
      "learning_rate": 4.988600193961177e-07,
      "loss": 0.0001,
      "reward": 1.7678572237491608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571790456772,
      "rewards/format_reward_func": 1.0,
      "step": 8032
    },
    {
      "completion_length": 243.38393688201904,
      "epoch": 1.3470388532629196,
      "grad_norm": 0.37835809624396216,
      "kl": 0.085357666015625,
      "learning_rate": 4.988588748470101e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 8034
    },
    {
      "completion_length": 238.62054634094238,
      "epoch": 1.3473741565027872,
      "grad_norm": 0.23108849325790354,
      "kl": 0.08673095703125,
      "learning_rate": 4.988577297249371e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428772062063,
      "rewards/format_reward_func": 1.0,
      "step": 8036
    },
    {
      "completion_length": 245.7053680419922,
      "epoch": 1.3477094597426547,
      "grad_norm": 0.0852437054563733,
      "kl": 0.0871124267578125,
      "learning_rate": 4.988565840299014e-07,
      "loss": 0.0001,
      "reward": 1.6982143744826317,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7026786133646965,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8038
    },
    {
      "completion_length": 256.2857246398926,
      "epoch": 1.3480447629825223,
      "grad_norm": 0.27334831001457666,
      "kl": 0.09234619140625,
      "learning_rate": 4.988554377619054e-07,
      "loss": 0.0001,
      "reward": 1.7214286252856255,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7214286103844643,
      "rewards/format_reward_func": 1.0,
      "step": 8040
    },
    {
      "completion_length": 240.62500953674316,
      "epoch": 1.34838006622239,
      "grad_norm": 0.3848916028092399,
      "kl": 0.1457366943359375,
      "learning_rate": 4.988542909209521e-07,
      "loss": 0.0001,
      "reward": 1.8000000789761543,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.800000037997961,
      "rewards/format_reward_func": 1.0,
      "step": 8042
    },
    {
      "completion_length": 237.4375123977661,
      "epoch": 1.3487153694622576,
      "grad_norm": 0.24599279522305645,
      "kl": 0.0876922607421875,
      "learning_rate": 4.988531435070438e-07,
      "loss": 0.0001,
      "reward": 1.8000000640749931,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000193715096,
      "rewards/format_reward_func": 1.0,
      "step": 8044
    },
    {
      "completion_length": 258.0759038925171,
      "epoch": 1.349050672702125,
      "grad_norm": 0.4030775726183148,
      "kl": 0.0973968505859375,
      "learning_rate": 4.988519955201834e-07,
      "loss": 0.0001,
      "reward": 1.710714377462864,
      "reward_std": 0.1060660146176815,
      "rewards/equation_reward_func": 0.7196428924798965,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8046
    },
    {
      "completion_length": 248.3214406967163,
      "epoch": 1.3493859759419926,
      "grad_norm": 0.30576816556802056,
      "kl": 0.12847900390625,
      "learning_rate": 4.988508469603735e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 8048
    },
    {
      "completion_length": 241.42858123779297,
      "epoch": 1.34972127918186,
      "grad_norm": 0.26936985666206786,
      "kl": 0.1039581298828125,
      "learning_rate": 4.988496978276166e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571850061417,
      "rewards/format_reward_func": 1.0,
      "step": 8050
    },
    {
      "completion_length": 255.56251525878906,
      "epoch": 1.3500565824217277,
      "grad_norm": 0.32989257568241814,
      "kl": 0.133758544921875,
      "learning_rate": 4.988485481219154e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7392857354134321,
      "rewards/format_reward_func": 1.0,
      "step": 8052
    },
    {
      "completion_length": 242.9866189956665,
      "epoch": 1.3503918856615953,
      "grad_norm": 0.44298681376559773,
      "kl": 0.0977783203125,
      "learning_rate": 4.988473978432725e-07,
      "loss": 0.0001,
      "reward": 1.8000000789761543,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000081956387,
      "rewards/format_reward_func": 1.0,
      "step": 8054
    },
    {
      "completion_length": 257.57590198516846,
      "epoch": 1.3507271889014627,
      "grad_norm": 0.380273329373798,
      "kl": 0.085113525390625,
      "learning_rate": 4.988462469916908e-07,
      "loss": 0.0001,
      "reward": 1.7982143312692642,
      "reward_std": 0.0732360603287816,
      "rewards/equation_reward_func": 0.8116071745753288,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8056
    },
    {
      "completion_length": 251.74108505249023,
      "epoch": 1.3510624921413303,
      "grad_norm": 0.23128181541135234,
      "kl": 0.1066436767578125,
      "learning_rate": 4.988450955671727e-07,
      "loss": 0.0001,
      "reward": 1.7696429044008255,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.774107176810503,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8058
    },
    {
      "completion_length": 239.02679824829102,
      "epoch": 1.3513977953811978,
      "grad_norm": 0.3129928101399441,
      "kl": 0.1588897705078125,
      "learning_rate": 4.988439435697209e-07,
      "loss": 0.0002,
      "reward": 1.7500000819563866,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7589286118745804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8060
    },
    {
      "completion_length": 244.73215293884277,
      "epoch": 1.3517330986210654,
      "grad_norm": 0.31500260650227535,
      "kl": 0.1183929443359375,
      "learning_rate": 4.98842790999338e-07,
      "loss": 0.0001,
      "reward": 1.7964286133646965,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7964286096394062,
      "rewards/format_reward_func": 1.0,
      "step": 8062
    },
    {
      "completion_length": 266.3928699493408,
      "epoch": 1.352068401860933,
      "grad_norm": 0.19326811694448875,
      "kl": 0.0916748046875,
      "learning_rate": 4.98841637856027e-07,
      "loss": 0.0001,
      "reward": 1.7910714820027351,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7955357357859612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8064
    },
    {
      "completion_length": 251.99554824829102,
      "epoch": 1.3524037051008007,
      "grad_norm": 0.28242915098566646,
      "kl": 0.0901641845703125,
      "learning_rate": 4.9884048413979e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 8066
    },
    {
      "completion_length": 254.74108409881592,
      "epoch": 1.352739008340668,
      "grad_norm": 0.21990591678680338,
      "kl": 0.1272125244140625,
      "learning_rate": 4.988393298506301e-07,
      "loss": 0.0001,
      "reward": 1.7267858013510704,
      "reward_std": 0.0732360603287816,
      "rewards/equation_reward_func": 0.7401785962283611,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8068
    },
    {
      "completion_length": 261.4910831451416,
      "epoch": 1.3530743115805357,
      "grad_norm": 0.19448140488390295,
      "kl": 0.101318359375,
      "learning_rate": 4.988381749885498e-07,
      "loss": 0.0001,
      "reward": 1.8000000417232513,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.800000037997961,
      "rewards/format_reward_func": 1.0,
      "step": 8070
    },
    {
      "completion_length": 260.87501335144043,
      "epoch": 1.3534096148204031,
      "grad_norm": 0.3282465087207066,
      "kl": 0.15740966796875,
      "learning_rate": 4.988370195535517e-07,
      "loss": 0.0002,
      "reward": 1.7375000640749931,
      "reward_std": 0.0883883461356163,
      "rewards/equation_reward_func": 0.7508928887546062,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8072
    },
    {
      "completion_length": 268.11162185668945,
      "epoch": 1.3537449180602708,
      "grad_norm": 1.0868446090043193,
      "kl": 0.2178192138671875,
      "learning_rate": 4.988358635456385e-07,
      "loss": 0.0002,
      "reward": 1.721428632736206,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7303571570664644,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8074
    },
    {
      "completion_length": 260.8928680419922,
      "epoch": 1.3540802213001384,
      "grad_norm": 0.16292994704342872,
      "kl": 0.0915069580078125,
      "learning_rate": 4.988347069648129e-07,
      "loss": 0.0001,
      "reward": 1.787500075995922,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7919643074274063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8076
    },
    {
      "completion_length": 269.1830472946167,
      "epoch": 1.3544155245400058,
      "grad_norm": 0.2766759911148509,
      "kl": 0.10125732421875,
      "learning_rate": 4.988335498110776e-07,
      "loss": 0.0001,
      "reward": 1.7321429252624512,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7410714700818062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8078
    },
    {
      "completion_length": 270.62947940826416,
      "epoch": 1.3547508277798734,
      "grad_norm": 0.24982035488239168,
      "kl": 0.084320068359375,
      "learning_rate": 4.988323920844352e-07,
      "loss": 0.0001,
      "reward": 1.7964286506175995,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 8080
    },
    {
      "completion_length": 262.83037090301514,
      "epoch": 1.3550861310197408,
      "grad_norm": 0.12586729994994952,
      "kl": 0.0821075439453125,
      "learning_rate": 4.988312337848883e-07,
      "loss": 0.0001,
      "reward": 1.7125000804662704,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7169643081724644,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8082
    },
    {
      "completion_length": 277.20983695983887,
      "epoch": 1.3554214342596085,
      "grad_norm": 0.35417812775222424,
      "kl": 0.243377685546875,
      "learning_rate": 4.988300749124397e-07,
      "loss": 0.0002,
      "reward": 1.6589286550879478,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.6723214648663998,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8084
    },
    {
      "completion_length": 268.1919755935669,
      "epoch": 1.3557567374994761,
      "grad_norm": 0.35492565059691283,
      "kl": 0.2265167236328125,
      "learning_rate": 4.988289154670919e-07,
      "loss": 0.0002,
      "reward": 1.6982143893837929,
      "reward_std": 0.11364216171205044,
      "rewards/equation_reward_func": 0.7205357477068901,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 8086
    },
    {
      "completion_length": 275.0803689956665,
      "epoch": 1.3560920407393438,
      "grad_norm": 0.12388342800816557,
      "kl": 0.2826995849609375,
      "learning_rate": 4.988277554488478e-07,
      "loss": 0.0003,
      "reward": 1.7517857998609543,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7562500312924385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8088
    },
    {
      "completion_length": 258.0134038925171,
      "epoch": 1.3564273439792112,
      "grad_norm": 0.3590281267788517,
      "kl": 0.2235107421875,
      "learning_rate": 4.988265948577099e-07,
      "loss": 0.0002,
      "reward": 1.7767857760190964,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.7901785932481289,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8090
    },
    {
      "completion_length": 263.65625953674316,
      "epoch": 1.3567626472190788,
      "grad_norm": 0.2747411302913466,
      "kl": 0.1129913330078125,
      "learning_rate": 4.98825433693681e-07,
      "loss": 0.0001,
      "reward": 1.7250000461935997,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7339285984635353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8092
    },
    {
      "completion_length": 279.2901906967163,
      "epoch": 1.3570979504589462,
      "grad_norm": 0.33908673465572003,
      "kl": 0.1176605224609375,
      "learning_rate": 4.988242719567636e-07,
      "loss": 0.0001,
      "reward": 1.6571429520845413,
      "reward_std": 0.10606601648032665,
      "rewards/equation_reward_func": 0.6839285977184772,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 8094
    },
    {
      "completion_length": 262.75894260406494,
      "epoch": 1.3574332536988138,
      "grad_norm": 0.245226238939219,
      "kl": 0.1348724365234375,
      "learning_rate": 4.988231096469606e-07,
      "loss": 0.0001,
      "reward": 1.7375000640749931,
      "reward_std": 0.07828682195395231,
      "rewards/equation_reward_func": 0.7508928775787354,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8096
    },
    {
      "completion_length": 273.8214416503906,
      "epoch": 1.3577685569386815,
      "grad_norm": 0.21272801489659302,
      "kl": 0.1975250244140625,
      "learning_rate": 4.988219467642743e-07,
      "loss": 0.0002,
      "reward": 1.7857143357396126,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 8098
    },
    {
      "completion_length": 266.92412090301514,
      "epoch": 1.358103860178549,
      "grad_norm": 0.1923993590207927,
      "kl": 0.10992431640625,
      "learning_rate": 4.988207833087078e-07,
      "loss": 0.0001,
      "reward": 1.7178572341799736,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7178571783006191,
      "rewards/format_reward_func": 1.0,
      "step": 8100
    },
    {
      "completion_length": 273.35715198516846,
      "epoch": 1.3584391634184165,
      "grad_norm": 0.2306174121200835,
      "kl": 0.5870361328125,
      "learning_rate": 4.988196192802636e-07,
      "loss": 0.0006,
      "reward": 1.7642857730388641,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7732143066823483,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8102
    },
    {
      "completion_length": 253.38393878936768,
      "epoch": 1.3587744666582842,
      "grad_norm": 0.2780585896084185,
      "kl": 0.104766845703125,
      "learning_rate": 4.988184546789444e-07,
      "loss": 0.0001,
      "reward": 1.7665179446339607,
      "reward_std": 0.037249374436214566,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 8104
    },
    {
      "completion_length": 264.9151887893677,
      "epoch": 1.3591097698981516,
      "grad_norm": 0.21535603932789843,
      "kl": 0.2724151611328125,
      "learning_rate": 4.988172895047528e-07,
      "loss": 0.0003,
      "reward": 1.7214286103844643,
      "reward_std": 0.0909137288108468,
      "rewards/equation_reward_func": 0.7392857559025288,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 8106
    },
    {
      "completion_length": 257.81697940826416,
      "epoch": 1.3594450731380192,
      "grad_norm": 0.1236021672041698,
      "kl": 0.0832977294921875,
      "learning_rate": 4.988161237576915e-07,
      "loss": 0.0001,
      "reward": 1.7625000476837158,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8108
    },
    {
      "completion_length": 266.21429920196533,
      "epoch": 1.3597803763778868,
      "grad_norm": 0.28031837613588595,
      "kl": 0.12738037109375,
      "learning_rate": 4.988149574377633e-07,
      "loss": 0.0001,
      "reward": 1.6785715073347092,
      "reward_std": 0.09091372694820166,
      "rewards/equation_reward_func": 0.6875000447034836,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8110
    },
    {
      "completion_length": 260.03572940826416,
      "epoch": 1.3601156796177543,
      "grad_norm": 0.24458932541296974,
      "kl": 0.3977203369140625,
      "learning_rate": 4.988137905449708e-07,
      "loss": 0.0004,
      "reward": 1.7678571939468384,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7767857424914837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8112
    },
    {
      "completion_length": 261.80804920196533,
      "epoch": 1.360450982857622,
      "grad_norm": 0.249895327320769,
      "kl": 0.093292236328125,
      "learning_rate": 4.988126230793167e-07,
      "loss": 0.0001,
      "reward": 1.7339286282658577,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7383928969502449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8114
    },
    {
      "completion_length": 255.57590293884277,
      "epoch": 1.3607862860974893,
      "grad_norm": 0.3183229100046573,
      "kl": 0.235931396484375,
      "learning_rate": 4.988114550408037e-07,
      "loss": 0.0002,
      "reward": 1.7607143372297287,
      "reward_std": 0.05555839091539383,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 8116
    },
    {
      "completion_length": 257.883939743042,
      "epoch": 1.361121589337357,
      "grad_norm": 0.215454386421447,
      "kl": 0.0789642333984375,
      "learning_rate": 4.988102864294344e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7910714522004128,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8118
    },
    {
      "completion_length": 252.44644165039062,
      "epoch": 1.3614568925772246,
      "grad_norm": 0.12456854677300534,
      "kl": 0.2279205322265625,
      "learning_rate": 4.988091172452117e-07,
      "loss": 0.0002,
      "reward": 1.730357214808464,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7348214536905289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8120
    },
    {
      "completion_length": 245.48215579986572,
      "epoch": 1.3617921958170922,
      "grad_norm": 0.32630944614624346,
      "kl": 0.2685546875,
      "learning_rate": 4.988079474881381e-07,
      "loss": 0.0003,
      "reward": 1.7839286476373672,
      "reward_std": 0.06313453521579504,
      "rewards/equation_reward_func": 0.7973214536905289,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8122
    },
    {
      "completion_length": 252.8169755935669,
      "epoch": 1.3621274990569596,
      "grad_norm": 0.24195474875844628,
      "kl": 0.3311767578125,
      "learning_rate": 4.988067771582163e-07,
      "loss": 0.0003,
      "reward": 1.7696429267525673,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7830357439815998,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8124
    },
    {
      "completion_length": 259.477689743042,
      "epoch": 1.3624628022968273,
      "grad_norm": 0.7045995087371681,
      "kl": 1.045379638671875,
      "learning_rate": 4.98805606255449e-07,
      "loss": 0.001,
      "reward": 1.7071429267525673,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7250000275671482,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 8126
    },
    {
      "completion_length": 275.4866199493408,
      "epoch": 1.3627981055366947,
      "grad_norm": 0.31304647795741647,
      "kl": 1.1280517578125,
      "learning_rate": 4.988044347798392e-07,
      "loss": 0.0011,
      "reward": 1.708928644657135,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7133928816765547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8128
    },
    {
      "completion_length": 248.54912090301514,
      "epoch": 1.3631334087765623,
      "grad_norm": 0.23159962235353818,
      "kl": 0.232452392578125,
      "learning_rate": 4.988032627313892e-07,
      "loss": 0.0002,
      "reward": 1.769642911851406,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8130
    },
    {
      "completion_length": 267.6607255935669,
      "epoch": 1.36346871201643,
      "grad_norm": 0.20785342484097794,
      "kl": 0.69158935546875,
      "learning_rate": 4.988020901101017e-07,
      "loss": 0.0007,
      "reward": 1.719642922282219,
      "reward_std": 0.09343910869210958,
      "rewards/equation_reward_func": 0.7330357357859612,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8132
    },
    {
      "completion_length": 255.8348331451416,
      "epoch": 1.3638040152562974,
      "grad_norm": 0.5485224272178171,
      "kl": 0.561676025390625,
      "learning_rate": 4.988009169159798e-07,
      "loss": 0.0006,
      "reward": 1.7339286357164383,
      "reward_std": 0.08333758544176817,
      "rewards/equation_reward_func": 0.7473214641213417,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8134
    },
    {
      "completion_length": 251.4196548461914,
      "epoch": 1.364139318496165,
      "grad_norm": 0.37869015687501495,
      "kl": 0.3833465576171875,
      "learning_rate": 4.987997431490257e-07,
      "loss": 0.0004,
      "reward": 1.7428571805357933,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7517857514321804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8136
    },
    {
      "completion_length": 246.7589349746704,
      "epoch": 1.3644746217360324,
      "grad_norm": 0.29526207045490876,
      "kl": 0.158782958984375,
      "learning_rate": 4.987985688092426e-07,
      "loss": 0.0002,
      "reward": 1.7982143387198448,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.820535734295845,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 8138
    },
    {
      "completion_length": 261.4687614440918,
      "epoch": 1.3648099249759,
      "grad_norm": 0.33114232485365375,
      "kl": 0.33135986328125,
      "learning_rate": 4.987973938966326e-07,
      "loss": 0.0003,
      "reward": 1.798214353621006,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.802678607404232,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8140
    },
    {
      "completion_length": 251.94197463989258,
      "epoch": 1.3651452282157677,
      "grad_norm": 0.4965083376527316,
      "kl": 0.1101531982421875,
      "learning_rate": 4.98796218411199e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.0858629634603858,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 1.0,
      "step": 8142
    },
    {
      "completion_length": 242.477689743042,
      "epoch": 1.3654805314556353,
      "grad_norm": 0.32604095360515667,
      "kl": 0.1768951416015625,
      "learning_rate": 4.987950423529442e-07,
      "loss": 0.0002,
      "reward": 1.8142857775092125,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8142857365310192,
      "rewards/format_reward_func": 1.0,
      "step": 8144
    },
    {
      "completion_length": 250.02679538726807,
      "epoch": 1.3658158346955027,
      "grad_norm": 0.28201790087899403,
      "kl": 0.1674957275390625,
      "learning_rate": 4.98793865721871e-07,
      "loss": 0.0002,
      "reward": 1.7303572222590446,
      "reward_std": 0.07828682195395231,
      "rewards/equation_reward_func": 0.7437500394880772,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8146
    },
    {
      "completion_length": 237.24554538726807,
      "epoch": 1.3661511379353704,
      "grad_norm": 0.5227818679067141,
      "kl": 0.1568145751953125,
      "learning_rate": 4.987926885179821e-07,
      "loss": 0.0002,
      "reward": 1.8000000566244125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000156462193,
      "rewards/format_reward_func": 1.0,
      "step": 8148
    },
    {
      "completion_length": 239.01786708831787,
      "epoch": 1.3664864411752378,
      "grad_norm": 0.20125727672007632,
      "kl": 0.110504150390625,
      "learning_rate": 4.987915107412801e-07,
      "loss": 0.0001,
      "reward": 1.769642911851406,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8150
    },
    {
      "completion_length": 231.80358409881592,
      "epoch": 1.3668217444151054,
      "grad_norm": 0.26540877315998423,
      "kl": 0.13861083984375,
      "learning_rate": 4.987903323917678e-07,
      "loss": 0.0001,
      "reward": 1.803571492433548,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.8035714477300644,
      "rewards/format_reward_func": 1.0,
      "step": 8152
    },
    {
      "completion_length": 234.2232265472412,
      "epoch": 1.367157047654973,
      "grad_norm": 2.742825030886559,
      "kl": 0.16217041015625,
      "learning_rate": 4.987891534694479e-07,
      "loss": 0.0002,
      "reward": 1.7303571999073029,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7348214704543352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8154
    },
    {
      "completion_length": 237.60268783569336,
      "epoch": 1.3674923508948404,
      "grad_norm": 0.8197647544954363,
      "kl": 0.338653564453125,
      "learning_rate": 4.987879739743232e-07,
      "loss": 0.0003,
      "reward": 1.7214286401867867,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7214285954833031,
      "rewards/format_reward_func": 1.0,
      "step": 8156
    },
    {
      "completion_length": 238.80358123779297,
      "epoch": 1.367827654134708,
      "grad_norm": 0.17306658774659567,
      "kl": 0.164398193359375,
      "learning_rate": 4.987867939063963e-07,
      "loss": 0.0002,
      "reward": 1.821428619325161,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8214285932481289,
      "rewards/format_reward_func": 1.0,
      "step": 8158
    },
    {
      "completion_length": 240.7366189956665,
      "epoch": 1.3681629573745755,
      "grad_norm": 0.2755680868704616,
      "kl": 0.3265380859375,
      "learning_rate": 4.987856132656701e-07,
      "loss": 0.0003,
      "reward": 1.7767857909202576,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7812500335276127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8160
    },
    {
      "completion_length": 242.43750953674316,
      "epoch": 1.3684982606144431,
      "grad_norm": 0.34220703415178927,
      "kl": 0.2494964599609375,
      "learning_rate": 4.987844320521469e-07,
      "loss": 0.0002,
      "reward": 1.7392857745289803,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7392857372760773,
      "rewards/format_reward_func": 1.0,
      "step": 8162
    },
    {
      "completion_length": 244.96875858306885,
      "epoch": 1.3688335638543108,
      "grad_norm": 0.3656314574025403,
      "kl": 0.13720703125,
      "learning_rate": 4.987832502658299e-07,
      "loss": 0.0001,
      "reward": 1.8178571835160255,
      "reward_std": 0.04545686487108469,
      "rewards/equation_reward_func": 0.8267857357859612,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8164
    },
    {
      "completion_length": 248.82143878936768,
      "epoch": 1.3691688670941784,
      "grad_norm": 0.5701026727666527,
      "kl": 0.2044219970703125,
      "learning_rate": 4.987820679067215e-07,
      "loss": 0.0002,
      "reward": 1.7839286401867867,
      "reward_std": 0.09343910869210958,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8166
    },
    {
      "completion_length": 248.11608219146729,
      "epoch": 1.3695041703340458,
      "grad_norm": 0.1886594138713511,
      "kl": 0.4144134521484375,
      "learning_rate": 4.987808849748246e-07,
      "loss": 0.0004,
      "reward": 1.7839286476373672,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7883928678929806,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8168
    },
    {
      "completion_length": 244.99108219146729,
      "epoch": 1.3698394735739134,
      "grad_norm": 0.21902865172820737,
      "kl": 0.2199249267578125,
      "learning_rate": 4.987797014701418e-07,
      "loss": 0.0002,
      "reward": 1.753571480512619,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7625000327825546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8170
    },
    {
      "completion_length": 245.6517972946167,
      "epoch": 1.3701747768137809,
      "grad_norm": 0.2876812085996151,
      "kl": 0.0823822021484375,
      "learning_rate": 4.98778517392676e-07,
      "loss": 0.0001,
      "reward": 1.7625000700354576,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8172
    },
    {
      "completion_length": 251.2901906967163,
      "epoch": 1.3705100800536485,
      "grad_norm": 0.23891560103695314,
      "kl": 0.3800506591796875,
      "learning_rate": 4.987773327424297e-07,
      "loss": 0.0004,
      "reward": 1.7535714879631996,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 8174
    },
    {
      "completion_length": 239.9553680419922,
      "epoch": 1.3708453832935161,
      "grad_norm": 0.16681507824174535,
      "kl": 0.07891845703125,
      "learning_rate": 4.987761475194058e-07,
      "loss": 0.0001,
      "reward": 1.7464286237955093,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.755357176065445,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8176
    },
    {
      "completion_length": 246.50001049041748,
      "epoch": 1.3711806865333838,
      "grad_norm": 0.1460767895717453,
      "kl": 0.605316162109375,
      "learning_rate": 4.98774961723607e-07,
      "loss": 0.0006,
      "reward": 1.728571504354477,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7375000230967999,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8178
    },
    {
      "completion_length": 240.90626049041748,
      "epoch": 1.3715159897732512,
      "grad_norm": 0.27316633410956814,
      "kl": 0.10211181640625,
      "learning_rate": 4.987737753550359e-07,
      "loss": 0.0001,
      "reward": 1.7125000432133675,
      "reward_std": 0.061871841782703996,
      "rewards/equation_reward_func": 0.7258928939700127,
      "rewards/format_reward_func": 0.9866071529686451,
      "step": 8180
    },
    {
      "completion_length": 247.2544765472412,
      "epoch": 1.3718512930131188,
      "grad_norm": 0.27321212078962503,
      "kl": 0.32537841796875,
      "learning_rate": 4.987725884136954e-07,
      "loss": 0.0003,
      "reward": 1.787500075995922,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7919643223285675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8182
    },
    {
      "completion_length": 235.66072273254395,
      "epoch": 1.3721865962529862,
      "grad_norm": 0.26702454806950976,
      "kl": 0.393646240234375,
      "learning_rate": 4.987714008995882e-07,
      "loss": 0.0004,
      "reward": 1.778571493923664,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 8184
    },
    {
      "completion_length": 236.3928689956665,
      "epoch": 1.3725218994928539,
      "grad_norm": 0.2859289731323632,
      "kl": 0.1675567626953125,
      "learning_rate": 4.987702128127169e-07,
      "loss": 0.0002,
      "reward": 1.7428572326898575,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 8186
    },
    {
      "completion_length": 235.7053680419922,
      "epoch": 1.3728572027327215,
      "grad_norm": 0.1957172849424329,
      "kl": 0.1398468017578125,
      "learning_rate": 4.987690241530844e-07,
      "loss": 0.0001,
      "reward": 1.7321429252624512,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321428842842579,
      "rewards/format_reward_func": 1.0,
      "step": 8188
    },
    {
      "completion_length": 238.45983028411865,
      "epoch": 1.373192505972589,
      "grad_norm": 0.15376641020518758,
      "kl": 0.145050048828125,
      "learning_rate": 4.987678349206933e-07,
      "loss": 0.0001,
      "reward": 1.751785770058632,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7562500387430191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8190
    },
    {
      "completion_length": 248.5044765472412,
      "epoch": 1.3735278092124565,
      "grad_norm": 0.27968755909602633,
      "kl": 0.077972412109375,
      "learning_rate": 4.987666451155465e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.7946428842842579,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8192
    },
    {
      "completion_length": 242.2500123977661,
      "epoch": 1.373863112452324,
      "grad_norm": 0.2436045067533531,
      "kl": 0.108734130859375,
      "learning_rate": 4.987654547376466e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 8194
    },
    {
      "completion_length": 240.0937614440918,
      "epoch": 1.3741984156921916,
      "grad_norm": 0.3174198813775443,
      "kl": 0.0972442626953125,
      "learning_rate": 4.987642637869965e-07,
      "loss": 0.0001,
      "reward": 1.7625000774860382,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643089175224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8196
    },
    {
      "completion_length": 241.19643878936768,
      "epoch": 1.3745337189320592,
      "grad_norm": 0.32104686249339653,
      "kl": 0.245880126953125,
      "learning_rate": 4.987630722635986e-07,
      "loss": 0.0002,
      "reward": 1.7839286550879478,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8198
    },
    {
      "completion_length": 238.9062614440918,
      "epoch": 1.3748690221719269,
      "grad_norm": 0.22119044107872407,
      "kl": 0.101409912109375,
      "learning_rate": 4.98761880167456e-07,
      "loss": 0.0001,
      "reward": 1.721428632736206,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7214286010712385,
      "rewards/format_reward_func": 1.0,
      "step": 8200
    },
    {
      "completion_length": 242.1294765472412,
      "epoch": 1.3752043254117943,
      "grad_norm": 0.19621369458051252,
      "kl": 0.0926971435546875,
      "learning_rate": 4.987606874985713e-07,
      "loss": 0.0001,
      "reward": 1.7750000432133675,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 8202
    },
    {
      "completion_length": 252.14287185668945,
      "epoch": 1.375539628651662,
      "grad_norm": 0.20777406668459847,
      "kl": 0.213043212890625,
      "learning_rate": 4.987594942569473e-07,
      "loss": 0.0002,
      "reward": 1.800000049173832,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 8204
    },
    {
      "completion_length": 246.38840198516846,
      "epoch": 1.3758749318915293,
      "grad_norm": 0.25030524166954016,
      "kl": 0.1303863525390625,
      "learning_rate": 4.987583004425867e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 8206
    },
    {
      "completion_length": 256.933048248291,
      "epoch": 1.376210235131397,
      "grad_norm": 0.21811990945534226,
      "kl": 0.104339599609375,
      "learning_rate": 4.987571060554922e-07,
      "loss": 0.0001,
      "reward": 1.7553572282195091,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7598214596509933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8208
    },
    {
      "completion_length": 247.2410831451416,
      "epoch": 1.3765455383712646,
      "grad_norm": 0.30362258829253413,
      "kl": 0.1866455078125,
      "learning_rate": 4.987559110956667e-07,
      "loss": 0.0002,
      "reward": 1.7946429029107094,
      "reward_std": 0.07828682195395231,
      "rewards/equation_reward_func": 0.808035746216774,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8210
    },
    {
      "completion_length": 244.8705472946167,
      "epoch": 1.376880841611132,
      "grad_norm": 0.14472449444708704,
      "kl": 0.1186981201171875,
      "learning_rate": 4.987547155631128e-07,
      "loss": 0.0001,
      "reward": 1.7589286416769028,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7633928954601288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8212
    },
    {
      "completion_length": 250.7812623977661,
      "epoch": 1.3772161448509996,
      "grad_norm": 0.21972446651081873,
      "kl": 0.090972900390625,
      "learning_rate": 4.987535194578333e-07,
      "loss": 0.0001,
      "reward": 1.7732143923640251,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7776786014437675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8214
    },
    {
      "completion_length": 243.51786708831787,
      "epoch": 1.377551448090867,
      "grad_norm": 0.23292783449066065,
      "kl": 0.0885467529296875,
      "learning_rate": 4.98752322779831e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 8216
    },
    {
      "completion_length": 252.18751049041748,
      "epoch": 1.3778867513307347,
      "grad_norm": 0.20195243651174724,
      "kl": 0.0833892822265625,
      "learning_rate": 4.987511255291087e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7946428954601288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8218
    },
    {
      "completion_length": 240.0446538925171,
      "epoch": 1.3782220545706023,
      "grad_norm": 0.17906265132759394,
      "kl": 0.0956573486328125,
      "learning_rate": 4.987499277056689e-07,
      "loss": 0.0001,
      "reward": 1.798214316368103,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.8026785962283611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8220
    },
    {
      "completion_length": 254.49108505249023,
      "epoch": 1.37855735781047,
      "grad_norm": 0.3025207003237902,
      "kl": 0.0892486572265625,
      "learning_rate": 4.987487293095148e-07,
      "loss": 0.0001,
      "reward": 1.7732143700122833,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7776785995811224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8222
    },
    {
      "completion_length": 254.6741180419922,
      "epoch": 1.3788926610503374,
      "grad_norm": 0.17416299428135643,
      "kl": 0.0872802734375,
      "learning_rate": 4.987475303406486e-07,
      "loss": 0.0001,
      "reward": 1.7053572237491608,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7098214644938707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8224
    },
    {
      "completion_length": 243.1026906967163,
      "epoch": 1.379227964290205,
      "grad_norm": 0.007376292518863739,
      "kl": 0.0882415771484375,
      "learning_rate": 4.987463307990735e-07,
      "loss": 0.0001,
      "reward": 1.7875000908970833,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7919643186032772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8226
    },
    {
      "completion_length": 243.65179634094238,
      "epoch": 1.3795632675300724,
      "grad_norm": 0.3491150701987759,
      "kl": 0.29400634765625,
      "learning_rate": 4.987451306847922e-07,
      "loss": 0.0003,
      "reward": 1.666071504354477,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.6705357562750578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8228
    },
    {
      "completion_length": 259.85268783569336,
      "epoch": 1.37989857076994,
      "grad_norm": 0.2254974501549908,
      "kl": 0.18572998046875,
      "learning_rate": 4.987439299978072e-07,
      "loss": 0.0002,
      "reward": 1.7500000596046448,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 8230
    },
    {
      "completion_length": 248.71876335144043,
      "epoch": 1.3802338740098077,
      "grad_norm": 0.005612214152327968,
      "kl": 0.0982513427734375,
      "learning_rate": 4.987427287381215e-07,
      "loss": 0.0001,
      "reward": 1.7446429058909416,
      "reward_std": 0.007576144300401211,
      "rewards/equation_reward_func": 0.74910718947649,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8232
    },
    {
      "completion_length": 242.35268783569336,
      "epoch": 1.380569177249675,
      "grad_norm": 0.15117705551384866,
      "kl": 0.1826324462890625,
      "learning_rate": 4.987415269057379e-07,
      "loss": 0.0002,
      "reward": 1.8178571984171867,
      "reward_std": 0.04545686487108469,
      "rewards/equation_reward_func": 0.8267857395112514,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8234
    },
    {
      "completion_length": 248.83036994934082,
      "epoch": 1.3809044804895427,
      "grad_norm": 0.16819016833733927,
      "kl": 0.1221923828125,
      "learning_rate": 4.987403245006591e-07,
      "loss": 0.0001,
      "reward": 1.7553571984171867,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214633762836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8236
    },
    {
      "completion_length": 257.6607303619385,
      "epoch": 1.3812397837294104,
      "grad_norm": 0.20559730867779458,
      "kl": 0.1548309326171875,
      "learning_rate": 4.987391215228878e-07,
      "loss": 0.0002,
      "reward": 1.7482143566012383,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8238
    },
    {
      "completion_length": 244.97769260406494,
      "epoch": 1.3815750869692778,
      "grad_norm": 0.14968388550478384,
      "kl": 0.2997283935546875,
      "learning_rate": 4.987379179724267e-07,
      "loss": 0.0003,
      "reward": 1.778571493923664,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 8240
    },
    {
      "completion_length": 253.08037090301514,
      "epoch": 1.3819103902091454,
      "grad_norm": 0.16791201111642334,
      "kl": 0.1606903076171875,
      "learning_rate": 4.987367138492787e-07,
      "loss": 0.0002,
      "reward": 1.7660715207457542,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7705357447266579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8242
    },
    {
      "completion_length": 250.44644260406494,
      "epoch": 1.382245693449013,
      "grad_norm": 0.2818567106401476,
      "kl": 0.3727569580078125,
      "learning_rate": 4.987355091534467e-07,
      "loss": 0.0004,
      "reward": 1.7714286521077156,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7714285776019096,
      "rewards/format_reward_func": 1.0,
      "step": 8244
    },
    {
      "completion_length": 241.80358219146729,
      "epoch": 1.3825809966888805,
      "grad_norm": 0.21332612942021542,
      "kl": 0.193572998046875,
      "learning_rate": 4.987343038849333e-07,
      "loss": 0.0002,
      "reward": 1.7571429461240768,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 8246
    },
    {
      "completion_length": 244.90179443359375,
      "epoch": 1.382916299928748,
      "grad_norm": 0.29629913715379597,
      "kl": 0.24725341796875,
      "learning_rate": 4.987330980437413e-07,
      "loss": 0.0002,
      "reward": 1.7750000432133675,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 8248
    },
    {
      "completion_length": 258.2053689956665,
      "epoch": 1.3832516031686155,
      "grad_norm": 0.3395190955566177,
      "kl": 0.3449859619140625,
      "learning_rate": 4.987318916298734e-07,
      "loss": 0.0003,
      "reward": 1.7517857775092125,
      "reward_std": 0.09848987217992544,
      "rewards/equation_reward_func": 0.765178594738245,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8250
    },
    {
      "completion_length": 249.21429634094238,
      "epoch": 1.3835869064084831,
      "grad_norm": 0.44822066564718804,
      "kl": 0.234832763671875,
      "learning_rate": 4.987306846433325e-07,
      "loss": 0.0002,
      "reward": 1.7446429282426834,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7491071820259094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8252
    },
    {
      "completion_length": 242.1071548461914,
      "epoch": 1.3839222096483508,
      "grad_norm": 0.3920342374420616,
      "kl": 0.464263916015625,
      "learning_rate": 4.987294770841214e-07,
      "loss": 0.0005,
      "reward": 1.733928643167019,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.7473214566707611,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8254
    },
    {
      "completion_length": 240.27233028411865,
      "epoch": 1.3842575128882184,
      "grad_norm": 0.1406635263351389,
      "kl": 0.174072265625,
      "learning_rate": 4.987282689522428e-07,
      "loss": 0.0002,
      "reward": 1.7464286386966705,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464285790920258,
      "rewards/format_reward_func": 1.0,
      "step": 8256
    },
    {
      "completion_length": 239.23215198516846,
      "epoch": 1.3845928161280858,
      "grad_norm": 0.179425887440818,
      "kl": 0.1930084228515625,
      "learning_rate": 4.987270602476995e-07,
      "loss": 0.0002,
      "reward": 1.7750000804662704,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 8258
    },
    {
      "completion_length": 242.12947273254395,
      "epoch": 1.3849281193679535,
      "grad_norm": 0.3154012365668129,
      "kl": 0.10003662109375,
      "learning_rate": 4.987258509704942e-07,
      "loss": 0.0001,
      "reward": 1.7392857819795609,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857559025288,
      "rewards/format_reward_func": 1.0,
      "step": 8260
    },
    {
      "completion_length": 234.19643783569336,
      "epoch": 1.3852634226078209,
      "grad_norm": 0.16562049076361046,
      "kl": 0.114166259765625,
      "learning_rate": 4.9872464112063e-07,
      "loss": 0.0001,
      "reward": 1.821428619325161,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8214285969734192,
      "rewards/format_reward_func": 1.0,
      "step": 8262
    },
    {
      "completion_length": 234.68750953674316,
      "epoch": 1.3855987258476885,
      "grad_norm": 0.7369731607621153,
      "kl": 0.1848907470703125,
      "learning_rate": 4.987234306981093e-07,
      "loss": 0.0002,
      "reward": 1.7428572326898575,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 8264
    },
    {
      "completion_length": 233.227689743042,
      "epoch": 1.3859340290875561,
      "grad_norm": 0.02821810625385667,
      "kl": 0.226776123046875,
      "learning_rate": 4.98722219702935e-07,
      "loss": 0.0002,
      "reward": 1.7732143253087997,
      "reward_std": 0.017677669413387775,
      "rewards/equation_reward_func": 0.7776786126196384,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8266
    },
    {
      "completion_length": 228.66072463989258,
      "epoch": 1.3862693323274236,
      "grad_norm": 0.19036897284281554,
      "kl": 0.1602020263671875,
      "learning_rate": 4.9872100813511e-07,
      "loss": 0.0002,
      "reward": 1.782142922282219,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 8268
    },
    {
      "completion_length": 240.06251049041748,
      "epoch": 1.3866046355672912,
      "grad_norm": 0.28445586166487996,
      "kl": 0.0912628173828125,
      "learning_rate": 4.98719795994637e-07,
      "loss": 0.0001,
      "reward": 1.7321429550647736,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.732142873108387,
      "rewards/format_reward_func": 1.0,
      "step": 8270
    },
    {
      "completion_length": 228.25001049041748,
      "epoch": 1.3869399388071586,
      "grad_norm": 0.36868253449781974,
      "kl": 0.4998779296875,
      "learning_rate": 4.987185832815188e-07,
      "loss": 0.0005,
      "reward": 1.7928572073578835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571626543999,
      "rewards/format_reward_func": 1.0,
      "step": 8272
    },
    {
      "completion_length": 231.93304634094238,
      "epoch": 1.3872752420470262,
      "grad_norm": 0.2135789845077452,
      "kl": 0.1878509521484375,
      "learning_rate": 4.987173699957582e-07,
      "loss": 0.0002,
      "reward": 1.7803571820259094,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7848214656114578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8274
    },
    {
      "completion_length": 233.14733219146729,
      "epoch": 1.3876105452868939,
      "grad_norm": 0.09741457644301535,
      "kl": 0.0953216552734375,
      "learning_rate": 4.98716156137358e-07,
      "loss": 0.0001,
      "reward": 1.775000050663948,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 8276
    },
    {
      "completion_length": 231.3259038925171,
      "epoch": 1.3879458485267615,
      "grad_norm": 0.3273387632249645,
      "kl": 0.5626983642578125,
      "learning_rate": 4.98714941706321e-07,
      "loss": 0.0006,
      "reward": 1.7857143506407738,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 8278
    },
    {
      "completion_length": 228.06697463989258,
      "epoch": 1.388281151766629,
      "grad_norm": 0.4202954088324719,
      "kl": 0.4383392333984375,
      "learning_rate": 4.9871372670265e-07,
      "loss": 0.0004,
      "reward": 1.7750000432133675,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.775000037625432,
      "rewards/format_reward_func": 1.0,
      "step": 8280
    },
    {
      "completion_length": 228.60715293884277,
      "epoch": 1.3886164550064966,
      "grad_norm": 0.28889603554498616,
      "kl": 0.09014892578125,
      "learning_rate": 4.987125111263477e-07,
      "loss": 0.0001,
      "reward": 1.7607143446803093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143241912127,
      "rewards/format_reward_func": 1.0,
      "step": 8282
    },
    {
      "completion_length": 225.5535831451416,
      "epoch": 1.388951758246364,
      "grad_norm": 0.33761219684242044,
      "kl": 0.2540435791015625,
      "learning_rate": 4.987112949774171e-07,
      "loss": 0.0003,
      "reward": 1.7214286550879478,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7214286141097546,
      "rewards/format_reward_func": 1.0,
      "step": 8284
    },
    {
      "completion_length": 226.7634048461914,
      "epoch": 1.3892870614862316,
      "grad_norm": 0.19009764938273654,
      "kl": 0.253997802734375,
      "learning_rate": 4.987100782558608e-07,
      "loss": 0.0003,
      "reward": 1.7678572237491608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 8286
    },
    {
      "completion_length": 235.06251049041748,
      "epoch": 1.3896223647260992,
      "grad_norm": 0.25891486225934435,
      "kl": 0.2195892333984375,
      "learning_rate": 4.987088609616818e-07,
      "loss": 0.0002,
      "reward": 1.7500000596046448,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 8288
    },
    {
      "completion_length": 226.99554538726807,
      "epoch": 1.3899576679659666,
      "grad_norm": 0.1975587721492033,
      "kl": 0.310394287109375,
      "learning_rate": 4.987076430948827e-07,
      "loss": 0.0003,
      "reward": 1.717857226729393,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7178571783006191,
      "rewards/format_reward_func": 1.0,
      "step": 8290
    },
    {
      "completion_length": 238.50447463989258,
      "epoch": 1.3902929712058343,
      "grad_norm": 0.2642575662903022,
      "kl": 0.089019775390625,
      "learning_rate": 4.987064246554664e-07,
      "loss": 0.0001,
      "reward": 1.7696429416537285,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7741071842610836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8292
    },
    {
      "completion_length": 230.88840293884277,
      "epoch": 1.3906282744457017,
      "grad_norm": 0.1691942612744324,
      "kl": 0.2000579833984375,
      "learning_rate": 4.987052056434357e-07,
      "loss": 0.0002,
      "reward": 1.7750000655651093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 8294
    },
    {
      "completion_length": 237.10268783569336,
      "epoch": 1.3909635776855693,
      "grad_norm": 0.330917499427202,
      "kl": 0.1829986572265625,
      "learning_rate": 4.987039860587933e-07,
      "loss": 0.0002,
      "reward": 1.7375000715255737,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7419643048197031,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8296
    },
    {
      "completion_length": 244.22322463989258,
      "epoch": 1.391298880925437,
      "grad_norm": 0.17135511845815518,
      "kl": 0.373077392578125,
      "learning_rate": 4.987027659015423e-07,
      "loss": 0.0004,
      "reward": 1.735714353621006,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7446428909897804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8298
    },
    {
      "completion_length": 252.75001335144043,
      "epoch": 1.3916341841653046,
      "grad_norm": 0.33960622184873085,
      "kl": 0.1415557861328125,
      "learning_rate": 4.987015451716853e-07,
      "loss": 0.0001,
      "reward": 1.6946429386734962,
      "reward_std": 0.09848987031728029,
      "rewards/equation_reward_func": 0.6991071663796902,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8300
    },
    {
      "completion_length": 245.071439743042,
      "epoch": 1.391969487405172,
      "grad_norm": 0.2723528900900101,
      "kl": 0.122833251953125,
      "learning_rate": 4.987003238692251e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000394880772,
      "rewards/format_reward_func": 1.0,
      "step": 8302
    },
    {
      "completion_length": 241.62054824829102,
      "epoch": 1.3923047906450396,
      "grad_norm": 0.2372067895679325,
      "kl": 0.095794677734375,
      "learning_rate": 4.986991019941644e-07,
      "loss": 0.0001,
      "reward": 1.7660714834928513,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7705357484519482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8304
    },
    {
      "completion_length": 236.24108409881592,
      "epoch": 1.392640093884907,
      "grad_norm": 0.27936471533894636,
      "kl": 0.105682373046875,
      "learning_rate": 4.986978795465063e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 8306
    },
    {
      "completion_length": 236.5089406967163,
      "epoch": 1.3929753971247747,
      "grad_norm": 0.23005213349580997,
      "kl": 0.1139984130859375,
      "learning_rate": 4.986966565262534e-07,
      "loss": 0.0001,
      "reward": 1.8178572058677673,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.817857164889574,
      "rewards/format_reward_func": 1.0,
      "step": 8308
    },
    {
      "completion_length": 232.9285831451416,
      "epoch": 1.3933107003646423,
      "grad_norm": 0.3397062470054958,
      "kl": 0.1307373046875,
      "learning_rate": 4.986954329334087e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857268452644,
      "rewards/format_reward_func": 1.0,
      "step": 8310
    },
    {
      "completion_length": 248.77679634094238,
      "epoch": 1.39364600360451,
      "grad_norm": 0.1851405173069177,
      "kl": 0.1040496826171875,
      "learning_rate": 4.986942087679749e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7714286148548126,
      "rewards/format_reward_func": 1.0,
      "step": 8312
    },
    {
      "completion_length": 254.46429824829102,
      "epoch": 1.3939813068443774,
      "grad_norm": 0.28245811292805495,
      "kl": 0.1212158203125,
      "learning_rate": 4.986929840299547e-07,
      "loss": 0.0001,
      "reward": 1.7410714998841286,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357424914837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8314
    },
    {
      "completion_length": 233.6875114440918,
      "epoch": 1.394316610084245,
      "grad_norm": 0.22962909738151074,
      "kl": 0.161712646484375,
      "learning_rate": 4.986917587193511e-07,
      "loss": 0.0002,
      "reward": 1.725000061094761,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000424683094,
      "rewards/format_reward_func": 1.0,
      "step": 8316
    },
    {
      "completion_length": 232.30358409881592,
      "epoch": 1.3946519133241124,
      "grad_norm": 0.3743388791517971,
      "kl": 0.281494140625,
      "learning_rate": 4.986905328361668e-07,
      "loss": 0.0003,
      "reward": 1.7785715013742447,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 8318
    },
    {
      "completion_length": 236.19197463989258,
      "epoch": 1.39498721656398,
      "grad_norm": 0.18783598559833184,
      "kl": 0.1347198486328125,
      "learning_rate": 4.986893063804048e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714469850063,
      "rewards/format_reward_func": 1.0,
      "step": 8320
    },
    {
      "completion_length": 230.33483219146729,
      "epoch": 1.3953225198038477,
      "grad_norm": 0.21152856279262738,
      "kl": 0.2019805908203125,
      "learning_rate": 4.986880793520677e-07,
      "loss": 0.0002,
      "reward": 1.7571429461240768,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 1.0,
      "step": 8322
    },
    {
      "completion_length": 230.64733219146729,
      "epoch": 1.395657823043715,
      "grad_norm": 0.1390615445266548,
      "kl": 0.3101959228515625,
      "learning_rate": 4.986868517511585e-07,
      "loss": 0.0003,
      "reward": 1.8000000566244125,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 8324
    },
    {
      "completion_length": 248.91518878936768,
      "epoch": 1.3959931262835827,
      "grad_norm": 0.24380398211878274,
      "kl": 0.0861358642578125,
      "learning_rate": 4.9868562357768e-07,
      "loss": 0.0001,
      "reward": 1.7964286357164383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286059141159,
      "rewards/format_reward_func": 1.0,
      "step": 8326
    },
    {
      "completion_length": 237.41965293884277,
      "epoch": 1.3963284295234502,
      "grad_norm": 0.2780829880264992,
      "kl": 0.08013916015625,
      "learning_rate": 4.98684394831635e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 8328
    },
    {
      "completion_length": 233.1473331451416,
      "epoch": 1.3966637327633178,
      "grad_norm": 0.25010869340527275,
      "kl": 0.0888214111328125,
      "learning_rate": 4.986831655130262e-07,
      "loss": 0.0001,
      "reward": 1.766071505844593,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7705357372760773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8330
    },
    {
      "completion_length": 228.9866189956665,
      "epoch": 1.3969990360031854,
      "grad_norm": 0.2859377036608154,
      "kl": 0.0903472900390625,
      "learning_rate": 4.986819356218565e-07,
      "loss": 0.0001,
      "reward": 1.8000000715255737,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 8332
    },
    {
      "completion_length": 236.1384048461914,
      "epoch": 1.397334339243053,
      "grad_norm": 0.2847175423965275,
      "kl": 0.1934661865234375,
      "learning_rate": 4.98680705158129e-07,
      "loss": 0.0002,
      "reward": 1.7535714879631996,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7535714656114578,
      "rewards/format_reward_func": 1.0,
      "step": 8334
    },
    {
      "completion_length": 239.97769260406494,
      "epoch": 1.3976696424829205,
      "grad_norm": 0.24022825278721455,
      "kl": 0.14080810546875,
      "learning_rate": 4.986794741218462e-07,
      "loss": 0.0001,
      "reward": 1.7196429520845413,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7241071723401546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8336
    },
    {
      "completion_length": 244.83037090301514,
      "epoch": 1.398004945722788,
      "grad_norm": 0.20960399495324336,
      "kl": 0.2378387451171875,
      "learning_rate": 4.986782425130111e-07,
      "loss": 0.0002,
      "reward": 1.7750000581145287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 8338
    },
    {
      "completion_length": 238.55804443359375,
      "epoch": 1.3983402489626555,
      "grad_norm": 0.1458231825218131,
      "kl": 0.10693359375,
      "learning_rate": 4.986770103316263e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 8340
    },
    {
      "completion_length": 245.87947750091553,
      "epoch": 1.3986755522025232,
      "grad_norm": 0.4447577806999814,
      "kl": 0.1026153564453125,
      "learning_rate": 4.98675777577695e-07,
      "loss": 0.0001,
      "reward": 1.6696429327130318,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.6830357573926449,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8342
    },
    {
      "completion_length": 246.63840198516846,
      "epoch": 1.3990108554423908,
      "grad_norm": 0.5346485602738086,
      "kl": 0.145660400390625,
      "learning_rate": 4.986745442512198e-07,
      "loss": 0.0001,
      "reward": 1.7982143387198448,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8026786036789417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8344
    },
    {
      "completion_length": 245.3125123977661,
      "epoch": 1.3993461586822582,
      "grad_norm": 0.2319599019557805,
      "kl": 0.086669921875,
      "learning_rate": 4.986733103522037e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857450991869,
      "rewards/format_reward_func": 1.0,
      "step": 8346
    },
    {
      "completion_length": 258.2634038925171,
      "epoch": 1.3996814619221258,
      "grad_norm": 0.42665677837153787,
      "kl": 0.156463623046875,
      "learning_rate": 4.986720758806493e-07,
      "loss": 0.0002,
      "reward": 1.7464286386966705,
      "reward_std": 0.08586296625435352,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 8348
    },
    {
      "completion_length": 239.5178680419922,
      "epoch": 1.4000167651619932,
      "grad_norm": 0.23925510003351777,
      "kl": 0.1253509521484375,
      "learning_rate": 4.986708408365596e-07,
      "loss": 0.0001,
      "reward": 1.7928572297096252,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571626543999,
      "rewards/format_reward_func": 1.0,
      "step": 8350
    },
    {
      "completion_length": 256.7232255935669,
      "epoch": 1.4003520684018609,
      "grad_norm": 0.29109935293740163,
      "kl": 0.097412109375,
      "learning_rate": 4.986696052199373e-07,
      "loss": 0.0001,
      "reward": 1.7464286163449287,
      "reward_std": 0.0858629634603858,
      "rewards/equation_reward_func": 0.7553571909666061,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8352
    },
    {
      "completion_length": 257.20537090301514,
      "epoch": 1.4006873716417285,
      "grad_norm": 0.17636591360078244,
      "kl": 0.255035400390625,
      "learning_rate": 4.986683690307856e-07,
      "loss": 0.0003,
      "reward": 1.6839286461472511,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.6883928999304771,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8354
    },
    {
      "completion_length": 260.008939743042,
      "epoch": 1.4010226748815962,
      "grad_norm": 0.20055391467891595,
      "kl": 0.430389404296875,
      "learning_rate": 4.986671322691071e-07,
      "loss": 0.0004,
      "reward": 1.6892857998609543,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.6982143372297287,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8356
    },
    {
      "completion_length": 259.1562604904175,
      "epoch": 1.4013579781214636,
      "grad_norm": 0.5636506182839675,
      "kl": 0.3524169921875,
      "learning_rate": 4.986658949349046e-07,
      "loss": 0.0004,
      "reward": 1.7232143580913544,
      "reward_std": 0.08838834706693888,
      "rewards/equation_reward_func": 0.7366071753203869,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8358
    },
    {
      "completion_length": 255.43304634094238,
      "epoch": 1.4016932813613312,
      "grad_norm": 0.5249611321278694,
      "kl": 0.516448974609375,
      "learning_rate": 4.98664657028181e-07,
      "loss": 0.0005,
      "reward": 1.7500000596046448,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7589286118745804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8360
    },
    {
      "completion_length": 260.5312623977661,
      "epoch": 1.4020285846011986,
      "grad_norm": 0.36025818419775324,
      "kl": 0.0961761474609375,
      "learning_rate": 4.986634185489391e-07,
      "loss": 0.0001,
      "reward": 1.7218750789761543,
      "reward_std": 0.09028238290920854,
      "rewards/equation_reward_func": 0.7276786044239998,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 8362
    },
    {
      "completion_length": 256.7142963409424,
      "epoch": 1.4023638878410662,
      "grad_norm": 0.19437306007039692,
      "kl": 0.405792236328125,
      "learning_rate": 4.986621794971819e-07,
      "loss": 0.0004,
      "reward": 1.7428572326898575,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7517857402563095,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8364
    },
    {
      "completion_length": 257.4955463409424,
      "epoch": 1.4026991910809339,
      "grad_norm": 0.5732167804087518,
      "kl": 0.6250762939453125,
      "learning_rate": 4.986609398729121e-07,
      "loss": 0.0006,
      "reward": 1.7714286670088768,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 8366
    },
    {
      "completion_length": 262.8080472946167,
      "epoch": 1.4030344943208013,
      "grad_norm": 0.33261515236998473,
      "kl": 0.738128662109375,
      "learning_rate": 4.986596996761327e-07,
      "loss": 0.0007,
      "reward": 1.7375000715255737,
      "reward_std": 0.0883883461356163,
      "rewards/equation_reward_func": 0.7419643104076385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8368
    },
    {
      "completion_length": 258.48662281036377,
      "epoch": 1.403369797560669,
      "grad_norm": 0.4386239574155109,
      "kl": 1.22271728515625,
      "learning_rate": 4.986584589068465e-07,
      "loss": 0.0012,
      "reward": 1.7758929282426834,
      "reward_std": 0.04419417306780815,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 0.9973214343190193,
      "step": 8370
    },
    {
      "completion_length": 256.2455472946167,
      "epoch": 1.4037051008005366,
      "grad_norm": 0.27849866199909995,
      "kl": 1.667144775390625,
      "learning_rate": 4.986572175650562e-07,
      "loss": 0.0017,
      "reward": 1.7357143610715866,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7446428872644901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8372
    },
    {
      "completion_length": 243.52233028411865,
      "epoch": 1.404040404040404,
      "grad_norm": 0.25705270920538403,
      "kl": 0.59259033203125,
      "learning_rate": 4.986559756507649e-07,
      "loss": 0.0006,
      "reward": 1.7625000774860382,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7669643107801676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8374
    },
    {
      "completion_length": 251.6294755935669,
      "epoch": 1.4043757072802716,
      "grad_norm": 0.15817553115601674,
      "kl": 1.261627197265625,
      "learning_rate": 4.986547331639753e-07,
      "loss": 0.0013,
      "reward": 1.7321429550647736,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7410714533179998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8376
    },
    {
      "completion_length": 253.94644165039062,
      "epoch": 1.4047110105201392,
      "grad_norm": 0.14483387736198614,
      "kl": 1.48773193359375,
      "learning_rate": 4.986534901046903e-07,
      "loss": 0.0015,
      "reward": 1.814285770058632,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857365310192,
      "rewards/format_reward_func": 1.0,
      "step": 8378
    },
    {
      "completion_length": 250.39733505249023,
      "epoch": 1.4050463137600067,
      "grad_norm": 0.20242507143435626,
      "kl": 0.5549163818359375,
      "learning_rate": 4.986522464729129e-07,
      "loss": 0.0006,
      "reward": 1.7616072297096252,
      "reward_std": 0.07071067788638175,
      "rewards/equation_reward_func": 0.7732143178582191,
      "rewards/format_reward_func": 0.9883928671479225,
      "step": 8380
    },
    {
      "completion_length": 261.4821548461914,
      "epoch": 1.4053816169998743,
      "grad_norm": 0.3677156368563927,
      "kl": 0.290008544921875,
      "learning_rate": 4.986510022686456e-07,
      "loss": 0.0003,
      "reward": 1.8062500581145287,
      "reward_std": 0.07197336968965828,
      "rewards/equation_reward_func": 0.8223214484751225,
      "rewards/format_reward_func": 0.9839285798370838,
      "step": 8382
    },
    {
      "completion_length": 269.2812623977661,
      "epoch": 1.4057169202397417,
      "grad_norm": 0.5339161928148323,
      "kl": 0.60052490234375,
      "learning_rate": 4.986497574918917e-07,
      "loss": 0.0006,
      "reward": 1.7718750163912773,
      "reward_std": 0.10669736191630363,
      "rewards/equation_reward_func": 0.8053571619093418,
      "rewards/format_reward_func": 0.9665178768336773,
      "step": 8384
    },
    {
      "completion_length": 261.0803689956665,
      "epoch": 1.4060522234796093,
      "grad_norm": 0.21607394771384367,
      "kl": 0.5733642578125,
      "learning_rate": 4.986485121426538e-07,
      "loss": 0.0006,
      "reward": 1.728571467101574,
      "reward_std": 0.07071067858487368,
      "rewards/equation_reward_func": 0.746428582817316,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 8386
    },
    {
      "completion_length": 281.7589416503906,
      "epoch": 1.406387526719477,
      "grad_norm": 0.23092201842398968,
      "kl": 0.65911865234375,
      "learning_rate": 4.986472662209348e-07,
      "loss": 0.0007,
      "reward": 1.6799107789993286,
      "reward_std": 0.11932427063584328,
      "rewards/equation_reward_func": 0.715178593993187,
      "rewards/format_reward_func": 0.9647321589291096,
      "step": 8388
    },
    {
      "completion_length": 263.16519355773926,
      "epoch": 1.4067228299593446,
      "grad_norm": 0.30303730808498497,
      "kl": 0.55340576171875,
      "learning_rate": 4.986460197267376e-07,
      "loss": 0.0006,
      "reward": 1.7633929401636124,
      "reward_std": 0.07197336852550507,
      "rewards/equation_reward_func": 0.779464315623045,
      "rewards/format_reward_func": 0.9839285835623741,
      "step": 8390
    },
    {
      "completion_length": 287.9866180419922,
      "epoch": 1.407058133199212,
      "grad_norm": 0.3113564380005515,
      "kl": 0.73895263671875,
      "learning_rate": 4.986447726600651e-07,
      "loss": 0.0007,
      "reward": 1.6491071954369545,
      "reward_std": 0.128794448915869,
      "rewards/equation_reward_func": 0.6919643133878708,
      "rewards/format_reward_func": 0.9571428783237934,
      "step": 8392
    },
    {
      "completion_length": 260.9285821914673,
      "epoch": 1.4073934364390797,
      "grad_norm": 0.5877392569400088,
      "kl": 0.58837890625,
      "learning_rate": 4.986435250209201e-07,
      "loss": 0.0006,
      "reward": 1.7267857939004898,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7312500365078449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8394
    },
    {
      "completion_length": 270.6071538925171,
      "epoch": 1.407728739678947,
      "grad_norm": 0.37487510306904775,
      "kl": 1.393402099609375,
      "learning_rate": 4.986422768093056e-07,
      "loss": 0.0014,
      "reward": 1.771875038743019,
      "reward_std": 0.05997780826874077,
      "rewards/equation_reward_func": 0.786607176065445,
      "rewards/format_reward_func": 0.9852678664028645,
      "step": 8396
    },
    {
      "completion_length": 257.44197273254395,
      "epoch": 1.4080640429188147,
      "grad_norm": 0.1787303931438916,
      "kl": 0.350555419921875,
      "learning_rate": 4.986410280252244e-07,
      "loss": 0.0004,
      "reward": 1.7986607998609543,
      "reward_std": 0.05240166233852506,
      "rewards/equation_reward_func": 0.8000000193715096,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 8398
    },
    {
      "completion_length": 261.2142972946167,
      "epoch": 1.4083993461586823,
      "grad_norm": 0.3143881061853852,
      "kl": 0.587188720703125,
      "learning_rate": 4.986397786686793e-07,
      "loss": 0.0006,
      "reward": 1.7607143595814705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 8400
    },
    {
      "completion_length": 263.7767972946167,
      "epoch": 1.4087346493985498,
      "grad_norm": 0.608426074512046,
      "kl": 0.341552734375,
      "learning_rate": 4.986385287396733e-07,
      "loss": 0.0003,
      "reward": 1.7392857894301414,
      "reward_std": 0.10606601554900408,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 8402
    },
    {
      "completion_length": 274.3080472946167,
      "epoch": 1.4090699526384174,
      "grad_norm": 0.6285536610000946,
      "kl": 0.3597412109375,
      "learning_rate": 4.986372782382092e-07,
      "loss": 0.0004,
      "reward": 1.692857213318348,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.701785746961832,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8404
    },
    {
      "completion_length": 260.9910831451416,
      "epoch": 1.4094052558782848,
      "grad_norm": 0.10934691631824027,
      "kl": 0.264923095703125,
      "learning_rate": 4.986360271642898e-07,
      "loss": 0.0003,
      "reward": 1.7553571984171867,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7687500305473804,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8406
    },
    {
      "completion_length": 262.0044755935669,
      "epoch": 1.4097405591181524,
      "grad_norm": 0.14820756319286144,
      "kl": 0.15960693359375,
      "learning_rate": 4.986347755179181e-07,
      "loss": 0.0002,
      "reward": 1.8107143267989159,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107142932713032,
      "rewards/format_reward_func": 1.0,
      "step": 8408
    },
    {
      "completion_length": 260.5714406967163,
      "epoch": 1.41007586235802,
      "grad_norm": 0.21250860071479305,
      "kl": 0.140960693359375,
      "learning_rate": 4.98633523299097e-07,
      "loss": 0.0001,
      "reward": 1.7303572222590446,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7348214592784643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8410
    },
    {
      "completion_length": 258.4509029388428,
      "epoch": 1.4104111655978877,
      "grad_norm": 0.355366517952804,
      "kl": 0.192138671875,
      "learning_rate": 4.986322705078294e-07,
      "loss": 0.0002,
      "reward": 1.7053572237491608,
      "reward_std": 0.12374368496239185,
      "rewards/equation_reward_func": 0.7187500242143869,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8412
    },
    {
      "completion_length": 248.47322845458984,
      "epoch": 1.4107464688377551,
      "grad_norm": 0.24897194525629096,
      "kl": 0.14501953125,
      "learning_rate": 4.98631017144118e-07,
      "loss": 0.0001,
      "reward": 1.7928571924567223,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 8414
    },
    {
      "completion_length": 233.80804634094238,
      "epoch": 1.4110817720776228,
      "grad_norm": 0.22718979397807326,
      "kl": 0.15289306640625,
      "learning_rate": 4.986297632079659e-07,
      "loss": 0.0002,
      "reward": 1.767857201397419,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571790456772,
      "rewards/format_reward_func": 1.0,
      "step": 8416
    },
    {
      "completion_length": 247.35268783569336,
      "epoch": 1.4114170753174902,
      "grad_norm": 0.1969943143254704,
      "kl": 0.1800537109375,
      "learning_rate": 4.986285086993759e-07,
      "loss": 0.0002,
      "reward": 1.7500000819563866,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 8418
    },
    {
      "completion_length": 248.67412090301514,
      "epoch": 1.4117523785573578,
      "grad_norm": 0.3579731369561279,
      "kl": 0.17572021484375,
      "learning_rate": 4.986272536183509e-07,
      "loss": 0.0002,
      "reward": 1.771428607404232,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 8420
    },
    {
      "completion_length": 245.44197463989258,
      "epoch": 1.4120876817972254,
      "grad_norm": 0.19579481260834086,
      "kl": 0.20904541015625,
      "learning_rate": 4.986259979648938e-07,
      "loss": 0.0002,
      "reward": 1.6642857864499092,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.6732143200933933,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8422
    },
    {
      "completion_length": 250.2053680419922,
      "epoch": 1.4124229850370928,
      "grad_norm": 0.4279326961042182,
      "kl": 0.214385986328125,
      "learning_rate": 4.986247417390074e-07,
      "loss": 0.0002,
      "reward": 1.7553572058677673,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7598214522004128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8424
    },
    {
      "completion_length": 239.1562623977661,
      "epoch": 1.4127582882769605,
      "grad_norm": 0.3513471847587535,
      "kl": 0.18414306640625,
      "learning_rate": 4.986234849406947e-07,
      "loss": 0.0002,
      "reward": 1.7629465088248253,
      "reward_std": 0.06250318652018905,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 8426
    },
    {
      "completion_length": 245.0714406967163,
      "epoch": 1.413093591516828,
      "grad_norm": 0.3502444509464687,
      "kl": 0.243865966796875,
      "learning_rate": 4.986222275699585e-07,
      "loss": 0.0002,
      "reward": 1.8375000655651093,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.8419643118977547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8428
    },
    {
      "completion_length": 245.3125114440918,
      "epoch": 1.4134288947566955,
      "grad_norm": 0.437842987118793,
      "kl": 0.224090576171875,
      "learning_rate": 4.986209696268018e-07,
      "loss": 0.0002,
      "reward": 1.739285796880722,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7392857559025288,
      "rewards/format_reward_func": 1.0,
      "step": 8430
    },
    {
      "completion_length": 247.5312623977661,
      "epoch": 1.4137641979965632,
      "grad_norm": 0.19685596006311468,
      "kl": 0.1822509765625,
      "learning_rate": 4.986197111112275e-07,
      "loss": 0.0002,
      "reward": 1.7392857819795609,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 1.0,
      "step": 8432
    },
    {
      "completion_length": 244.63393783569336,
      "epoch": 1.4140995012364308,
      "grad_norm": 0.16053489041924518,
      "kl": 0.25244140625,
      "learning_rate": 4.986184520232383e-07,
      "loss": 0.0003,
      "reward": 1.7696429044008255,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7741071842610836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8434
    },
    {
      "completion_length": 242.8973331451416,
      "epoch": 1.4144348044762982,
      "grad_norm": 0.11632207449364086,
      "kl": 0.202850341796875,
      "learning_rate": 4.986171923628373e-07,
      "loss": 0.0002,
      "reward": 1.8035715073347092,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 8436
    },
    {
      "completion_length": 248.72769165039062,
      "epoch": 1.4147701077161658,
      "grad_norm": 0.3687579739976894,
      "kl": 0.3116455078125,
      "learning_rate": 4.986159321300274e-07,
      "loss": 0.0003,
      "reward": 1.7767857685685158,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7812500298023224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8438
    },
    {
      "completion_length": 246.8928680419922,
      "epoch": 1.4151054109560333,
      "grad_norm": 0.25257908439965243,
      "kl": 0.21636962890625,
      "learning_rate": 4.986146713248115e-07,
      "loss": 0.0002,
      "reward": 1.7642857804894447,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 8440
    },
    {
      "completion_length": 254.77679538726807,
      "epoch": 1.415440714195901,
      "grad_norm": 0.3316500270046741,
      "kl": 0.380126953125,
      "learning_rate": 4.986134099471923e-07,
      "loss": 0.0004,
      "reward": 1.757142923772335,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428846567869,
      "rewards/format_reward_func": 1.0,
      "step": 8442
    },
    {
      "completion_length": 244.6116189956665,
      "epoch": 1.4157760174357685,
      "grad_norm": 0.13862563902299166,
      "kl": 0.159881591796875,
      "learning_rate": 4.986121479971729e-07,
      "loss": 0.0002,
      "reward": 1.7357143312692642,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7357142996042967,
      "rewards/format_reward_func": 1.0,
      "step": 8444
    },
    {
      "completion_length": 241.01340770721436,
      "epoch": 1.4161113206756362,
      "grad_norm": 0.18927875554614454,
      "kl": 0.397674560546875,
      "learning_rate": 4.986108854747561e-07,
      "loss": 0.0004,
      "reward": 1.7620536461472511,
      "reward_std": 0.05366435460746288,
      "rewards/equation_reward_func": 0.7696428932249546,
      "rewards/format_reward_func": 0.9924107193946838,
      "step": 8446
    },
    {
      "completion_length": 251.79465579986572,
      "epoch": 1.4164466239155036,
      "grad_norm": 0.1946390969329135,
      "kl": 0.354339599609375,
      "learning_rate": 4.986096223799449e-07,
      "loss": 0.0004,
      "reward": 1.805357202887535,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.8187500312924385,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8448
    },
    {
      "completion_length": 262.28126525878906,
      "epoch": 1.4167819271553712,
      "grad_norm": 0.15546482453488383,
      "kl": 0.262237548828125,
      "learning_rate": 4.986083587127423e-07,
      "loss": 0.0003,
      "reward": 1.7732143476605415,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7776785977184772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8450
    },
    {
      "completion_length": 250.01786994934082,
      "epoch": 1.4171172303952386,
      "grad_norm": 0.25181161802683366,
      "kl": 0.218536376953125,
      "learning_rate": 4.98607094473151e-07,
      "loss": 0.0002,
      "reward": 1.796428643167019,
      "reward_std": 0.08586296625435352,
      "rewards/equation_reward_func": 0.8053571656346321,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8452
    },
    {
      "completion_length": 255.62500858306885,
      "epoch": 1.4174525336351063,
      "grad_norm": 0.3271275806533319,
      "kl": 0.414764404296875,
      "learning_rate": 4.98605829661174e-07,
      "loss": 0.0004,
      "reward": 1.7678571939468384,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.776785746216774,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8454
    },
    {
      "completion_length": 256.46875858306885,
      "epoch": 1.417787836874974,
      "grad_norm": 0.47129680996280954,
      "kl": 0.260772705078125,
      "learning_rate": 4.986045642768141e-07,
      "loss": 0.0003,
      "reward": 1.764285758137703,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 8456
    },
    {
      "completion_length": 259.5714387893677,
      "epoch": 1.4181231401148413,
      "grad_norm": 0.09692164782946838,
      "kl": 0.162628173828125,
      "learning_rate": 4.986032983200745e-07,
      "loss": 0.0002,
      "reward": 1.789285771548748,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7982143089175224,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8458
    },
    {
      "completion_length": 256.38840198516846,
      "epoch": 1.418458443354709,
      "grad_norm": 0.23052555459158877,
      "kl": 0.2418212890625,
      "learning_rate": 4.986020317909577e-07,
      "loss": 0.0002,
      "reward": 1.7785714864730835,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7875000275671482,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8460
    },
    {
      "completion_length": 259.8259048461914,
      "epoch": 1.4187937465945764,
      "grad_norm": 0.21238298783978266,
      "kl": 0.22589111328125,
      "learning_rate": 4.98600764689467e-07,
      "loss": 0.0002,
      "reward": 1.7500000894069672,
      "reward_std": 0.04040610231459141,
      "rewards/equation_reward_func": 0.7589285895228386,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8462
    },
    {
      "completion_length": 255.8125123977661,
      "epoch": 1.419129049834444,
      "grad_norm": 0.10720861997222418,
      "kl": 0.178863525390625,
      "learning_rate": 4.985994970156052e-07,
      "loss": 0.0002,
      "reward": 1.7410714775323868,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357313156128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8464
    },
    {
      "completion_length": 263.75001335144043,
      "epoch": 1.4194643530743116,
      "grad_norm": 0.2796565830384278,
      "kl": 0.134246826171875,
      "learning_rate": 4.985982287693751e-07,
      "loss": 0.0001,
      "reward": 1.7803572043776512,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7848214693367481,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8466
    },
    {
      "completion_length": 269.77679920196533,
      "epoch": 1.4197996563141793,
      "grad_norm": 0.1424100872454886,
      "kl": 0.282989501953125,
      "learning_rate": 4.985969599507797e-07,
      "loss": 0.0003,
      "reward": 1.741071492433548,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.745535746216774,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8468
    },
    {
      "completion_length": 255.5759038925171,
      "epoch": 1.4201349595540467,
      "grad_norm": 0.10532280309336838,
      "kl": 0.143096923828125,
      "learning_rate": 4.98595690559822e-07,
      "loss": 0.0001,
      "reward": 1.7446429505944252,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7491071708500385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8470
    },
    {
      "completion_length": 252.64733505249023,
      "epoch": 1.4204702627939143,
      "grad_norm": 0.4641752137066568,
      "kl": 0.17669677734375,
      "learning_rate": 4.985944205965048e-07,
      "loss": 0.0002,
      "reward": 1.723214365541935,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7276786006987095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8472
    },
    {
      "completion_length": 248.41965198516846,
      "epoch": 1.4208055660337817,
      "grad_norm": 0.20406659597155857,
      "kl": 0.146026611328125,
      "learning_rate": 4.985931500608311e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285980910063,
      "rewards/format_reward_func": 1.0,
      "step": 8474
    },
    {
      "completion_length": 246.37500953674316,
      "epoch": 1.4211408692736494,
      "grad_norm": 0.25091824175709215,
      "kl": 0.225311279296875,
      "learning_rate": 4.985918789528037e-07,
      "loss": 0.0002,
      "reward": 1.732142947614193,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7321428842842579,
      "rewards/format_reward_func": 1.0,
      "step": 8476
    },
    {
      "completion_length": 252.33929920196533,
      "epoch": 1.421476172513517,
      "grad_norm": 0.3030296561947238,
      "kl": 0.157745361328125,
      "learning_rate": 4.985906072724257e-07,
      "loss": 0.0002,
      "reward": 1.735714353621006,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7357143219560385,
      "rewards/format_reward_func": 1.0,
      "step": 8478
    },
    {
      "completion_length": 263.6250114440918,
      "epoch": 1.4218114757533844,
      "grad_norm": 0.3695755041875649,
      "kl": 0.115570068359375,
      "learning_rate": 4.985893350196999e-07,
      "loss": 0.0001,
      "reward": 1.7125000804662704,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7169643212109804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8480
    },
    {
      "completion_length": 256.9732255935669,
      "epoch": 1.422146778993252,
      "grad_norm": 0.1439424708583325,
      "kl": 0.1053314208984375,
      "learning_rate": 4.985880621946294e-07,
      "loss": 0.0001,
      "reward": 1.7392858043313026,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.748214315623045,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8482
    },
    {
      "completion_length": 263.15626335144043,
      "epoch": 1.4224820822331194,
      "grad_norm": 0.281594071406156,
      "kl": 0.112548828125,
      "learning_rate": 4.985867887972169e-07,
      "loss": 0.0001,
      "reward": 1.6857143715023994,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7035714685916901,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 8484
    },
    {
      "completion_length": 253.27679920196533,
      "epoch": 1.422817385472987,
      "grad_norm": 0.20893025538573007,
      "kl": 0.092681884765625,
      "learning_rate": 4.985855148274655e-07,
      "loss": 0.0001,
      "reward": 1.7767857611179352,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7812500298023224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8486
    },
    {
      "completion_length": 250.0714406967163,
      "epoch": 1.4231526887128547,
      "grad_norm": 0.16251693933127415,
      "kl": 0.122161865234375,
      "learning_rate": 4.985842402853781e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857357859612,
      "rewards/format_reward_func": 1.0,
      "step": 8488
    },
    {
      "completion_length": 253.47769260406494,
      "epoch": 1.4234879919527224,
      "grad_norm": 0.2043458697797928,
      "kl": 0.140777587890625,
      "learning_rate": 4.985829651709575e-07,
      "loss": 0.0001,
      "reward": 1.8303571939468384,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8348214440047741,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8490
    },
    {
      "completion_length": 246.39733123779297,
      "epoch": 1.4238232951925898,
      "grad_norm": 0.20184167414718476,
      "kl": 0.106292724609375,
      "learning_rate": 4.985816894842069e-07,
      "loss": 0.0001,
      "reward": 1.8035714849829674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714626312256,
      "rewards/format_reward_func": 1.0,
      "step": 8492
    },
    {
      "completion_length": 245.11608219146729,
      "epoch": 1.4241585984324574,
      "grad_norm": 0.1438843833872152,
      "kl": 0.119415283203125,
      "learning_rate": 4.985804132251289e-07,
      "loss": 0.0001,
      "reward": 1.769642911851406,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7741071749478579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8494
    },
    {
      "completion_length": 258.9151887893677,
      "epoch": 1.4244939016723248,
      "grad_norm": 0.26308992695053623,
      "kl": 0.12542724609375,
      "learning_rate": 4.985791363937267e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 8496
    },
    {
      "completion_length": 251.2410831451416,
      "epoch": 1.4248292049121924,
      "grad_norm": 0.1744191837380716,
      "kl": 0.1085205078125,
      "learning_rate": 4.985778589900032e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 8498
    },
    {
      "completion_length": 242.3348331451416,
      "epoch": 1.42516450815206,
      "grad_norm": 0.25796914063104437,
      "kl": 0.091278076171875,
      "learning_rate": 4.985765810139614e-07,
      "loss": 0.0001,
      "reward": 1.7660714834928513,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8500
    },
    {
      "completion_length": 243.51340198516846,
      "epoch": 1.4254998113919275,
      "grad_norm": 0.27267829180547887,
      "kl": 0.1027984619140625,
      "learning_rate": 4.98575302465604e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000193715096,
      "rewards/format_reward_func": 1.0,
      "step": 8502
    },
    {
      "completion_length": 253.4553689956665,
      "epoch": 1.4258351146317951,
      "grad_norm": 0.25539325970142795,
      "kl": 0.093658447265625,
      "learning_rate": 4.985740233449341e-07,
      "loss": 0.0001,
      "reward": 1.7482143491506577,
      "reward_std": 0.10354063473641872,
      "rewards/equation_reward_func": 0.7616071701049805,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8504
    },
    {
      "completion_length": 227.38393688201904,
      "epoch": 1.4261704178716628,
      "grad_norm": 0.2182021184194296,
      "kl": 0.0866546630859375,
      "learning_rate": 4.985727436519547e-07,
      "loss": 0.0001,
      "reward": 1.7821428999304771,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428962051868,
      "rewards/format_reward_func": 1.0,
      "step": 8506
    },
    {
      "completion_length": 233.50447368621826,
      "epoch": 1.4265057211115302,
      "grad_norm": 0.12847280880524686,
      "kl": 0.1015472412109375,
      "learning_rate": 4.985714633866685e-07,
      "loss": 0.0001,
      "reward": 1.728571504354477,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7285714484751225,
      "rewards/format_reward_func": 1.0,
      "step": 8508
    },
    {
      "completion_length": 247.74108219146729,
      "epoch": 1.4268410243513978,
      "grad_norm": 0.17515847834387244,
      "kl": 0.113861083984375,
      "learning_rate": 4.985701825490789e-07,
      "loss": 0.0001,
      "reward": 1.7125000953674316,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.716964328661561,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8510
    },
    {
      "completion_length": 241.46429824829102,
      "epoch": 1.4271763275912654,
      "grad_norm": 0.18851509196918953,
      "kl": 0.089111328125,
      "learning_rate": 4.985689011391884e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428999304771,
      "rewards/format_reward_func": 1.0,
      "step": 8512
    },
    {
      "completion_length": 236.56697463989258,
      "epoch": 1.4275116308311329,
      "grad_norm": 0.1949627540989246,
      "kl": 0.0927734375,
      "learning_rate": 4.985676191570001e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 8514
    },
    {
      "completion_length": 228.66518878936768,
      "epoch": 1.4278469340710005,
      "grad_norm": 0.2164410413970927,
      "kl": 0.0898284912109375,
      "learning_rate": 4.985663366025171e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 8516
    },
    {
      "completion_length": 230.446439743042,
      "epoch": 1.428182237310868,
      "grad_norm": 0.10347510631654368,
      "kl": 0.11016845703125,
      "learning_rate": 4.985650534757421e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 8518
    },
    {
      "completion_length": 238.00447368621826,
      "epoch": 1.4285175405507355,
      "grad_norm": 0.1783372713511173,
      "kl": 0.0781707763671875,
      "learning_rate": 4.985637697766783e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.767857177183032,
      "rewards/format_reward_func": 1.0,
      "step": 8520
    },
    {
      "completion_length": 246.2767972946167,
      "epoch": 1.4288528437906032,
      "grad_norm": 0.2876657858107366,
      "kl": 0.0819549560546875,
      "learning_rate": 4.985624855053286e-07,
      "loss": 0.0001,
      "reward": 1.7321429178118706,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321428935974836,
      "rewards/format_reward_func": 1.0,
      "step": 8522
    },
    {
      "completion_length": 230.7544755935669,
      "epoch": 1.4291881470304708,
      "grad_norm": 0.22398410695846024,
      "kl": 0.0896148681640625,
      "learning_rate": 4.985612006616957e-07,
      "loss": 0.0001,
      "reward": 1.7892857939004898,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 8524
    },
    {
      "completion_length": 241.4241189956665,
      "epoch": 1.4295234502703382,
      "grad_norm": 0.5155656556852308,
      "kl": 0.111846923828125,
      "learning_rate": 4.985599152457829e-07,
      "loss": 0.0001,
      "reward": 1.764285795390606,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 8526
    },
    {
      "completion_length": 237.36608219146729,
      "epoch": 1.4298587535102059,
      "grad_norm": 0.12106024005112531,
      "kl": 0.085052490234375,
      "learning_rate": 4.985586292575929e-07,
      "loss": 0.0001,
      "reward": 1.7678571790456772,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571697324514,
      "rewards/format_reward_func": 1.0,
      "step": 8528
    },
    {
      "completion_length": 227.28572845458984,
      "epoch": 1.4301940567500733,
      "grad_norm": 0.15222247755437548,
      "kl": 0.0845947265625,
      "learning_rate": 4.985573426971289e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 8530
    },
    {
      "completion_length": 240.3705472946167,
      "epoch": 1.430529359989941,
      "grad_norm": 0.23859971475872568,
      "kl": 0.091400146484375,
      "learning_rate": 4.985560555643937e-07,
      "loss": 0.0001,
      "reward": 1.78035718947649,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8532
    },
    {
      "completion_length": 243.41072463989258,
      "epoch": 1.4308646632298085,
      "grad_norm": 0.22536579496061887,
      "kl": 0.081878662109375,
      "learning_rate": 4.985547678593903e-07,
      "loss": 0.0001,
      "reward": 1.7232143357396126,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7276785969734192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8534
    },
    {
      "completion_length": 247.3750123977661,
      "epoch": 1.431199966469676,
      "grad_norm": 0.2802477726314655,
      "kl": 0.079833984375,
      "learning_rate": 4.985534795821217e-07,
      "loss": 0.0001,
      "reward": 1.728571504354477,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7285714633762836,
      "rewards/format_reward_func": 1.0,
      "step": 8536
    },
    {
      "completion_length": 244.95983123779297,
      "epoch": 1.4315352697095436,
      "grad_norm": 0.22365417759999096,
      "kl": 0.0858612060546875,
      "learning_rate": 4.985521907325907e-07,
      "loss": 0.0001,
      "reward": 1.7160715386271477,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7205357328057289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8538
    },
    {
      "completion_length": 247.64286708831787,
      "epoch": 1.431870572949411,
      "grad_norm": 0.29738940177670603,
      "kl": 0.0897216796875,
      "learning_rate": 4.985509013108005e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 8540
    },
    {
      "completion_length": 242.60269260406494,
      "epoch": 1.4322058761892786,
      "grad_norm": 0.12314884860826378,
      "kl": 0.092803955078125,
      "learning_rate": 4.985496113167539e-07,
      "loss": 0.0001,
      "reward": 1.8321429044008255,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8321428932249546,
      "rewards/format_reward_func": 1.0,
      "step": 8542
    },
    {
      "completion_length": 236.54018878936768,
      "epoch": 1.4325411794291463,
      "grad_norm": 0.28216533234311264,
      "kl": 0.093536376953125,
      "learning_rate": 4.985483207504541e-07,
      "loss": 0.0001,
      "reward": 1.7428572326898575,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 8544
    },
    {
      "completion_length": 234.81250953674316,
      "epoch": 1.432876482669014,
      "grad_norm": 0.22984926963704252,
      "kl": 0.110687255859375,
      "learning_rate": 4.985470296119038e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 8546
    },
    {
      "completion_length": 232.24554634094238,
      "epoch": 1.4332117859088813,
      "grad_norm": 0.11927480159230068,
      "kl": 0.120819091796875,
      "learning_rate": 4.985457379011061e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 8548
    },
    {
      "completion_length": 230.32143783569336,
      "epoch": 1.433547089148749,
      "grad_norm": 0.20098209710952783,
      "kl": 0.1114654541015625,
      "learning_rate": 4.98544445618064e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428719907999,
      "rewards/format_reward_func": 1.0,
      "step": 8550
    },
    {
      "completion_length": 237.6294755935669,
      "epoch": 1.4338823923886164,
      "grad_norm": 0.12974919344947425,
      "kl": 0.113372802734375,
      "learning_rate": 4.985431527627804e-07,
      "loss": 0.0001,
      "reward": 1.800000049173832,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.8089285865426064,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8552
    },
    {
      "completion_length": 242.09375953674316,
      "epoch": 1.434217695628484,
      "grad_norm": 0.1779069025850947,
      "kl": 0.11090087890625,
      "learning_rate": 4.985418593352583e-07,
      "loss": 0.0001,
      "reward": 1.801785796880722,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.8062500208616257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8554
    },
    {
      "completion_length": 253.00894165039062,
      "epoch": 1.4345529988683516,
      "grad_norm": 0.39904585725146463,
      "kl": 0.11761474609375,
      "learning_rate": 4.985405653355006e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7732143327593803,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8556
    },
    {
      "completion_length": 241.35268783569336,
      "epoch": 1.434888302108219,
      "grad_norm": 0.20989425840657888,
      "kl": 0.116424560546875,
      "learning_rate": 4.985392707635104e-07,
      "loss": 0.0001,
      "reward": 1.7535714656114578,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714600235224,
      "rewards/format_reward_func": 1.0,
      "step": 8558
    },
    {
      "completion_length": 254.6339406967163,
      "epoch": 1.4352236053480867,
      "grad_norm": 0.5080392816810477,
      "kl": 0.169158935546875,
      "learning_rate": 4.985379756192908e-07,
      "loss": 0.0002,
      "reward": 1.7428572177886963,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7517857514321804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8560
    },
    {
      "completion_length": 250.40625953674316,
      "epoch": 1.435558908587954,
      "grad_norm": 0.40974130131501063,
      "kl": 0.1434326171875,
      "learning_rate": 4.985366799028445e-07,
      "loss": 0.0001,
      "reward": 1.7361607775092125,
      "reward_std": 0.05997780663892627,
      "rewards/equation_reward_func": 0.741964302957058,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 8562
    },
    {
      "completion_length": 238.5759038925171,
      "epoch": 1.4358942118278217,
      "grad_norm": 0.3081459571029257,
      "kl": 0.13323974609375,
      "learning_rate": 4.985353836141746e-07,
      "loss": 0.0001,
      "reward": 1.742857187986374,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7517857477068901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8564
    },
    {
      "completion_length": 245.1116189956665,
      "epoch": 1.4362295150676894,
      "grad_norm": 0.17198720339414358,
      "kl": 0.1402435302734375,
      "learning_rate": 4.985340867532841e-07,
      "loss": 0.0001,
      "reward": 1.7285714969038963,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7285714708268642,
      "rewards/format_reward_func": 1.0,
      "step": 8566
    },
    {
      "completion_length": 245.3437614440918,
      "epoch": 1.436564818307557,
      "grad_norm": 0.19688696870277422,
      "kl": 0.13861083984375,
      "learning_rate": 4.98532789320176e-07,
      "loss": 0.0001,
      "reward": 1.7464286237955093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464286163449287,
      "rewards/format_reward_func": 1.0,
      "step": 8568
    },
    {
      "completion_length": 236.9910831451416,
      "epoch": 1.4369001215474244,
      "grad_norm": 0.1986506605536204,
      "kl": 0.130035400390625,
      "learning_rate": 4.985314913148534e-07,
      "loss": 0.0001,
      "reward": 1.7910715118050575,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7955357432365417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8570
    },
    {
      "completion_length": 236.37054634094238,
      "epoch": 1.437235424787292,
      "grad_norm": 0.1790702839307691,
      "kl": 0.146270751953125,
      "learning_rate": 4.98530192737319e-07,
      "loss": 0.0001,
      "reward": 1.7964286133646965,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.796428594738245,
      "rewards/format_reward_func": 1.0,
      "step": 8572
    },
    {
      "completion_length": 252.28126049041748,
      "epoch": 1.4375707280271595,
      "grad_norm": 0.224827230192582,
      "kl": 0.1280517578125,
      "learning_rate": 4.98528893587576e-07,
      "loss": 0.0001,
      "reward": 1.7526786476373672,
      "reward_std": 0.05682107945904136,
      "rewards/equation_reward_func": 0.7589285969734192,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 8574
    },
    {
      "completion_length": 242.94643878936768,
      "epoch": 1.437906031267027,
      "grad_norm": 0.40589939211072706,
      "kl": 0.128082275390625,
      "learning_rate": 4.985275938656273e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.11111677810549736,
      "rewards/equation_reward_func": 0.7750000394880772,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 8576
    },
    {
      "completion_length": 240.07590293884277,
      "epoch": 1.4382413345068947,
      "grad_norm": 0.2912371749687889,
      "kl": 0.328521728515625,
      "learning_rate": 4.98526293571476e-07,
      "loss": 0.0003,
      "reward": 1.705357238650322,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7098214663565159,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8578
    },
    {
      "completion_length": 242.60715293884277,
      "epoch": 1.4385766377467624,
      "grad_norm": 0.21713702252679135,
      "kl": 0.2657470703125,
      "learning_rate": 4.98524992705125e-07,
      "loss": 0.0003,
      "reward": 1.742857202887535,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 8580
    },
    {
      "completion_length": 248.51340293884277,
      "epoch": 1.4389119409866298,
      "grad_norm": 0.34041115823987345,
      "kl": 0.33697509765625,
      "learning_rate": 4.985236912665773e-07,
      "loss": 0.0003,
      "reward": 1.7482143640518188,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7526785917580128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8582
    },
    {
      "completion_length": 252.89733219146729,
      "epoch": 1.4392472442264974,
      "grad_norm": 0.3167723579300486,
      "kl": 0.2979736328125,
      "learning_rate": 4.985223892558359e-07,
      "loss": 0.0003,
      "reward": 1.757142923772335,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 8584
    },
    {
      "completion_length": 232.28125953674316,
      "epoch": 1.4395825474663648,
      "grad_norm": 0.14006470671982402,
      "kl": 0.265777587890625,
      "learning_rate": 4.985210866729038e-07,
      "loss": 0.0003,
      "reward": 1.7053572237491608,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.709821455180645,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8586
    },
    {
      "completion_length": 240.96429347991943,
      "epoch": 1.4399178507062325,
      "grad_norm": 0.5398088833976759,
      "kl": 0.422210693359375,
      "learning_rate": 4.985197835177841e-07,
      "loss": 0.0004,
      "reward": 1.7107143625617027,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.719642885029316,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8588
    },
    {
      "completion_length": 252.01340293884277,
      "epoch": 1.4402531539461,
      "grad_norm": 0.19388137471372102,
      "kl": 0.404083251953125,
      "learning_rate": 4.985184797904797e-07,
      "loss": 0.0004,
      "reward": 1.7017858028411865,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.706250037997961,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8590
    },
    {
      "completion_length": 249.28572750091553,
      "epoch": 1.4405884571859675,
      "grad_norm": 0.1371797585529847,
      "kl": 0.428375244140625,
      "learning_rate": 4.985171754909936e-07,
      "loss": 0.0004,
      "reward": 1.7428572252392769,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7517857365310192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8592
    },
    {
      "completion_length": 244.2500114440918,
      "epoch": 1.4409237604258351,
      "grad_norm": 0.23078455389106672,
      "kl": 0.218597412109375,
      "learning_rate": 4.985158706193287e-07,
      "loss": 0.0002,
      "reward": 1.7750000581145287,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7839285936206579,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8594
    },
    {
      "completion_length": 251.49108219146729,
      "epoch": 1.4412590636657026,
      "grad_norm": 0.2085646350562671,
      "kl": 0.250274658203125,
      "learning_rate": 4.985145651754883e-07,
      "loss": 0.0003,
      "reward": 1.7625000551342964,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8596
    },
    {
      "completion_length": 253.1517972946167,
      "epoch": 1.4415943669055702,
      "grad_norm": 0.19498147177236294,
      "kl": 0.826019287109375,
      "learning_rate": 4.985132591594751e-07,
      "loss": 0.0008,
      "reward": 1.7857143357396126,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143022119999,
      "rewards/format_reward_func": 1.0,
      "step": 8598
    },
    {
      "completion_length": 234.71876049041748,
      "epoch": 1.4419296701454378,
      "grad_norm": 0.544497343527771,
      "kl": 0.1741943359375,
      "learning_rate": 4.985119525712924e-07,
      "loss": 0.0002,
      "reward": 1.76071435213089,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143297791481,
      "rewards/format_reward_func": 1.0,
      "step": 8600
    },
    {
      "completion_length": 230.98661708831787,
      "epoch": 1.4422649733853055,
      "grad_norm": 0.2627565818988088,
      "kl": 0.214813232421875,
      "learning_rate": 4.98510645410943e-07,
      "loss": 0.0002,
      "reward": 1.716071479022503,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7205357514321804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8602
    },
    {
      "completion_length": 233.70090103149414,
      "epoch": 1.4426002766251729,
      "grad_norm": 0.6021051909045801,
      "kl": 0.244903564453125,
      "learning_rate": 4.985093376784298e-07,
      "loss": 0.0002,
      "reward": 1.7660715132951736,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7705357521772385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8604
    },
    {
      "completion_length": 235.54911518096924,
      "epoch": 1.4429355798650405,
      "grad_norm": 0.2665148810564702,
      "kl": 0.25433349609375,
      "learning_rate": 4.98508029373756e-07,
      "loss": 0.0003,
      "reward": 1.7142858058214188,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7142857424914837,
      "rewards/format_reward_func": 1.0,
      "step": 8606
    },
    {
      "completion_length": 232.60268688201904,
      "epoch": 1.443270883104908,
      "grad_norm": 0.34224759146463907,
      "kl": 0.1395263671875,
      "learning_rate": 4.985067204969247e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 8608
    },
    {
      "completion_length": 233.10715579986572,
      "epoch": 1.4436061863447756,
      "grad_norm": 0.19504296285660103,
      "kl": 0.133758544921875,
      "learning_rate": 4.985054110479387e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 8610
    },
    {
      "completion_length": 239.98215293884277,
      "epoch": 1.4439414895846432,
      "grad_norm": 0.17531776073423436,
      "kl": 0.121978759765625,
      "learning_rate": 4.985041010268011e-07,
      "loss": 0.0001,
      "reward": 1.7250000685453415,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000387430191,
      "rewards/format_reward_func": 1.0,
      "step": 8612
    },
    {
      "completion_length": 219.0044755935669,
      "epoch": 1.4442767928245106,
      "grad_norm": 0.5596495225303464,
      "kl": 0.151397705078125,
      "learning_rate": 4.98502790433515e-07,
      "loss": 0.0002,
      "reward": 1.8142857402563095,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8142857477068901,
      "rewards/format_reward_func": 1.0,
      "step": 8614
    },
    {
      "completion_length": 229.80358219146729,
      "epoch": 1.4446120960643782,
      "grad_norm": 0.15586386632307309,
      "kl": 0.11224365234375,
      "learning_rate": 4.985014792680833e-07,
      "loss": 0.0001,
      "reward": 1.7035714834928513,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.703571455553174,
      "rewards/format_reward_func": 1.0,
      "step": 8616
    },
    {
      "completion_length": 245.8526906967163,
      "epoch": 1.4449473993042456,
      "grad_norm": 0.13640709963299535,
      "kl": 0.118865966796875,
      "learning_rate": 4.98500167530509e-07,
      "loss": 0.0001,
      "reward": 1.6883929371833801,
      "reward_std": 0.06692260596901178,
      "rewards/equation_reward_func": 0.6946429014205933,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 8618
    },
    {
      "completion_length": 227.4687614440918,
      "epoch": 1.4452827025441133,
      "grad_norm": 0.2478581313144655,
      "kl": 0.0968017578125,
      "learning_rate": 4.984988552207952e-07,
      "loss": 0.0001,
      "reward": 1.7535715252161026,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 8620
    },
    {
      "completion_length": 242.321439743042,
      "epoch": 1.445618005783981,
      "grad_norm": 0.4948357753835724,
      "kl": 0.118927001953125,
      "learning_rate": 4.984975423389449e-07,
      "loss": 0.0001,
      "reward": 1.7375000789761543,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643253087997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8622
    },
    {
      "completion_length": 230.17858123779297,
      "epoch": 1.4459533090238486,
      "grad_norm": 0.13151233430921153,
      "kl": 0.130401611328125,
      "learning_rate": 4.984962288849611e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.767857164144516,
      "rewards/format_reward_func": 1.0,
      "step": 8624
    },
    {
      "completion_length": 234.3616189956665,
      "epoch": 1.446288612263716,
      "grad_norm": 0.3987436135276987,
      "kl": 0.14691162109375,
      "learning_rate": 4.98494914858847e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7553571704775095,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8626
    },
    {
      "completion_length": 239.37054920196533,
      "epoch": 1.4466239155035836,
      "grad_norm": 0.5486163983262353,
      "kl": 0.307373046875,
      "learning_rate": 4.984936002606053e-07,
      "loss": 0.0003,
      "reward": 1.8000000566244125,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000342726707,
      "rewards/format_reward_func": 1.0,
      "step": 8628
    },
    {
      "completion_length": 230.57590675354004,
      "epoch": 1.446959218743451,
      "grad_norm": 0.12681436673323587,
      "kl": 0.11016845703125,
      "learning_rate": 4.984922850902392e-07,
      "loss": 0.0001,
      "reward": 1.7375000715255737,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643197208643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8630
    },
    {
      "completion_length": 241.89286994934082,
      "epoch": 1.4472945219833186,
      "grad_norm": 0.18086561065557644,
      "kl": 0.263641357421875,
      "learning_rate": 4.984909693477518e-07,
      "loss": 0.0003,
      "reward": 1.7285714894533157,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7285714596509933,
      "rewards/format_reward_func": 1.0,
      "step": 8632
    },
    {
      "completion_length": 242.1562623977661,
      "epoch": 1.4476298252231863,
      "grad_norm": 0.2787489245249078,
      "kl": 0.3875732421875,
      "learning_rate": 4.98489653033146e-07,
      "loss": 0.0004,
      "reward": 1.7125000804662704,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7169643305242062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8634
    },
    {
      "completion_length": 245.22322463989258,
      "epoch": 1.4479651284630537,
      "grad_norm": 0.2814785129957571,
      "kl": 0.5368499755859375,
      "learning_rate": 4.98488336146425e-07,
      "loss": 0.0005,
      "reward": 1.7464286237955093,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 8636
    },
    {
      "completion_length": 240.5000114440918,
      "epoch": 1.4483004317029213,
      "grad_norm": 0.24754685535957013,
      "kl": 0.8140869140625,
      "learning_rate": 4.984870186875916e-07,
      "loss": 0.0008,
      "reward": 1.7821429297327995,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 8638
    },
    {
      "completion_length": 232.0669755935669,
      "epoch": 1.448635734942789,
      "grad_norm": 0.24005853574977867,
      "kl": 0.136199951171875,
      "learning_rate": 4.984857006566489e-07,
      "loss": 0.0001,
      "reward": 1.7071429565548897,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7071428876370192,
      "rewards/format_reward_func": 1.0,
      "step": 8640
    },
    {
      "completion_length": 239.9776906967163,
      "epoch": 1.4489710381826564,
      "grad_norm": 0.3229400482830985,
      "kl": 0.5002593994140625,
      "learning_rate": 4.984843820536002e-07,
      "loss": 0.0005,
      "reward": 1.7500000596046448,
      "reward_std": 0.08081220369786024,
      "rewards/equation_reward_func": 0.7589285913854837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8642
    },
    {
      "completion_length": 246.5714406967163,
      "epoch": 1.449306341422524,
      "grad_norm": 0.19913141938149045,
      "kl": 0.5619049072265625,
      "learning_rate": 4.984830628784482e-07,
      "loss": 0.0006,
      "reward": 1.74642863124609,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7553571835160255,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8644
    },
    {
      "completion_length": 256.48215675354004,
      "epoch": 1.4496416446623916,
      "grad_norm": 0.1987641534881554,
      "kl": 0.1080780029296875,
      "learning_rate": 4.984817431311961e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7767857350409031,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8646
    },
    {
      "completion_length": 243.52679634094238,
      "epoch": 1.449976947902259,
      "grad_norm": 0.388932344482841,
      "kl": 0.1192169189453125,
      "learning_rate": 4.984804228118468e-07,
      "loss": 0.0001,
      "reward": 1.741071492433548,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 0.987500011920929,
      "step": 8648
    },
    {
      "completion_length": 253.69643688201904,
      "epoch": 1.4503122511421267,
      "grad_norm": 0.12055179889101425,
      "kl": 0.193603515625,
      "learning_rate": 4.984791019204034e-07,
      "loss": 0.0002,
      "reward": 1.7267857640981674,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7401786055415869,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8650
    },
    {
      "completion_length": 256.64287185668945,
      "epoch": 1.450647554381994,
      "grad_norm": 0.45740297022280063,
      "kl": 0.192108154296875,
      "learning_rate": 4.984777804568692e-07,
      "loss": 0.0002,
      "reward": 1.785714328289032,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 8652
    },
    {
      "completion_length": 255.66519165039062,
      "epoch": 1.4509828576218617,
      "grad_norm": 0.17801379402730066,
      "kl": 0.08984375,
      "learning_rate": 4.984764584212469e-07,
      "loss": 0.0001,
      "reward": 1.7803572192788124,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214618861675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8654
    },
    {
      "completion_length": 267.5491199493408,
      "epoch": 1.4513181608617294,
      "grad_norm": 0.11665353846037967,
      "kl": 0.127777099609375,
      "learning_rate": 4.984751358135396e-07,
      "loss": 0.0001,
      "reward": 1.714285783469677,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7232143133878708,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8656
    },
    {
      "completion_length": 263.14287090301514,
      "epoch": 1.451653464101597,
      "grad_norm": 0.25408979673938825,
      "kl": 0.13909912109375,
      "learning_rate": 4.984738126337504e-07,
      "loss": 0.0001,
      "reward": 1.6875000521540642,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7008928880095482,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8658
    },
    {
      "completion_length": 272.9107255935669,
      "epoch": 1.4519887673414644,
      "grad_norm": 0.18260584338052813,
      "kl": 0.2103271484375,
      "learning_rate": 4.984724888818824e-07,
      "loss": 0.0002,
      "reward": 1.7982143387198448,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8026785962283611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8660
    },
    {
      "completion_length": 264.65625953674316,
      "epoch": 1.452324070581332,
      "grad_norm": 0.06142023819391897,
      "kl": 0.0996246337890625,
      "learning_rate": 4.984711645579387e-07,
      "loss": 0.0001,
      "reward": 1.7571429088711739,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7660714685916901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8662
    },
    {
      "completion_length": 263.11162090301514,
      "epoch": 1.4526593738211995,
      "grad_norm": 0.1881334188221849,
      "kl": 0.0886993408203125,
      "learning_rate": 4.984698396619221e-07,
      "loss": 0.0001,
      "reward": 1.7464286163449287,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 8664
    },
    {
      "completion_length": 257.37500953674316,
      "epoch": 1.452994677061067,
      "grad_norm": 0.18317443366305186,
      "kl": 0.09564208984375,
      "learning_rate": 4.98468514193836e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7875000275671482,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8666
    },
    {
      "completion_length": 268.60715770721436,
      "epoch": 1.4533299803009347,
      "grad_norm": 0.4507416161216136,
      "kl": 0.1070556640625,
      "learning_rate": 4.984671881536831e-07,
      "loss": 0.0001,
      "reward": 1.6857143715023994,
      "reward_std": 0.11111677810549736,
      "rewards/equation_reward_func": 0.7035714704543352,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 8668
    },
    {
      "completion_length": 267.8214387893677,
      "epoch": 1.4536652835408022,
      "grad_norm": 0.1606135561413552,
      "kl": 0.0889129638671875,
      "learning_rate": 4.984658615414666e-07,
      "loss": 0.0001,
      "reward": 1.812500074505806,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8169643022119999,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8670
    },
    {
      "completion_length": 254.62947463989258,
      "epoch": 1.4540005867806698,
      "grad_norm": 0.13900528444515145,
      "kl": 0.0863189697265625,
      "learning_rate": 4.984645343571896e-07,
      "loss": 0.0001,
      "reward": 1.7875000685453415,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7919643148779869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8672
    },
    {
      "completion_length": 272.78126525878906,
      "epoch": 1.4543358900205372,
      "grad_norm": 0.20889585300332983,
      "kl": 0.0862579345703125,
      "learning_rate": 4.984632066008551e-07,
      "loss": 0.0001,
      "reward": 1.7571428939700127,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 8674
    },
    {
      "completion_length": 270.0491189956665,
      "epoch": 1.4546711932604048,
      "grad_norm": 0.23130697896746683,
      "kl": 0.1115264892578125,
      "learning_rate": 4.984618782724662e-07,
      "loss": 0.0001,
      "reward": 1.7473215013742447,
      "reward_std": 0.06439722282812,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 8676
    },
    {
      "completion_length": 267.8303699493408,
      "epoch": 1.4550064965002725,
      "grad_norm": 0.32041604349225783,
      "kl": 0.11407470703125,
      "learning_rate": 4.98460549372026e-07,
      "loss": 0.0001,
      "reward": 1.7392858117818832,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7482143118977547,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8678
    },
    {
      "completion_length": 265.58037185668945,
      "epoch": 1.45534179974014,
      "grad_norm": 0.14919159431849408,
      "kl": 0.1207122802734375,
      "learning_rate": 4.984592198995373e-07,
      "loss": 0.0001,
      "reward": 1.7125000581145287,
      "reward_std": 0.06313453428447247,
      "rewards/equation_reward_func": 0.725892897695303,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8680
    },
    {
      "completion_length": 265.38394260406494,
      "epoch": 1.4556771029800075,
      "grad_norm": 0.3035788406369713,
      "kl": 0.1330718994140625,
      "learning_rate": 4.984578898550035e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 8682
    },
    {
      "completion_length": 254.71875953674316,
      "epoch": 1.4560124062198752,
      "grad_norm": 0.19167303109979877,
      "kl": 0.13134765625,
      "learning_rate": 4.984565592384275e-07,
      "loss": 0.0001,
      "reward": 1.6964286416769028,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7053571902215481,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8684
    },
    {
      "completion_length": 265.8303689956665,
      "epoch": 1.4563477094597426,
      "grad_norm": 0.18865444325673764,
      "kl": 0.0995635986328125,
      "learning_rate": 4.984552280498124e-07,
      "loss": 0.0001,
      "reward": 1.8321429044008255,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.832142885774374,
      "rewards/format_reward_func": 1.0,
      "step": 8686
    },
    {
      "completion_length": 256.9241247177124,
      "epoch": 1.4566830126996102,
      "grad_norm": 0.8027363666604862,
      "kl": 0.3691558837890625,
      "learning_rate": 4.984538962891612e-07,
      "loss": 0.0004,
      "reward": 1.7750000655651093,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 8688
    },
    {
      "completion_length": 261.95536708831787,
      "epoch": 1.4570183159394778,
      "grad_norm": 0.19079471714756488,
      "kl": 0.086334228515625,
      "learning_rate": 4.984525639564771e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7875000350177288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8690
    },
    {
      "completion_length": 242.62947463989258,
      "epoch": 1.4573536191793452,
      "grad_norm": 0.16115868086484084,
      "kl": 0.103118896484375,
      "learning_rate": 4.984512310517629e-07,
      "loss": 0.0001,
      "reward": 1.764732226729393,
      "reward_std": 0.056189734023064375,
      "rewards/equation_reward_func": 0.770535733550787,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 8692
    },
    {
      "completion_length": 257.6428699493408,
      "epoch": 1.4576889224192129,
      "grad_norm": 0.18748173456319275,
      "kl": 0.13623046875,
      "learning_rate": 4.98449897575022e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 8694
    },
    {
      "completion_length": 253.9732255935669,
      "epoch": 1.4580242256590803,
      "grad_norm": 0.0670366887196531,
      "kl": 0.087493896484375,
      "learning_rate": 4.984485635262573e-07,
      "loss": 0.0001,
      "reward": 1.7625000476837158,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7669643275439739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8696
    },
    {
      "completion_length": 255.08483123779297,
      "epoch": 1.458359528898948,
      "grad_norm": 0.46439941711117166,
      "kl": 0.146087646484375,
      "learning_rate": 4.984472289054718e-07,
      "loss": 0.0001,
      "reward": 1.7660714983940125,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7705357521772385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8698
    },
    {
      "completion_length": 258.120548248291,
      "epoch": 1.4586948321388156,
      "grad_norm": 0.13445293436482703,
      "kl": 0.2231903076171875,
      "learning_rate": 4.984458937126687e-07,
      "loss": 0.0002,
      "reward": 1.782142922282219,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 8700
    },
    {
      "completion_length": 253.5178689956665,
      "epoch": 1.4590301353786832,
      "grad_norm": 0.2825612021039923,
      "kl": 0.203125,
      "learning_rate": 4.984445579478512e-07,
      "loss": 0.0002,
      "reward": 1.7535715103149414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 8702
    },
    {
      "completion_length": 255.8616180419922,
      "epoch": 1.4593654386185506,
      "grad_norm": 0.0662004696664487,
      "kl": 0.160614013671875,
      "learning_rate": 4.98443221611022e-07,
      "loss": 0.0002,
      "reward": 1.7928571701049805,
      "reward_std": 0.030304577201604843,
      "rewards/equation_reward_func": 0.8017857428640127,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8704
    },
    {
      "completion_length": 247.165189743042,
      "epoch": 1.4597007418584182,
      "grad_norm": 0.14048445465509524,
      "kl": 0.146881103515625,
      "learning_rate": 4.984418847021845e-07,
      "loss": 0.0001,
      "reward": 1.8321429044008255,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.832142885774374,
      "rewards/format_reward_func": 1.0,
      "step": 8706
    },
    {
      "completion_length": 253.55804538726807,
      "epoch": 1.4600360450982857,
      "grad_norm": 0.13718770727586044,
      "kl": 0.219635009765625,
      "learning_rate": 4.984405472213415e-07,
      "loss": 0.0002,
      "reward": 1.789285771548748,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 8708
    },
    {
      "completion_length": 253.86161613464355,
      "epoch": 1.4603713483381533,
      "grad_norm": 0.17726014798555442,
      "kl": 0.115692138671875,
      "learning_rate": 4.984392091684965e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 8710
    },
    {
      "completion_length": 258.8660840988159,
      "epoch": 1.460706651578021,
      "grad_norm": 0.10583912091027659,
      "kl": 0.15997314453125,
      "learning_rate": 4.984378705436521e-07,
      "loss": 0.0002,
      "reward": 1.7464286610484123,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 8712
    },
    {
      "completion_length": 246.37054347991943,
      "epoch": 1.4610419548178886,
      "grad_norm": 0.19567685530098944,
      "kl": 0.1878814697265625,
      "learning_rate": 4.984365313468117e-07,
      "loss": 0.0002,
      "reward": 1.7928571626543999,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.8017857372760773,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8714
    },
    {
      "completion_length": 252.37054824829102,
      "epoch": 1.461377258057756,
      "grad_norm": 0.2787780610294813,
      "kl": 0.104888916015625,
      "learning_rate": 4.984351915779783e-07,
      "loss": 0.0001,
      "reward": 1.7214286401867867,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7303571738302708,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8716
    },
    {
      "completion_length": 252.12947750091553,
      "epoch": 1.4617125612976236,
      "grad_norm": 0.3540296381410723,
      "kl": 1.019378662109375,
      "learning_rate": 4.98433851237155e-07,
      "loss": 0.001,
      "reward": 1.7285714894533157,
      "reward_std": 0.03030457627028227,
      "rewards/equation_reward_func": 0.7375000361353159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8718
    },
    {
      "completion_length": 251.0625114440918,
      "epoch": 1.462047864537491,
      "grad_norm": 0.29658882871831993,
      "kl": 0.2493743896484375,
      "learning_rate": 4.984325103243448e-07,
      "loss": 0.0002,
      "reward": 1.7125000730156898,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.716964315623045,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8720
    },
    {
      "completion_length": 255.5446538925171,
      "epoch": 1.4623831677773587,
      "grad_norm": 0.15793259635685353,
      "kl": 0.204254150390625,
      "learning_rate": 4.984311688395508e-07,
      "loss": 0.0002,
      "reward": 1.789285771548748,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.7982143051922321,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8722
    },
    {
      "completion_length": 245.4553689956665,
      "epoch": 1.4627184710172263,
      "grad_norm": 0.20221390732780958,
      "kl": 0.2862548828125,
      "learning_rate": 4.984298267827763e-07,
      "loss": 0.0003,
      "reward": 1.735714353621006,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7357143256813288,
      "rewards/format_reward_func": 1.0,
      "step": 8724
    },
    {
      "completion_length": 256.0401906967163,
      "epoch": 1.4630537742570937,
      "grad_norm": 0.18251496359166494,
      "kl": 0.3800048828125,
      "learning_rate": 4.984284841540242e-07,
      "loss": 0.0004,
      "reward": 1.7053572162985802,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7098214589059353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8726
    },
    {
      "completion_length": 239.9241189956665,
      "epoch": 1.4633890774969613,
      "grad_norm": 0.16568506817127004,
      "kl": 0.2138671875,
      "learning_rate": 4.984271409532975e-07,
      "loss": 0.0002,
      "reward": 1.6928572282195091,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.6928571704775095,
      "rewards/format_reward_func": 1.0,
      "step": 8728
    },
    {
      "completion_length": 242.13840198516846,
      "epoch": 1.4637243807368288,
      "grad_norm": 0.1784867004860529,
      "kl": 0.3534393310546875,
      "learning_rate": 4.984257971805995e-07,
      "loss": 0.0004,
      "reward": 1.7678571939468384,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571827709675,
      "rewards/format_reward_func": 1.0,
      "step": 8730
    },
    {
      "completion_length": 248.75447368621826,
      "epoch": 1.4640596839766964,
      "grad_norm": 0.1337191378943348,
      "kl": 0.240386962890625,
      "learning_rate": 4.984244528359332e-07,
      "loss": 0.0002,
      "reward": 1.7339286357164383,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7383929006755352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8732
    },
    {
      "completion_length": 245.96429824829102,
      "epoch": 1.464394987216564,
      "grad_norm": 0.2954717061176378,
      "kl": 0.18841552734375,
      "learning_rate": 4.984231079193016e-07,
      "loss": 0.0002,
      "reward": 1.7285715341567993,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7285714633762836,
      "rewards/format_reward_func": 1.0,
      "step": 8734
    },
    {
      "completion_length": 241.88393783569336,
      "epoch": 1.4647302904564317,
      "grad_norm": 0.15578577592582596,
      "kl": 0.1372833251953125,
      "learning_rate": 4.98421762430708e-07,
      "loss": 0.0001,
      "reward": 1.7482143491506577,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7526786103844643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8736
    },
    {
      "completion_length": 246.76340293884277,
      "epoch": 1.465065593696299,
      "grad_norm": 0.18416701370666422,
      "kl": 0.2396087646484375,
      "learning_rate": 4.984204163701554e-07,
      "loss": 0.0002,
      "reward": 1.7892857789993286,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 8738
    },
    {
      "completion_length": 235.80358028411865,
      "epoch": 1.4654008969361667,
      "grad_norm": 0.27698587198850594,
      "kl": 0.2167816162109375,
      "learning_rate": 4.984190697376469e-07,
      "loss": 0.0002,
      "reward": 1.7446429207921028,
      "reward_std": 0.09848987311124802,
      "rewards/equation_reward_func": 0.7580357454717159,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 8740
    },
    {
      "completion_length": 236.6919755935669,
      "epoch": 1.4657362001760341,
      "grad_norm": 0.41206796353337627,
      "kl": 0.5396728515625,
      "learning_rate": 4.984177225331856e-07,
      "loss": 0.0005,
      "reward": 1.733928643167019,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7383928820490837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8742
    },
    {
      "completion_length": 242.49108505249023,
      "epoch": 1.4660715034159018,
      "grad_norm": 0.7406925620636272,
      "kl": 0.42578125,
      "learning_rate": 4.984163747567745e-07,
      "loss": 0.0004,
      "reward": 1.7964286357164383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.796428594738245,
      "rewards/format_reward_func": 1.0,
      "step": 8744
    },
    {
      "completion_length": 230.26786613464355,
      "epoch": 1.4664068066557694,
      "grad_norm": 0.15986850405720687,
      "kl": 0.114776611328125,
      "learning_rate": 4.984150264084169e-07,
      "loss": 0.0001,
      "reward": 1.798214353621006,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8017857447266579,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 8746
    },
    {
      "completion_length": 246.60715579986572,
      "epoch": 1.4667421098956368,
      "grad_norm": 0.21336797543590305,
      "kl": 0.48760986328125,
      "learning_rate": 4.984136774881158e-07,
      "loss": 0.0005,
      "reward": 1.7857143506407738,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 8748
    },
    {
      "completion_length": 232.2544755935669,
      "epoch": 1.4670774131355044,
      "grad_norm": 0.2915242898738858,
      "kl": 0.111358642578125,
      "learning_rate": 4.984123279958742e-07,
      "loss": 0.0001,
      "reward": 1.7160715013742447,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.720535745844245,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8750
    },
    {
      "completion_length": 236.7544755935669,
      "epoch": 1.4674127163753719,
      "grad_norm": 0.43406847290546435,
      "kl": 0.15557861328125,
      "learning_rate": 4.984109779316955e-07,
      "loss": 0.0002,
      "reward": 1.750000074505806,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 8752
    },
    {
      "completion_length": 233.5937623977661,
      "epoch": 1.4677480196152395,
      "grad_norm": 0.1601888555299674,
      "kl": 0.16204833984375,
      "learning_rate": 4.984096272955825e-07,
      "loss": 0.0002,
      "reward": 1.7892857640981674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857398837805,
      "rewards/format_reward_func": 1.0,
      "step": 8754
    },
    {
      "completion_length": 237.0669755935669,
      "epoch": 1.4680833228551071,
      "grad_norm": 0.1986376468777605,
      "kl": 0.101776123046875,
      "learning_rate": 4.984082760875383e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 8756
    },
    {
      "completion_length": 230.04465198516846,
      "epoch": 1.4684186260949748,
      "grad_norm": 0.11074341297077231,
      "kl": 0.114471435546875,
      "learning_rate": 4.984069243075663e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143245637417,
      "rewards/format_reward_func": 1.0,
      "step": 8758
    },
    {
      "completion_length": 236.65179824829102,
      "epoch": 1.4687539293348422,
      "grad_norm": 0.15408371005615754,
      "kl": 0.10626220703125,
      "learning_rate": 4.984055719556695e-07,
      "loss": 0.0001,
      "reward": 1.8178571984171867,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8178571704775095,
      "rewards/format_reward_func": 1.0,
      "step": 8760
    },
    {
      "completion_length": 239.63840198516846,
      "epoch": 1.4690892325747098,
      "grad_norm": 0.21080123821184735,
      "kl": 0.1241455078125,
      "learning_rate": 4.984042190318509e-07,
      "loss": 0.0001,
      "reward": 1.760714367032051,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143260538578,
      "rewards/format_reward_func": 1.0,
      "step": 8762
    },
    {
      "completion_length": 237.87054443359375,
      "epoch": 1.4694245358145772,
      "grad_norm": 0.2850582997015366,
      "kl": 0.1470184326171875,
      "learning_rate": 4.984028655361138e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 8764
    },
    {
      "completion_length": 240.77679824829102,
      "epoch": 1.4697598390544448,
      "grad_norm": 0.2321030600462851,
      "kl": 0.164703369140625,
      "learning_rate": 4.98401511468461e-07,
      "loss": 0.0002,
      "reward": 1.7357143461704254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.735714316368103,
      "rewards/format_reward_func": 1.0,
      "step": 8766
    },
    {
      "completion_length": 236.86608123779297,
      "epoch": 1.4700951422943125,
      "grad_norm": 0.3982264017600752,
      "kl": 0.1109619140625,
      "learning_rate": 4.984001568288961e-07,
      "loss": 0.0001,
      "reward": 1.7464286163449287,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 8768
    },
    {
      "completion_length": 239.758939743042,
      "epoch": 1.47043044553418,
      "grad_norm": 0.5475701173802654,
      "kl": 0.11822509765625,
      "learning_rate": 4.983988016174216e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 8770
    },
    {
      "completion_length": 238.2901906967163,
      "epoch": 1.4707657487740475,
      "grad_norm": 0.13873530815564392,
      "kl": 0.13739013671875,
      "learning_rate": 4.983974458340412e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000223517418,
      "rewards/format_reward_func": 1.0,
      "step": 8772
    },
    {
      "completion_length": 240.0044765472412,
      "epoch": 1.4711010520139152,
      "grad_norm": 0.20762954442612053,
      "kl": 0.152435302734375,
      "learning_rate": 4.983960894787577e-07,
      "loss": 0.0002,
      "reward": 1.73392865806818,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7383928913623095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8774
    },
    {
      "completion_length": 240.7009048461914,
      "epoch": 1.4714363552537826,
      "grad_norm": 0.24434572005939045,
      "kl": 0.20208740234375,
      "learning_rate": 4.983947325515743e-07,
      "loss": 0.0002,
      "reward": 1.7535714879631996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 8776
    },
    {
      "completion_length": 237.5937623977661,
      "epoch": 1.4717716584936502,
      "grad_norm": 0.14532140615552122,
      "kl": 0.213592529296875,
      "learning_rate": 4.983933750524941e-07,
      "loss": 0.0002,
      "reward": 1.7642857730388641,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 8778
    },
    {
      "completion_length": 237.67858219146729,
      "epoch": 1.4721069617335178,
      "grad_norm": 0.21694863813568013,
      "kl": 0.153778076171875,
      "learning_rate": 4.983920169815203e-07,
      "loss": 0.0002,
      "reward": 1.782142922282219,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 8780
    },
    {
      "completion_length": 237.61608505249023,
      "epoch": 1.4724422649733853,
      "grad_norm": 0.1438045442306162,
      "kl": 0.290191650390625,
      "learning_rate": 4.98390658338656e-07,
      "loss": 0.0003,
      "reward": 1.7607143819332123,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 8782
    },
    {
      "completion_length": 238.18751049041748,
      "epoch": 1.472777568213253,
      "grad_norm": 0.21501601267495266,
      "kl": 0.1873779296875,
      "learning_rate": 4.983892991239043e-07,
      "loss": 0.0002,
      "reward": 1.750000074505806,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 8784
    },
    {
      "completion_length": 237.9419755935669,
      "epoch": 1.4731128714531203,
      "grad_norm": 0.22531915179161718,
      "kl": 0.19091796875,
      "learning_rate": 4.983879393372683e-07,
      "loss": 0.0002,
      "reward": 1.8071429133415222,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.8071428909897804,
      "rewards/format_reward_func": 1.0,
      "step": 8786
    },
    {
      "completion_length": 240.008939743042,
      "epoch": 1.473448174692988,
      "grad_norm": 0.26067253479642094,
      "kl": 0.281646728515625,
      "learning_rate": 4.983865789787512e-07,
      "loss": 0.0003,
      "reward": 1.7196429371833801,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7241071835160255,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8788
    },
    {
      "completion_length": 238.67858123779297,
      "epoch": 1.4737834779328556,
      "grad_norm": 0.08043931843097116,
      "kl": 0.195220947265625,
      "learning_rate": 4.983852180483561e-07,
      "loss": 0.0002,
      "reward": 1.7964286133646965,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.805357163771987,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8790
    },
    {
      "completion_length": 239.61161708831787,
      "epoch": 1.4741187811727232,
      "grad_norm": 0.13354989199160783,
      "kl": 0.142486572265625,
      "learning_rate": 4.983838565460861e-07,
      "loss": 0.0001,
      "reward": 1.8071429282426834,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428798139095,
      "rewards/format_reward_func": 1.0,
      "step": 8792
    },
    {
      "completion_length": 224.60715293884277,
      "epoch": 1.4744540844125906,
      "grad_norm": 0.08819330980406609,
      "kl": 0.14404296875,
      "learning_rate": 4.983824944719445e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857142984867096,
      "rewards/format_reward_func": 1.0,
      "step": 8794
    },
    {
      "completion_length": 235.8928680419922,
      "epoch": 1.4747893876524583,
      "grad_norm": 0.1696130647851362,
      "kl": 0.256622314453125,
      "learning_rate": 4.983811318259341e-07,
      "loss": 0.0003,
      "reward": 1.782142922282219,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 8796
    },
    {
      "completion_length": 251.71429538726807,
      "epoch": 1.4751246908923257,
      "grad_norm": 0.24888454972454868,
      "kl": 0.257476806640625,
      "learning_rate": 4.983797686080584e-07,
      "loss": 0.0003,
      "reward": 1.74642863124609,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464285995811224,
      "rewards/format_reward_func": 1.0,
      "step": 8798
    },
    {
      "completion_length": 242.61161518096924,
      "epoch": 1.4754599941321933,
      "grad_norm": 0.24204832950998906,
      "kl": 0.138885498046875,
      "learning_rate": 4.983784048183203e-07,
      "loss": 0.0001,
      "reward": 1.7357143834233284,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 8800
    },
    {
      "completion_length": 247.0357265472412,
      "epoch": 1.475795297372061,
      "grad_norm": 0.2644117340159772,
      "kl": 0.148284912109375,
      "learning_rate": 4.98377040456723e-07,
      "loss": 0.0001,
      "reward": 1.7910714820027351,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7955357432365417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8802
    },
    {
      "completion_length": 244.20983219146729,
      "epoch": 1.4761306006119284,
      "grad_norm": 0.10373343324288181,
      "kl": 0.2305908203125,
      "learning_rate": 4.983756755232698e-07,
      "loss": 0.0002,
      "reward": 1.7196429520845413,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7241071723401546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8804
    },
    {
      "completion_length": 245.05804634094238,
      "epoch": 1.476465903851796,
      "grad_norm": 0.2235878076098335,
      "kl": 0.140228271484375,
      "learning_rate": 4.983743100179635e-07,
      "loss": 0.0001,
      "reward": 1.7553572282195091,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.759821455925703,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8806
    },
    {
      "completion_length": 237.69644165039062,
      "epoch": 1.4768012070916634,
      "grad_norm": 0.40044982762353,
      "kl": 0.15948486328125,
      "learning_rate": 4.983729439408076e-07,
      "loss": 0.0002,
      "reward": 1.84285718947649,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8428571783006191,
      "rewards/format_reward_func": 1.0,
      "step": 8808
    },
    {
      "completion_length": 237.32590675354004,
      "epoch": 1.477136510331531,
      "grad_norm": 0.3943765992765898,
      "kl": 0.201751708984375,
      "learning_rate": 4.983715772918051e-07,
      "loss": 0.0002,
      "reward": 1.7571429386734962,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 8810
    },
    {
      "completion_length": 228.9642972946167,
      "epoch": 1.4774718135713987,
      "grad_norm": 0.257684754189661,
      "kl": 0.1728515625,
      "learning_rate": 4.98370210070959e-07,
      "loss": 0.0002,
      "reward": 1.77946437895298,
      "reward_std": 0.049244935158640146,
      "rewards/equation_reward_func": 0.7812500223517418,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 8812
    },
    {
      "completion_length": 240.70983219146729,
      "epoch": 1.4778071168112663,
      "grad_norm": 0.2669589546020435,
      "kl": 0.183319091796875,
      "learning_rate": 4.983688422782727e-07,
      "loss": 0.0002,
      "reward": 1.7642857730388641,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 8814
    },
    {
      "completion_length": 237.05358219146729,
      "epoch": 1.4781424200511337,
      "grad_norm": 0.18092645813421443,
      "kl": 0.289398193359375,
      "learning_rate": 4.983674739137492e-07,
      "loss": 0.0003,
      "reward": 1.7142857909202576,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857536673546,
      "rewards/format_reward_func": 1.0,
      "step": 8816
    },
    {
      "completion_length": 242.8750114440918,
      "epoch": 1.4784777232910014,
      "grad_norm": 0.12930487883190817,
      "kl": 0.164398193359375,
      "learning_rate": 4.983661049773918e-07,
      "loss": 0.0002,
      "reward": 1.7821428999304771,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 8818
    },
    {
      "completion_length": 226.96429824829102,
      "epoch": 1.4788130265308688,
      "grad_norm": 0.4132107732816038,
      "kl": 0.1771240234375,
      "learning_rate": 4.983647354692034e-07,
      "loss": 0.0002,
      "reward": 1.81428574770689,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8142857551574707,
      "rewards/format_reward_func": 1.0,
      "step": 8820
    },
    {
      "completion_length": 243.04018783569336,
      "epoch": 1.4791483297707364,
      "grad_norm": 0.39982853624464665,
      "kl": 0.329010009765625,
      "learning_rate": 4.983633653891872e-07,
      "loss": 0.0003,
      "reward": 1.7446429207921028,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8822
    },
    {
      "completion_length": 252.696439743042,
      "epoch": 1.479483633010604,
      "grad_norm": 0.5636208493610942,
      "kl": 0.17095947265625,
      "learning_rate": 4.983619947373467e-07,
      "loss": 0.0002,
      "reward": 1.7678572461009026,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.767857164144516,
      "rewards/format_reward_func": 1.0,
      "step": 8824
    },
    {
      "completion_length": 242.7857255935669,
      "epoch": 1.4798189362504715,
      "grad_norm": 0.22129015934686244,
      "kl": 0.28814697265625,
      "learning_rate": 4.983606235136847e-07,
      "loss": 0.0003,
      "reward": 1.7571429088711739,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 8826
    },
    {
      "completion_length": 237.4687614440918,
      "epoch": 1.480154239490339,
      "grad_norm": 0.2303722888334993,
      "kl": 0.1688385009765625,
      "learning_rate": 4.983592517182044e-07,
      "loss": 0.0002,
      "reward": 1.7089286521077156,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7133928760886192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8828
    },
    {
      "completion_length": 235.96429634094238,
      "epoch": 1.4804895427302065,
      "grad_norm": 0.2179046244872161,
      "kl": 0.308685302734375,
      "learning_rate": 4.98357879350909e-07,
      "loss": 0.0003,
      "reward": 1.7357143685221672,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7446428909897804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8830
    },
    {
      "completion_length": 221.7366189956665,
      "epoch": 1.4808248459700741,
      "grad_norm": 0.12126952172765075,
      "kl": 0.15643310546875,
      "learning_rate": 4.983565064118017e-07,
      "loss": 0.0002,
      "reward": 1.7750000655651093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 8832
    },
    {
      "completion_length": 234.1428689956665,
      "epoch": 1.4811601492099418,
      "grad_norm": 0.31648037804345736,
      "kl": 0.201019287109375,
      "learning_rate": 4.983551329008858e-07,
      "loss": 0.0002,
      "reward": 1.7232143506407738,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7276786137372255,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8834
    },
    {
      "completion_length": 233.80804538726807,
      "epoch": 1.4814954524498094,
      "grad_norm": 0.22828808628135586,
      "kl": 0.19964599609375,
      "learning_rate": 4.983537588181641e-07,
      "loss": 0.0002,
      "reward": 1.678571529686451,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.678571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 8836
    },
    {
      "completion_length": 223.83036708831787,
      "epoch": 1.4818307556896768,
      "grad_norm": 0.25069049217082295,
      "kl": 0.148529052734375,
      "learning_rate": 4.9835238416364e-07,
      "loss": 0.0001,
      "reward": 1.8178571909666061,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.817857176065445,
      "rewards/format_reward_func": 1.0,
      "step": 8838
    },
    {
      "completion_length": 228.49554634094238,
      "epoch": 1.4821660589295444,
      "grad_norm": 0.18364389650314558,
      "kl": 0.13134765625,
      "learning_rate": 4.983510089373167e-07,
      "loss": 0.0001,
      "reward": 1.7410714849829674,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357424914837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8840
    },
    {
      "completion_length": 227.50447368621826,
      "epoch": 1.4825013621694119,
      "grad_norm": 0.3162348652446139,
      "kl": 0.1231842041015625,
      "learning_rate": 4.983496331391973e-07,
      "loss": 0.0001,
      "reward": 1.7250000908970833,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7250000350177288,
      "rewards/format_reward_func": 1.0,
      "step": 8842
    },
    {
      "completion_length": 225.16072463989258,
      "epoch": 1.4828366654092795,
      "grad_norm": 0.1262953317962862,
      "kl": 0.11572265625,
      "learning_rate": 4.983482567692848e-07,
      "loss": 0.0001,
      "reward": 1.8107143342494965,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 8844
    },
    {
      "completion_length": 236.4375114440918,
      "epoch": 1.4831719686491471,
      "grad_norm": 0.348430568533721,
      "kl": 0.1246185302734375,
      "learning_rate": 4.983468798275827e-07,
      "loss": 0.0001,
      "reward": 1.7142857983708382,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.714285746216774,
      "rewards/format_reward_func": 1.0,
      "step": 8846
    },
    {
      "completion_length": 222.82143688201904,
      "epoch": 1.4835072718890145,
      "grad_norm": 0.10965358167212523,
      "kl": 0.1109466552734375,
      "learning_rate": 4.98345502314094e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 8848
    },
    {
      "completion_length": 237.7366180419922,
      "epoch": 1.4838425751288822,
      "grad_norm": 0.2539963287004196,
      "kl": 0.1054534912109375,
      "learning_rate": 4.983441242288219e-07,
      "loss": 0.0001,
      "reward": 1.7464286088943481,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464285902678967,
      "rewards/format_reward_func": 1.0,
      "step": 8850
    },
    {
      "completion_length": 236.00447750091553,
      "epoch": 1.4841778783687498,
      "grad_norm": 0.2132569693177203,
      "kl": 0.104949951171875,
      "learning_rate": 4.983427455717694e-07,
      "loss": 0.0001,
      "reward": 1.7000000849366188,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7000000309199095,
      "rewards/format_reward_func": 1.0,
      "step": 8852
    },
    {
      "completion_length": 235.29019355773926,
      "epoch": 1.4845131816086172,
      "grad_norm": 0.178193772294305,
      "kl": 0.125152587890625,
      "learning_rate": 4.983413663429399e-07,
      "loss": 0.0001,
      "reward": 1.7482143715023994,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8854
    },
    {
      "completion_length": 231.33929538726807,
      "epoch": 1.4848484848484849,
      "grad_norm": 0.37231971986180934,
      "kl": 0.13079833984375,
      "learning_rate": 4.983399865423365e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 8856
    },
    {
      "completion_length": 230.4821538925171,
      "epoch": 1.4851837880883525,
      "grad_norm": 0.23070799188104596,
      "kl": 0.146514892578125,
      "learning_rate": 4.983386061699624e-07,
      "loss": 0.0001,
      "reward": 1.7732143476605415,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7776785977184772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8858
    },
    {
      "completion_length": 249.3571538925171,
      "epoch": 1.48551909132822,
      "grad_norm": 0.4236600611044629,
      "kl": 0.24822998046875,
      "learning_rate": 4.983372252258207e-07,
      "loss": 0.0002,
      "reward": 1.6915179640054703,
      "reward_std": 0.0625031883828342,
      "rewards/equation_reward_func": 0.7017857395112514,
      "rewards/format_reward_func": 0.9897321499884129,
      "step": 8860
    },
    {
      "completion_length": 248.9151906967163,
      "epoch": 1.4858543945680875,
      "grad_norm": 0.39799241693952037,
      "kl": 0.156219482421875,
      "learning_rate": 4.983358437099147e-07,
      "loss": 0.0002,
      "reward": 1.7125000730156898,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7169643193483353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8862
    },
    {
      "completion_length": 245.5714406967163,
      "epoch": 1.486189697807955,
      "grad_norm": 0.17593234268182617,
      "kl": 0.192962646484375,
      "learning_rate": 4.983344616222475e-07,
      "loss": 0.0002,
      "reward": 1.6982143595814705,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7026786133646965,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8864
    },
    {
      "completion_length": 234.52233123779297,
      "epoch": 1.4865250010478226,
      "grad_norm": 0.13123695463945143,
      "kl": 0.311370849609375,
      "learning_rate": 4.983330789628221e-07,
      "loss": 0.0003,
      "reward": 1.8107143491506577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143044471741,
      "rewards/format_reward_func": 1.0,
      "step": 8866
    },
    {
      "completion_length": 239.7678680419922,
      "epoch": 1.4868603042876902,
      "grad_norm": 0.12358910005630577,
      "kl": 0.30950927734375,
      "learning_rate": 4.983316957316421e-07,
      "loss": 0.0003,
      "reward": 1.7535714879631996,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 8868
    },
    {
      "completion_length": 232.96875953674316,
      "epoch": 1.4871956075275579,
      "grad_norm": 0.17331399232615474,
      "kl": 0.1680908203125,
      "learning_rate": 4.983303119287104e-07,
      "loss": 0.0002,
      "reward": 1.76607146859169,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7705357633531094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8870
    },
    {
      "completion_length": 232.16072368621826,
      "epoch": 1.4875309107674253,
      "grad_norm": 0.26597933576708865,
      "kl": 0.197265625,
      "learning_rate": 4.983289275540302e-07,
      "loss": 0.0002,
      "reward": 1.732142947614193,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428880095482,
      "rewards/format_reward_func": 1.0,
      "step": 8872
    },
    {
      "completion_length": 233.13840579986572,
      "epoch": 1.487866214007293,
      "grad_norm": 0.1892458028270941,
      "kl": 0.406097412109375,
      "learning_rate": 4.983275426076048e-07,
      "loss": 0.0004,
      "reward": 1.7678572088479996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 8874
    },
    {
      "completion_length": 235.5982255935669,
      "epoch": 1.4882015172471603,
      "grad_norm": 0.23906095397175636,
      "kl": 0.17352294921875,
      "learning_rate": 4.983261570894373e-07,
      "loss": 0.0002,
      "reward": 1.7660714983940125,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7705357298254967,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8876
    },
    {
      "completion_length": 230.30358123779297,
      "epoch": 1.488536820487028,
      "grad_norm": 0.2635429378102701,
      "kl": 0.173614501953125,
      "learning_rate": 4.98324770999531e-07,
      "loss": 0.0002,
      "reward": 1.7392858117818832,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7392857521772385,
      "rewards/format_reward_func": 1.0,
      "step": 8878
    },
    {
      "completion_length": 244.73215866088867,
      "epoch": 1.4888721237268956,
      "grad_norm": 0.20412065409466093,
      "kl": 0.16748046875,
      "learning_rate": 4.983233843378889e-07,
      "loss": 0.0002,
      "reward": 1.7553571984171867,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214708268642,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8880
    },
    {
      "completion_length": 231.58036994934082,
      "epoch": 1.489207426966763,
      "grad_norm": 0.39266692298473066,
      "kl": 0.111083984375,
      "learning_rate": 4.983219971045143e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571790456772,
      "rewards/format_reward_func": 1.0,
      "step": 8882
    },
    {
      "completion_length": 222.5982255935669,
      "epoch": 1.4895427302066306,
      "grad_norm": 0.15750322995007576,
      "kl": 0.106170654296875,
      "learning_rate": 4.983206092994104e-07,
      "loss": 0.0001,
      "reward": 1.841071456670761,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8455357328057289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8884
    },
    {
      "completion_length": 234.8125123977661,
      "epoch": 1.489878033446498,
      "grad_norm": 0.20708568340897027,
      "kl": 0.2044830322265625,
      "learning_rate": 4.983192209225805e-07,
      "loss": 0.0002,
      "reward": 1.7428572326898575,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571674972773,
      "rewards/format_reward_func": 1.0,
      "step": 8886
    },
    {
      "completion_length": 238.50447368621826,
      "epoch": 1.4902133366863657,
      "grad_norm": 0.3230453112977611,
      "kl": 0.141143798828125,
      "learning_rate": 4.983178319740276e-07,
      "loss": 0.0001,
      "reward": 1.7625000700354576,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8888
    },
    {
      "completion_length": 231.5401906967163,
      "epoch": 1.4905486399262333,
      "grad_norm": 0.23606147243547548,
      "kl": 0.140899658203125,
      "learning_rate": 4.983164424537549e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 8890
    },
    {
      "completion_length": 232.88840293884277,
      "epoch": 1.490883943166101,
      "grad_norm": 0.21897116798986696,
      "kl": 0.1290740966796875,
      "learning_rate": 4.983150523617658e-07,
      "loss": 0.0001,
      "reward": 1.789285771548748,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 8892
    },
    {
      "completion_length": 237.89287185668945,
      "epoch": 1.4912192464059684,
      "grad_norm": 0.21563554948529737,
      "kl": 0.155853271484375,
      "learning_rate": 4.983136616980633e-07,
      "loss": 0.0002,
      "reward": 1.730357214808464,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7348214499652386,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8894
    },
    {
      "completion_length": 246.1384048461914,
      "epoch": 1.491554549645836,
      "grad_norm": 0.303244848945181,
      "kl": 0.187408447265625,
      "learning_rate": 4.983122704626507e-07,
      "loss": 0.0002,
      "reward": 1.6964286342263222,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6964286155998707,
      "rewards/format_reward_func": 1.0,
      "step": 8896
    },
    {
      "completion_length": 242.7634048461914,
      "epoch": 1.4918898528857034,
      "grad_norm": 0.3079019603547445,
      "kl": 0.136077880859375,
      "learning_rate": 4.983108786555312e-07,
      "loss": 0.0001,
      "reward": 1.8160714730620384,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.8205357454717159,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8898
    },
    {
      "completion_length": 242.7500114440918,
      "epoch": 1.492225156125571,
      "grad_norm": 0.37232604700225147,
      "kl": 0.154022216796875,
      "learning_rate": 4.98309486276708e-07,
      "loss": 0.0002,
      "reward": 1.760714367032051,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7607143297791481,
      "rewards/format_reward_func": 1.0,
      "step": 8900
    },
    {
      "completion_length": 242.83483123779297,
      "epoch": 1.4925604593654387,
      "grad_norm": 0.1844494338084748,
      "kl": 0.201995849609375,
      "learning_rate": 4.983080933261842e-07,
      "loss": 0.0002,
      "reward": 1.7375000715255737,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7419643178582191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8902
    },
    {
      "completion_length": 236.4776906967163,
      "epoch": 1.492895762605306,
      "grad_norm": 0.25817305870319135,
      "kl": 0.271881103515625,
      "learning_rate": 4.983066998039632e-07,
      "loss": 0.0003,
      "reward": 1.7357143759727478,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7357143349945545,
      "rewards/format_reward_func": 1.0,
      "step": 8904
    },
    {
      "completion_length": 243.10715293884277,
      "epoch": 1.4932310658451737,
      "grad_norm": 0.20426186955697542,
      "kl": 0.264862060546875,
      "learning_rate": 4.983053057100482e-07,
      "loss": 0.0003,
      "reward": 1.725000075995922,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000312924385,
      "rewards/format_reward_func": 1.0,
      "step": 8906
    },
    {
      "completion_length": 247.6919755935669,
      "epoch": 1.4935663690850411,
      "grad_norm": 0.1531023894494314,
      "kl": 0.41204833984375,
      "learning_rate": 4.983039110444422e-07,
      "loss": 0.0004,
      "reward": 1.6678572371602058,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.6767857410013676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 8908
    },
    {
      "completion_length": 231.68304538726807,
      "epoch": 1.4939016723249088,
      "grad_norm": 0.2368336569619978,
      "kl": 0.259857177734375,
      "learning_rate": 4.983025158071485e-07,
      "loss": 0.0003,
      "reward": 1.775000087916851,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 8910
    },
    {
      "completion_length": 249.6250114440918,
      "epoch": 1.4942369755647764,
      "grad_norm": 0.1959684567937748,
      "kl": 0.44342041015625,
      "learning_rate": 4.983011199981704e-07,
      "loss": 0.0004,
      "reward": 1.7464286386966705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 8912
    },
    {
      "completion_length": 249.63393878936768,
      "epoch": 1.494572278804644,
      "grad_norm": 0.20096044788713058,
      "kl": 0.470428466796875,
      "learning_rate": 4.982997236175111e-07,
      "loss": 0.0005,
      "reward": 1.7053572088479996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7089286148548126,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 8914
    },
    {
      "completion_length": 250.508939743042,
      "epoch": 1.4949075820445115,
      "grad_norm": 0.37251539691184704,
      "kl": 0.377716064453125,
      "learning_rate": 4.982983266651737e-07,
      "loss": 0.0004,
      "reward": 1.7589286416769028,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928954601288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8916
    },
    {
      "completion_length": 235.14733219146729,
      "epoch": 1.495242885284379,
      "grad_norm": 0.11145203648541949,
      "kl": 0.121337890625,
      "learning_rate": 4.982969291411615e-07,
      "loss": 0.0001,
      "reward": 1.801785759627819,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.8062500171363354,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8918
    },
    {
      "completion_length": 231.8884048461914,
      "epoch": 1.4955781885242465,
      "grad_norm": 0.19749947364265377,
      "kl": 0.156341552734375,
      "learning_rate": 4.982955310454777e-07,
      "loss": 0.0002,
      "reward": 1.7714286148548126,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286223053932,
      "rewards/format_reward_func": 1.0,
      "step": 8920
    },
    {
      "completion_length": 233.74108219146729,
      "epoch": 1.4959134917641141,
      "grad_norm": 0.10809478877179468,
      "kl": 0.143829345703125,
      "learning_rate": 4.982941323781255e-07,
      "loss": 0.0001,
      "reward": 1.7571429461240768,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 8922
    },
    {
      "completion_length": 224.13393878936768,
      "epoch": 1.4962487950039818,
      "grad_norm": 0.25690059581071395,
      "kl": 0.108489990234375,
      "learning_rate": 4.982927331391083e-07,
      "loss": 0.0001,
      "reward": 1.8178571611642838,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8178571704775095,
      "rewards/format_reward_func": 1.0,
      "step": 8924
    },
    {
      "completion_length": 225.33483219146729,
      "epoch": 1.4965840982438494,
      "grad_norm": 0.21596429702799877,
      "kl": 0.1488037109375,
      "learning_rate": 4.98291333328429e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857447266579,
      "rewards/format_reward_func": 1.0,
      "step": 8926
    },
    {
      "completion_length": 225.4509038925171,
      "epoch": 1.4969194014837168,
      "grad_norm": 0.3113412682431212,
      "kl": 0.1666259765625,
      "learning_rate": 4.982899329460911e-07,
      "loss": 0.0002,
      "reward": 1.7535715028643608,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 8928
    },
    {
      "completion_length": 228.73215198516846,
      "epoch": 1.4972547047235845,
      "grad_norm": 0.10233896625503296,
      "kl": 0.1457366943359375,
      "learning_rate": 4.982885319920977e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 8930
    },
    {
      "completion_length": 236.79465293884277,
      "epoch": 1.4975900079634519,
      "grad_norm": 0.24018858070649615,
      "kl": 0.164825439453125,
      "learning_rate": 4.98287130466452e-07,
      "loss": 0.0002,
      "reward": 1.7625000700354576,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8932
    },
    {
      "completion_length": 220.77679538726807,
      "epoch": 1.4979253112033195,
      "grad_norm": 0.15697397655397416,
      "kl": 0.1427001953125,
      "learning_rate": 4.982857283691572e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714488476515,
      "rewards/format_reward_func": 1.0,
      "step": 8934
    },
    {
      "completion_length": 236.42858028411865,
      "epoch": 1.4982606144431871,
      "grad_norm": 0.12975874262251802,
      "kl": 0.12359619140625,
      "learning_rate": 4.982843257002167e-07,
      "loss": 0.0001,
      "reward": 1.742857202887535,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7428571749478579,
      "rewards/format_reward_func": 1.0,
      "step": 8936
    },
    {
      "completion_length": 226.4241180419922,
      "epoch": 1.4985959176830546,
      "grad_norm": 0.33443616062865283,
      "kl": 0.098114013671875,
      "learning_rate": 4.982829224596337e-07,
      "loss": 0.0001,
      "reward": 1.7589286416769028,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7633928805589676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8938
    },
    {
      "completion_length": 219.69197273254395,
      "epoch": 1.4989312209229222,
      "grad_norm": 0.6156013334847034,
      "kl": 0.187744140625,
      "learning_rate": 4.982815186474111e-07,
      "loss": 0.0002,
      "reward": 1.7571429312229156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 8940
    },
    {
      "completion_length": 227.27233219146729,
      "epoch": 1.4992665241627896,
      "grad_norm": 0.11772230548243262,
      "kl": 0.121368408203125,
      "learning_rate": 4.982801142635526e-07,
      "loss": 0.0001,
      "reward": 1.8107143342494965,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8107142969965935,
      "rewards/format_reward_func": 1.0,
      "step": 8942
    },
    {
      "completion_length": 223.13393783569336,
      "epoch": 1.4996018274026572,
      "grad_norm": 0.21252309938680947,
      "kl": 0.1238250732421875,
      "learning_rate": 4.982787093080612e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 8944
    },
    {
      "completion_length": 227.83036422729492,
      "epoch": 1.4999371306425249,
      "grad_norm": 0.26532050624733566,
      "kl": 0.21636962890625,
      "learning_rate": 4.982773037809402e-07,
      "loss": 0.0002,
      "reward": 1.796428643167019,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285910129547,
      "rewards/format_reward_func": 1.0,
      "step": 8946
    },
    {
      "completion_length": 218.97322273254395,
      "epoch": 1.5002724338823925,
      "grad_norm": 0.26540863359360023,
      "kl": 0.182037353515625,
      "learning_rate": 4.982758976821927e-07,
      "loss": 0.0002,
      "reward": 1.730357214808464,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7348214723169804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 8948
    },
    {
      "completion_length": 221.95983123779297,
      "epoch": 1.50060773712226,
      "grad_norm": 0.20609867460584388,
      "kl": 0.11181640625,
      "learning_rate": 4.982744910118221e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7857143059372902,
      "rewards/format_reward_func": 1.0,
      "step": 8950
    },
    {
      "completion_length": 216.87947463989258,
      "epoch": 1.5009430403621273,
      "grad_norm": 0.2493354640229781,
      "kl": 0.1196746826171875,
      "learning_rate": 4.982730837698314e-07,
      "loss": 0.0001,
      "reward": 1.7535715252161026,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714562982321,
      "rewards/format_reward_func": 1.0,
      "step": 8952
    },
    {
      "completion_length": 215.49108219146729,
      "epoch": 1.501278343601995,
      "grad_norm": 0.2773946950947669,
      "kl": 0.125640869140625,
      "learning_rate": 4.982716759562242e-07,
      "loss": 0.0001,
      "reward": 1.760714367032051,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143074274063,
      "rewards/format_reward_func": 1.0,
      "step": 8954
    },
    {
      "completion_length": 220.76340293884277,
      "epoch": 1.5016136468418626,
      "grad_norm": 0.21977779618397725,
      "kl": 0.131927490234375,
      "learning_rate": 4.982702675710034e-07,
      "loss": 0.0001,
      "reward": 1.7000000923871994,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7000000383704901,
      "rewards/format_reward_func": 1.0,
      "step": 8956
    },
    {
      "completion_length": 218.67858123779297,
      "epoch": 1.5019489500817302,
      "grad_norm": 0.23496077922526748,
      "kl": 0.100067138671875,
      "learning_rate": 4.982688586141725e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.757142897695303,
      "rewards/format_reward_func": 1.0,
      "step": 8958
    },
    {
      "completion_length": 211.27233219146729,
      "epoch": 1.5022842533215979,
      "grad_norm": 0.24280331098890967,
      "kl": 0.1251220703125,
      "learning_rate": 4.982674490857346e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 8960
    },
    {
      "completion_length": 225.0759038925171,
      "epoch": 1.5026195565614653,
      "grad_norm": 0.16089475492518973,
      "kl": 0.10797119140625,
      "learning_rate": 4.98266038985693e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 8962
    },
    {
      "completion_length": 212.95090007781982,
      "epoch": 1.5029548598013327,
      "grad_norm": 0.17764220536121444,
      "kl": 0.0988922119140625,
      "learning_rate": 4.982646283140509e-07,
      "loss": 0.0001,
      "reward": 1.8107143267989159,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143044471741,
      "rewards/format_reward_func": 1.0,
      "step": 8964
    },
    {
      "completion_length": 213.56697368621826,
      "epoch": 1.5032901630412003,
      "grad_norm": 0.17043077666424078,
      "kl": 0.0966796875,
      "learning_rate": 4.982632170708117e-07,
      "loss": 0.0001,
      "reward": 1.8607143312692642,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8607143089175224,
      "rewards/format_reward_func": 1.0,
      "step": 8966
    },
    {
      "completion_length": 221.25893592834473,
      "epoch": 1.503625466281068,
      "grad_norm": 0.11552382456613353,
      "kl": 0.10955810546875,
      "learning_rate": 4.982618052559783e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 8968
    },
    {
      "completion_length": 218.31251049041748,
      "epoch": 1.5039607695209356,
      "grad_norm": 0.24386668329689457,
      "kl": 0.12744140625,
      "learning_rate": 4.982603928695543e-07,
      "loss": 0.0001,
      "reward": 1.8357143178582191,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8357143066823483,
      "rewards/format_reward_func": 1.0,
      "step": 8970
    },
    {
      "completion_length": 210.50000858306885,
      "epoch": 1.504296072760803,
      "grad_norm": 0.24801821338106403,
      "kl": 0.089141845703125,
      "learning_rate": 4.982589799115429e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428999304771,
      "rewards/format_reward_func": 1.0,
      "step": 8972
    },
    {
      "completion_length": 212.86608123779297,
      "epoch": 1.5046313760006707,
      "grad_norm": 0.11105424263655118,
      "kl": 0.097076416015625,
      "learning_rate": 4.982575663819471e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7714285925030708,
      "rewards/format_reward_func": 1.0,
      "step": 8974
    },
    {
      "completion_length": 216.56251049041748,
      "epoch": 1.504966679240538,
      "grad_norm": 0.18365734597731484,
      "kl": 0.113494873046875,
      "learning_rate": 4.982561522807705e-07,
      "loss": 0.0001,
      "reward": 1.7750000804662704,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 8976
    },
    {
      "completion_length": 222.071439743042,
      "epoch": 1.5053019824804057,
      "grad_norm": 0.19870798357054514,
      "kl": 0.116790771484375,
      "learning_rate": 4.982547376080161e-07,
      "loss": 0.0001,
      "reward": 1.8107143566012383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8107143081724644,
      "rewards/format_reward_func": 1.0,
      "step": 8978
    },
    {
      "completion_length": 218.92411708831787,
      "epoch": 1.5056372857202733,
      "grad_norm": 0.28292647884657346,
      "kl": 0.10693359375,
      "learning_rate": 4.982533223636872e-07,
      "loss": 0.0001,
      "reward": 1.8535714820027351,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8535714522004128,
      "rewards/format_reward_func": 1.0,
      "step": 8980
    },
    {
      "completion_length": 230.99554538726807,
      "epoch": 1.505972588960141,
      "grad_norm": 0.24335191206106543,
      "kl": 0.146392822265625,
      "learning_rate": 4.982519065477873e-07,
      "loss": 0.0001,
      "reward": 1.7678571939468384,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571864962578,
      "rewards/format_reward_func": 1.0,
      "step": 8982
    },
    {
      "completion_length": 231.1205472946167,
      "epoch": 1.5063078922000084,
      "grad_norm": 0.24953749066660705,
      "kl": 0.19580078125,
      "learning_rate": 4.982504901603192e-07,
      "loss": 0.0002,
      "reward": 1.753571517765522,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 8984
    },
    {
      "completion_length": 217.58482933044434,
      "epoch": 1.5066431954398758,
      "grad_norm": 0.31805716184155647,
      "kl": 0.17828369140625,
      "learning_rate": 4.982490732012867e-07,
      "loss": 0.0002,
      "reward": 1.7821429297327995,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428775787354,
      "rewards/format_reward_func": 1.0,
      "step": 8986
    },
    {
      "completion_length": 227.37501049041748,
      "epoch": 1.5069784986797434,
      "grad_norm": 0.0028478271534510976,
      "kl": 0.110992431640625,
      "learning_rate": 4.982476556706926e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7464286126196384,
      "rewards/format_reward_func": 1.0,
      "step": 8988
    },
    {
      "completion_length": 229.70090198516846,
      "epoch": 1.507313801919611,
      "grad_norm": 0.3544715140069262,
      "kl": 0.20654296875,
      "learning_rate": 4.982462375685404e-07,
      "loss": 0.0002,
      "reward": 1.7821429297327995,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.782142873853445,
      "rewards/format_reward_func": 1.0,
      "step": 8990
    },
    {
      "completion_length": 215.92858123779297,
      "epoch": 1.5076491051594787,
      "grad_norm": 0.23203895801518692,
      "kl": 0.116851806640625,
      "learning_rate": 4.982448188948333e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 8992
    },
    {
      "completion_length": 227.571439743042,
      "epoch": 1.507984408399346,
      "grad_norm": 0.2609133394304276,
      "kl": 0.152923583984375,
      "learning_rate": 4.982433996495747e-07,
      "loss": 0.0002,
      "reward": 1.7071429267525673,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7071428913623095,
      "rewards/format_reward_func": 1.0,
      "step": 8994
    },
    {
      "completion_length": 229.30804634094238,
      "epoch": 1.5083197116392137,
      "grad_norm": 0.1709798535313269,
      "kl": 0.137298583984375,
      "learning_rate": 4.982419798327676e-07,
      "loss": 0.0001,
      "reward": 1.6964286640286446,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.6964286174625158,
      "rewards/format_reward_func": 1.0,
      "step": 8996
    },
    {
      "completion_length": 227.90626049041748,
      "epoch": 1.5086550148790812,
      "grad_norm": 0.24870973516291606,
      "kl": 0.157684326171875,
      "learning_rate": 4.982405594444155e-07,
      "loss": 0.0002,
      "reward": 1.7857143431901932,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 8998
    },
    {
      "completion_length": 222.56250762939453,
      "epoch": 1.5089903181189488,
      "grad_norm": 0.5034636051894814,
      "kl": 0.1951904296875,
      "learning_rate": 4.982391384845216e-07,
      "loss": 0.0002,
      "reward": 1.7392858117818832,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857372760773,
      "rewards/format_reward_func": 1.0,
      "step": 9000
    },
    {
      "completion_length": 227.89286613464355,
      "epoch": 1.5093256213588164,
      "grad_norm": 0.19870862664022154,
      "kl": 0.1785888671875,
      "learning_rate": 4.982377169530892e-07,
      "loss": 0.0002,
      "reward": 1.7107143551111221,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.71071432903409,
      "rewards/format_reward_func": 1.0,
      "step": 9002
    },
    {
      "completion_length": 227.68750953674316,
      "epoch": 1.509660924598684,
      "grad_norm": 0.2525564638952568,
      "kl": 0.20538330078125,
      "learning_rate": 4.982362948501214e-07,
      "loss": 0.0002,
      "reward": 1.7035715207457542,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7035714536905289,
      "rewards/format_reward_func": 1.0,
      "step": 9004
    },
    {
      "completion_length": 233.28125953674316,
      "epoch": 1.5099962278385515,
      "grad_norm": 0.23294504953174144,
      "kl": 0.242095947265625,
      "learning_rate": 4.982348721756217e-07,
      "loss": 0.0002,
      "reward": 1.726785771548748,
      "reward_std": 0.04293148126453161,
      "rewards/equation_reward_func": 0.7312500271946192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9006
    },
    {
      "completion_length": 233.90179634094238,
      "epoch": 1.5103315310784189,
      "grad_norm": 0.4282919691411335,
      "kl": 0.20703125,
      "learning_rate": 4.982334489295933e-07,
      "loss": 0.0002,
      "reward": 1.7750000655651093,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7750000394880772,
      "rewards/format_reward_func": 1.0,
      "step": 9008
    },
    {
      "completion_length": 236.61607933044434,
      "epoch": 1.5106668343182865,
      "grad_norm": 0.492780322237796,
      "kl": 0.288299560546875,
      "learning_rate": 4.982320251120395e-07,
      "loss": 0.0003,
      "reward": 1.7571429312229156,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 9010
    },
    {
      "completion_length": 242.3750114440918,
      "epoch": 1.5110021375581542,
      "grad_norm": 0.233727013707272,
      "kl": 0.2298583984375,
      "learning_rate": 4.982306007229634e-07,
      "loss": 0.0002,
      "reward": 1.7089286521077156,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7133928909897804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9012
    },
    {
      "completion_length": 234.3080472946167,
      "epoch": 1.5113374407980218,
      "grad_norm": 0.1943769671417021,
      "kl": 0.2056884765625,
      "learning_rate": 4.982291757623685e-07,
      "loss": 0.0002,
      "reward": 1.7357143610715866,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7357143275439739,
      "rewards/format_reward_func": 1.0,
      "step": 9014
    },
    {
      "completion_length": 244.28572463989258,
      "epoch": 1.5116727440378894,
      "grad_norm": 0.592650509147069,
      "kl": 0.294830322265625,
      "learning_rate": 4.98227750230258e-07,
      "loss": 0.0003,
      "reward": 1.6875000819563866,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.6919643245637417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9016
    },
    {
      "completion_length": 233.68304824829102,
      "epoch": 1.5120080472777568,
      "grad_norm": 0.19249539570632418,
      "kl": 0.17291259765625,
      "learning_rate": 4.982263241266353e-07,
      "loss": 0.0002,
      "reward": 1.8250000476837158,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8250000290572643,
      "rewards/format_reward_func": 1.0,
      "step": 9018
    },
    {
      "completion_length": 221.12947463989258,
      "epoch": 1.5123433505176243,
      "grad_norm": 0.23756867356380112,
      "kl": 0.20294189453125,
      "learning_rate": 4.982248974515033e-07,
      "loss": 0.0002,
      "reward": 1.8178571835160255,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8178571723401546,
      "rewards/format_reward_func": 1.0,
      "step": 9020
    },
    {
      "completion_length": 232.18751049041748,
      "epoch": 1.5126786537574919,
      "grad_norm": 0.16497460287344504,
      "kl": 0.15875244140625,
      "learning_rate": 4.982234702048658e-07,
      "loss": 0.0002,
      "reward": 1.7571429312229156,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 1.0,
      "step": 9022
    },
    {
      "completion_length": 223.31251049041748,
      "epoch": 1.5130139569973595,
      "grad_norm": 0.20123506275968653,
      "kl": 0.142669677734375,
      "learning_rate": 4.982220423867257e-07,
      "loss": 0.0001,
      "reward": 1.7821428999304771,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428962051868,
      "rewards/format_reward_func": 1.0,
      "step": 9024
    },
    {
      "completion_length": 242.10268878936768,
      "epoch": 1.5133492602372272,
      "grad_norm": 0.20078758437765387,
      "kl": 0.135955810546875,
      "learning_rate": 4.982206139970865e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571823984385,
      "rewards/format_reward_func": 1.0,
      "step": 9026
    },
    {
      "completion_length": 230.040189743042,
      "epoch": 1.5136845634770946,
      "grad_norm": 0.2897700898573048,
      "kl": 0.130035400390625,
      "learning_rate": 4.982191850359514e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 9028
    },
    {
      "completion_length": 245.99108123779297,
      "epoch": 1.5140198667169622,
      "grad_norm": 0.23874503537560982,
      "kl": 0.152069091796875,
      "learning_rate": 4.982177555033236e-07,
      "loss": 0.0002,
      "reward": 1.7522322237491608,
      "reward_std": 0.0473508988507092,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 9030
    },
    {
      "completion_length": 237.39733219146729,
      "epoch": 1.5143551699568296,
      "grad_norm": 0.10178434262825313,
      "kl": 0.1427001953125,
      "learning_rate": 4.982163253992066e-07,
      "loss": 0.0001,
      "reward": 1.7035715132951736,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7035714574158192,
      "rewards/format_reward_func": 1.0,
      "step": 9032
    },
    {
      "completion_length": 244.5625123977661,
      "epoch": 1.5146904731966973,
      "grad_norm": 0.23837435651216773,
      "kl": 0.157928466796875,
      "learning_rate": 4.982148947236036e-07,
      "loss": 0.0002,
      "reward": 1.7464286237955093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 9034
    },
    {
      "completion_length": 237.78126335144043,
      "epoch": 1.5150257764365649,
      "grad_norm": 0.41831335496057137,
      "kl": 0.15301513671875,
      "learning_rate": 4.982134634765178e-07,
      "loss": 0.0002,
      "reward": 1.7714286223053932,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 9036
    },
    {
      "completion_length": 243.93304824829102,
      "epoch": 1.5153610796764325,
      "grad_norm": 0.19669736268949578,
      "kl": 0.146759033203125,
      "learning_rate": 4.982120316579527e-07,
      "loss": 0.0001,
      "reward": 1.6750001087784767,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.6750000193715096,
      "rewards/format_reward_func": 1.0,
      "step": 9038
    },
    {
      "completion_length": 249.227689743042,
      "epoch": 1.5156963829163,
      "grad_norm": 0.11420041596317355,
      "kl": 0.165985107421875,
      "learning_rate": 4.982105992679113e-07,
      "loss": 0.0002,
      "reward": 1.7678571939468384,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7678571827709675,
      "rewards/format_reward_func": 1.0,
      "step": 9040
    },
    {
      "completion_length": 253.27233219146729,
      "epoch": 1.5160316861561673,
      "grad_norm": 0.5580972834206843,
      "kl": 0.196990966796875,
      "learning_rate": 4.982091663063972e-07,
      "loss": 0.0002,
      "reward": 1.7785715013742447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714454948902,
      "rewards/format_reward_func": 1.0,
      "step": 9042
    },
    {
      "completion_length": 249.9330472946167,
      "epoch": 1.516366989396035,
      "grad_norm": 0.3001462222001796,
      "kl": 0.154052734375,
      "learning_rate": 4.982077327734135e-07,
      "loss": 0.0002,
      "reward": 1.7803572192788124,
      "reward_std": 0.09848987124860287,
      "rewards/equation_reward_func": 0.7848214581608772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9044
    },
    {
      "completion_length": 249.7946548461914,
      "epoch": 1.5167022926359026,
      "grad_norm": 0.2560114961097288,
      "kl": 0.13873291015625,
      "learning_rate": 4.982062986689637e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 9046
    },
    {
      "completion_length": 249.946439743042,
      "epoch": 1.5170375958757703,
      "grad_norm": 0.22287832163678353,
      "kl": 0.1337890625,
      "learning_rate": 4.982048639930509e-07,
      "loss": 0.0001,
      "reward": 1.7285714969038963,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714484751225,
      "rewards/format_reward_func": 1.0,
      "step": 9048
    },
    {
      "completion_length": 241.42858409881592,
      "epoch": 1.5173728991156377,
      "grad_norm": 0.17462550405638447,
      "kl": 0.124725341796875,
      "learning_rate": 4.982034287456784e-07,
      "loss": 0.0001,
      "reward": 1.725000075995922,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7339286059141159,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9050
    },
    {
      "completion_length": 241.64287090301514,
      "epoch": 1.5177082023555053,
      "grad_norm": 0.31667186359080757,
      "kl": 0.12017822265625,
      "learning_rate": 4.982019929268496e-07,
      "loss": 0.0001,
      "reward": 1.792857214808464,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 9052
    },
    {
      "completion_length": 244.20983219146729,
      "epoch": 1.5180435055953727,
      "grad_norm": 0.7155592391967859,
      "kl": 0.233062744140625,
      "learning_rate": 4.982005565365678e-07,
      "loss": 0.0002,
      "reward": 1.7428571954369545,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7428571693599224,
      "rewards/format_reward_func": 1.0,
      "step": 9054
    },
    {
      "completion_length": 234.5044755935669,
      "epoch": 1.5183788088352403,
      "grad_norm": 0.19988312655943996,
      "kl": 0.1508026123046875,
      "learning_rate": 4.981991195748363e-07,
      "loss": 0.0002,
      "reward": 1.741071492433548,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7455357573926449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9056
    },
    {
      "completion_length": 238.2053680419922,
      "epoch": 1.518714112075108,
      "grad_norm": 0.2720902850609489,
      "kl": 0.113433837890625,
      "learning_rate": 4.981976820416584e-07,
      "loss": 0.0001,
      "reward": 1.7285715192556381,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714596509933,
      "rewards/format_reward_func": 1.0,
      "step": 9058
    },
    {
      "completion_length": 232.69197750091553,
      "epoch": 1.5190494153149756,
      "grad_norm": 0.20592691107488253,
      "kl": 0.165008544921875,
      "learning_rate": 4.981962439370374e-07,
      "loss": 0.0002,
      "reward": 1.8250000476837158,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8250000327825546,
      "rewards/format_reward_func": 1.0,
      "step": 9060
    },
    {
      "completion_length": 236.633939743042,
      "epoch": 1.519384718554843,
      "grad_norm": 0.31761312113051826,
      "kl": 0.1402587890625,
      "learning_rate": 4.981948052609767e-07,
      "loss": 0.0001,
      "reward": 1.8000000417232513,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8000000342726707,
      "rewards/format_reward_func": 1.0,
      "step": 9062
    },
    {
      "completion_length": 235.4241189956665,
      "epoch": 1.5197200217947104,
      "grad_norm": 0.1862511774783427,
      "kl": 0.136566162109375,
      "learning_rate": 4.981933660134795e-07,
      "loss": 0.0001,
      "reward": 1.7517857924103737,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7562500238418579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9064
    },
    {
      "completion_length": 234.3928689956665,
      "epoch": 1.520055325034578,
      "grad_norm": 0.22103160635442895,
      "kl": 0.165435791015625,
      "learning_rate": 4.981919261945491e-07,
      "loss": 0.0002,
      "reward": 1.7767857685685158,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7812500204890966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9066
    },
    {
      "completion_length": 234.696439743042,
      "epoch": 1.5203906282744457,
      "grad_norm": 0.5237816922040125,
      "kl": 0.127655029296875,
      "learning_rate": 4.981904858041889e-07,
      "loss": 0.0001,
      "reward": 1.6714286729693413,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.6714286096394062,
      "rewards/format_reward_func": 1.0,
      "step": 9068
    },
    {
      "completion_length": 233.1116180419922,
      "epoch": 1.5207259315143133,
      "grad_norm": 0.2222210056755971,
      "kl": 0.108612060546875,
      "learning_rate": 4.981890448424021e-07,
      "loss": 0.0001,
      "reward": 1.7642857655882835,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 9070
    },
    {
      "completion_length": 227.2232265472412,
      "epoch": 1.5210612347541808,
      "grad_norm": 0.3268976894652414,
      "kl": 0.09259033203125,
      "learning_rate": 4.981876033091922e-07,
      "loss": 0.0001,
      "reward": 1.766071505844593,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7696428820490837,
      "rewards/format_reward_func": 0.9964285790920258,
      "step": 9072
    },
    {
      "completion_length": 221.62947368621826,
      "epoch": 1.5213965379940484,
      "grad_norm": 0.11221103222961556,
      "kl": 0.10101318359375,
      "learning_rate": 4.981861612045624e-07,
      "loss": 0.0001,
      "reward": 1.8357143253087997,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8357143104076385,
      "rewards/format_reward_func": 1.0,
      "step": 9074
    },
    {
      "completion_length": 229.33483123779297,
      "epoch": 1.5217318412339158,
      "grad_norm": 0.16956317672941165,
      "kl": 0.0871429443359375,
      "learning_rate": 4.98184718528516e-07,
      "loss": 0.0001,
      "reward": 1.8178572058677673,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8178571723401546,
      "rewards/format_reward_func": 1.0,
      "step": 9076
    },
    {
      "completion_length": 228.6741180419922,
      "epoch": 1.5220671444737834,
      "grad_norm": 0.11080000675820717,
      "kl": 0.113494873046875,
      "learning_rate": 4.981832752810564e-07,
      "loss": 0.0001,
      "reward": 1.7642857506871223,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7642857357859612,
      "rewards/format_reward_func": 1.0,
      "step": 9078
    },
    {
      "completion_length": 231.3660831451416,
      "epoch": 1.522402447713651,
      "grad_norm": 0.074742168877599,
      "kl": 0.09735107421875,
      "learning_rate": 4.981818314621868e-07,
      "loss": 0.0001,
      "reward": 1.787500075995922,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7919643148779869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9080
    },
    {
      "completion_length": 230.1785831451416,
      "epoch": 1.5227377509535187,
      "grad_norm": 0.2248487524696226,
      "kl": 0.111541748046875,
      "learning_rate": 4.981803870719107e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7821428775787354,
      "rewards/format_reward_func": 1.0,
      "step": 9082
    },
    {
      "completion_length": 240.22322273254395,
      "epoch": 1.5230730541933861,
      "grad_norm": 0.2183842155755358,
      "kl": 0.0999603271484375,
      "learning_rate": 4.981789421102313e-07,
      "loss": 0.0001,
      "reward": 1.8250000327825546,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8250000216066837,
      "rewards/format_reward_func": 1.0,
      "step": 9084
    },
    {
      "completion_length": 231.1384048461914,
      "epoch": 1.5234083574332535,
      "grad_norm": 0.20518936982415142,
      "kl": 0.0991363525390625,
      "learning_rate": 4.98177496577152e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714767873287,
      "rewards/format_reward_func": 1.0,
      "step": 9086
    },
    {
      "completion_length": 229.571439743042,
      "epoch": 1.5237436606731212,
      "grad_norm": 0.255223191177945,
      "kl": 0.094573974609375,
      "learning_rate": 4.981760504726759e-07,
      "loss": 0.0001,
      "reward": 1.8250000476837158,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8250000402331352,
      "rewards/format_reward_func": 1.0,
      "step": 9088
    },
    {
      "completion_length": 244.84822463989258,
      "epoch": 1.5240789639129888,
      "grad_norm": 0.24607461370798528,
      "kl": 0.1124420166015625,
      "learning_rate": 4.981746037968068e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.767857164144516,
      "rewards/format_reward_func": 1.0,
      "step": 9090
    },
    {
      "completion_length": 242.04911708831787,
      "epoch": 1.5244142671528564,
      "grad_norm": 0.09837423239495768,
      "kl": 0.1160888671875,
      "learning_rate": 4.981731565495475e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 9092
    },
    {
      "completion_length": 240.37947463989258,
      "epoch": 1.524749570392724,
      "grad_norm": 0.19651923643307115,
      "kl": 0.097412109375,
      "learning_rate": 4.981717087309018e-07,
      "loss": 0.0001,
      "reward": 1.79642865806818,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 9094
    },
    {
      "completion_length": 237.21875858306885,
      "epoch": 1.5250848736325915,
      "grad_norm": 0.14018062890921712,
      "kl": 0.09429931640625,
      "learning_rate": 4.981702603408726e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 9096
    },
    {
      "completion_length": 230.48215198516846,
      "epoch": 1.525420176872459,
      "grad_norm": 0.21898535817201434,
      "kl": 0.1555023193359375,
      "learning_rate": 4.981688113794636e-07,
      "loss": 0.0002,
      "reward": 1.803571492433548,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8035714477300644,
      "rewards/format_reward_func": 1.0,
      "step": 9098
    },
    {
      "completion_length": 244.7991180419922,
      "epoch": 1.5257554801123265,
      "grad_norm": 0.16935366185685477,
      "kl": 0.0987396240234375,
      "learning_rate": 4.981673618466779e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 9100
    },
    {
      "completion_length": 241.27233123779297,
      "epoch": 1.5260907833521942,
      "grad_norm": 0.33333039079707383,
      "kl": 0.0966796875,
      "learning_rate": 4.981659117425189e-07,
      "loss": 0.0001,
      "reward": 1.6571429520845413,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.6571428943425417,
      "rewards/format_reward_func": 1.0,
      "step": 9102
    },
    {
      "completion_length": 235.34376049041748,
      "epoch": 1.5264260865920618,
      "grad_norm": 0.2157871403374037,
      "kl": 0.09100341796875,
      "learning_rate": 4.9816446106699e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 9104
    },
    {
      "completion_length": 235.5714406967163,
      "epoch": 1.5267613898319292,
      "grad_norm": 0.21741295384543297,
      "kl": 0.0951690673828125,
      "learning_rate": 4.981630098200946e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.760714303702116,
      "rewards/format_reward_func": 1.0,
      "step": 9106
    },
    {
      "completion_length": 239.915189743042,
      "epoch": 1.5270966930717969,
      "grad_norm": 0.3244411047147427,
      "kl": 0.1490478515625,
      "learning_rate": 4.981615580018357e-07,
      "loss": 0.0001,
      "reward": 1.7910714820027351,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7955357395112514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9108
    },
    {
      "completion_length": 240.0937614440918,
      "epoch": 1.5274319963116643,
      "grad_norm": 0.17487077329756923,
      "kl": 0.125091552734375,
      "learning_rate": 4.98160105612217e-07,
      "loss": 0.0001,
      "reward": 1.8000000342726707,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000342726707,
      "rewards/format_reward_func": 1.0,
      "step": 9110
    },
    {
      "completion_length": 251.28125953674316,
      "epoch": 1.527767299551532,
      "grad_norm": 0.26637816520531216,
      "kl": 0.140777587890625,
      "learning_rate": 4.981586526512418e-07,
      "loss": 0.0001,
      "reward": 1.8000000417232513,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.800000037997961,
      "rewards/format_reward_func": 1.0,
      "step": 9112
    },
    {
      "completion_length": 251.94197463989258,
      "epoch": 1.5281026027913995,
      "grad_norm": 0.20245928613813574,
      "kl": 0.1270904541015625,
      "learning_rate": 4.981571991189133e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 9114
    },
    {
      "completion_length": 242.29018878936768,
      "epoch": 1.5284379060312672,
      "grad_norm": 0.17394760960328057,
      "kl": 0.153533935546875,
      "learning_rate": 4.98155745015235e-07,
      "loss": 0.0002,
      "reward": 1.8214286118745804,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8214285969734192,
      "rewards/format_reward_func": 1.0,
      "step": 9116
    },
    {
      "completion_length": 247.1160831451416,
      "epoch": 1.5287732092711346,
      "grad_norm": 0.2279073006313851,
      "kl": 0.10235595703125,
      "learning_rate": 4.9815429034021e-07,
      "loss": 0.0001,
      "reward": 1.741071492433548,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357536673546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9118
    },
    {
      "completion_length": 242.0178689956665,
      "epoch": 1.529108512511002,
      "grad_norm": 0.2639212009108935,
      "kl": 0.1070709228515625,
      "learning_rate": 4.981528350938419e-07,
      "loss": 0.0001,
      "reward": 1.7017857879400253,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7062500435858965,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9120
    },
    {
      "completion_length": 241.91965293884277,
      "epoch": 1.5294438157508696,
      "grad_norm": 0.22109535656033716,
      "kl": 0.105133056640625,
      "learning_rate": 4.98151379276134e-07,
      "loss": 0.0001,
      "reward": 1.796428605914116,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286096394062,
      "rewards/format_reward_func": 1.0,
      "step": 9122
    },
    {
      "completion_length": 249.31697463989258,
      "epoch": 1.5297791189907373,
      "grad_norm": 0.15664011211006298,
      "kl": 0.098358154296875,
      "learning_rate": 4.981499228870895e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.785714304074645,
      "rewards/format_reward_func": 1.0,
      "step": 9124
    },
    {
      "completion_length": 242.37054634094238,
      "epoch": 1.530114422230605,
      "grad_norm": 0.18290893825178986,
      "kl": 0.1169281005859375,
      "learning_rate": 4.981484659267121e-07,
      "loss": 0.0001,
      "reward": 1.775000087916851,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 9126
    },
    {
      "completion_length": 251.2991189956665,
      "epoch": 1.5304497254704723,
      "grad_norm": 0.25195483521072903,
      "kl": 0.1036376953125,
      "learning_rate": 4.981470083950047e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571674972773,
      "rewards/format_reward_func": 1.0,
      "step": 9128
    },
    {
      "completion_length": 239.30358219146729,
      "epoch": 1.53078502871034,
      "grad_norm": 0.12518929441684795,
      "kl": 0.095672607421875,
      "learning_rate": 4.98145550291971e-07,
      "loss": 0.0001,
      "reward": 1.7642857432365417,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857600003481,
      "rewards/format_reward_func": 1.0,
      "step": 9130
    },
    {
      "completion_length": 237.91072463989258,
      "epoch": 1.5311203319502074,
      "grad_norm": 0.20083257203715613,
      "kl": 0.092926025390625,
      "learning_rate": 4.981440916176142e-07,
      "loss": 0.0001,
      "reward": 1.7250000983476639,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7250000238418579,
      "rewards/format_reward_func": 1.0,
      "step": 9132
    },
    {
      "completion_length": 247.0937614440918,
      "epoch": 1.531455635190075,
      "grad_norm": 0.22650875766293518,
      "kl": 0.103668212890625,
      "learning_rate": 4.981426323719377e-07,
      "loss": 0.0001,
      "reward": 1.728571504354477,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7285714708268642,
      "rewards/format_reward_func": 1.0,
      "step": 9134
    },
    {
      "completion_length": 243.86162090301514,
      "epoch": 1.5317909384299426,
      "grad_norm": 0.08014615946695847,
      "kl": 0.09442138671875,
      "learning_rate": 4.981411725549449e-07,
      "loss": 0.0001,
      "reward": 1.710714377462864,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7107143178582191,
      "rewards/format_reward_func": 1.0,
      "step": 9136
    },
    {
      "completion_length": 237.8616180419922,
      "epoch": 1.5321262416698103,
      "grad_norm": 0.20283591817892682,
      "kl": 0.085479736328125,
      "learning_rate": 4.98139712166639e-07,
      "loss": 0.0001,
      "reward": 1.7803572043776512,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9138
    },
    {
      "completion_length": 243.31250953674316,
      "epoch": 1.5324615449096777,
      "grad_norm": 0.2042566525715254,
      "kl": 0.0798492431640625,
      "learning_rate": 4.981382512070235e-07,
      "loss": 0.0001,
      "reward": 1.7464286461472511,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 9140
    },
    {
      "completion_length": 237.3437614440918,
      "epoch": 1.532796848149545,
      "grad_norm": 0.2004883026580934,
      "kl": 0.088653564453125,
      "learning_rate": 4.981367896761019e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 9142
    },
    {
      "completion_length": 238.80358219146729,
      "epoch": 1.5331321513894127,
      "grad_norm": 0.19496765967296867,
      "kl": 0.090240478515625,
      "learning_rate": 4.981353275738772e-07,
      "loss": 0.0001,
      "reward": 1.7285714969038963,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7285714447498322,
      "rewards/format_reward_func": 1.0,
      "step": 9144
    },
    {
      "completion_length": 231.03125953674316,
      "epoch": 1.5334674546292804,
      "grad_norm": 0.17878598708437798,
      "kl": 0.0771331787109375,
      "learning_rate": 4.981338649003531e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 9146
    },
    {
      "completion_length": 241.41965579986572,
      "epoch": 1.533802757869148,
      "grad_norm": 0.1694416729522096,
      "kl": 0.092437744140625,
      "learning_rate": 4.981324016555328e-07,
      "loss": 0.0001,
      "reward": 1.7142857909202576,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7142857573926449,
      "rewards/format_reward_func": 1.0,
      "step": 9148
    },
    {
      "completion_length": 241.62947368621826,
      "epoch": 1.5341380611090156,
      "grad_norm": 0.3325622452794708,
      "kl": 0.0935211181640625,
      "learning_rate": 4.981309378394197e-07,
      "loss": 0.0001,
      "reward": 1.6928572282195091,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.6928571686148643,
      "rewards/format_reward_func": 1.0,
      "step": 9150
    },
    {
      "completion_length": 233.10268878936768,
      "epoch": 1.534473364348883,
      "grad_norm": 0.23292539109008156,
      "kl": 0.086761474609375,
      "learning_rate": 4.981294734520172e-07,
      "loss": 0.0001,
      "reward": 1.767857238650322,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 9152
    },
    {
      "completion_length": 237.7544765472412,
      "epoch": 1.5348086675887505,
      "grad_norm": 0.20881513610441563,
      "kl": 0.0846099853515625,
      "learning_rate": 4.981280084933287e-07,
      "loss": 0.0001,
      "reward": 1.78035718947649,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214730620384,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9154
    },
    {
      "completion_length": 233.49554824829102,
      "epoch": 1.535143970828618,
      "grad_norm": 0.2662071000635271,
      "kl": 0.089324951171875,
      "learning_rate": 4.981265429633575e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 9156
    },
    {
      "completion_length": 238.50001049041748,
      "epoch": 1.5354792740684857,
      "grad_norm": 0.22534214293420757,
      "kl": 0.102813720703125,
      "learning_rate": 4.98125076862107e-07,
      "loss": 0.0001,
      "reward": 1.7357143461704254,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 9158
    },
    {
      "completion_length": 226.07590293884277,
      "epoch": 1.5358145773083534,
      "grad_norm": 0.2740503946863582,
      "kl": 0.1274871826171875,
      "learning_rate": 4.981236101895806e-07,
      "loss": 0.0001,
      "reward": 1.7607143446803093,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143260538578,
      "rewards/format_reward_func": 1.0,
      "step": 9160
    },
    {
      "completion_length": 219.53572463989258,
      "epoch": 1.5361498805482208,
      "grad_norm": 0.206061237532187,
      "kl": 0.0881195068359375,
      "learning_rate": 4.981221429457815e-07,
      "loss": 0.0001,
      "reward": 1.753571517765522,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714488476515,
      "rewards/format_reward_func": 1.0,
      "step": 9162
    },
    {
      "completion_length": 222.23661708831787,
      "epoch": 1.5364851837880884,
      "grad_norm": 0.1439280155706265,
      "kl": 0.093841552734375,
      "learning_rate": 4.981206751307135e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 9164
    },
    {
      "completion_length": 228.29018878936768,
      "epoch": 1.5368204870279558,
      "grad_norm": 0.24340325547414882,
      "kl": 0.092681884765625,
      "learning_rate": 4.981192067443795e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428991854191,
      "rewards/format_reward_func": 1.0,
      "step": 9166
    },
    {
      "completion_length": 223.34376049041748,
      "epoch": 1.5371557902678235,
      "grad_norm": 0.18917360158958488,
      "kl": 0.10345458984375,
      "learning_rate": 4.981177377867831e-07,
      "loss": 0.0001,
      "reward": 1.7892857566475868,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857361584902,
      "rewards/format_reward_func": 1.0,
      "step": 9168
    },
    {
      "completion_length": 238.33483600616455,
      "epoch": 1.537491093507691,
      "grad_norm": 0.273360201970595,
      "kl": 0.10321044921875,
      "learning_rate": 4.981162682579278e-07,
      "loss": 0.0001,
      "reward": 1.7071429267525673,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7071428801864386,
      "rewards/format_reward_func": 1.0,
      "step": 9170
    },
    {
      "completion_length": 226.3482255935669,
      "epoch": 1.5378263967475587,
      "grad_norm": 0.32470792042468705,
      "kl": 0.098785400390625,
      "learning_rate": 4.981147981578168e-07,
      "loss": 0.0001,
      "reward": 1.8178571984171867,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.8178571574389935,
      "rewards/format_reward_func": 1.0,
      "step": 9172
    },
    {
      "completion_length": 234.78572273254395,
      "epoch": 1.5381616999874261,
      "grad_norm": 0.2515031157446506,
      "kl": 0.1017303466796875,
      "learning_rate": 4.981133274864535e-07,
      "loss": 0.0001,
      "reward": 1.7665178999304771,
      "reward_std": 0.027147849323228,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 9174
    },
    {
      "completion_length": 229.47768688201904,
      "epoch": 1.5384970032272935,
      "grad_norm": 0.16447082251076406,
      "kl": 0.092315673828125,
      "learning_rate": 4.981118562438414e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 9176
    },
    {
      "completion_length": 234.4107255935669,
      "epoch": 1.5388323064671612,
      "grad_norm": 0.21182922552358208,
      "kl": 0.107757568359375,
      "learning_rate": 4.981103844299837e-07,
      "loss": 0.0001,
      "reward": 1.7750000804662704,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 9178
    },
    {
      "completion_length": 232.08483123779297,
      "epoch": 1.5391676097070288,
      "grad_norm": 0.1271282836822565,
      "kl": 0.09710693359375,
      "learning_rate": 4.981089120448839e-07,
      "loss": 0.0001,
      "reward": 1.8000000268220901,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.8000000342726707,
      "rewards/format_reward_func": 1.0,
      "step": 9180
    },
    {
      "completion_length": 251.49554920196533,
      "epoch": 1.5395029129468965,
      "grad_norm": 0.08852330164511354,
      "kl": 0.103973388671875,
      "learning_rate": 4.981074390885455e-07,
      "loss": 0.0001,
      "reward": 1.7607143372297287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143223285675,
      "rewards/format_reward_func": 1.0,
      "step": 9182
    },
    {
      "completion_length": 247.26787090301514,
      "epoch": 1.5398382161867639,
      "grad_norm": 0.18896351681271314,
      "kl": 0.09765625,
      "learning_rate": 4.981059655609717e-07,
      "loss": 0.0001,
      "reward": 1.7857143580913544,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 9184
    },
    {
      "completion_length": 245.33036708831787,
      "epoch": 1.5401735194266315,
      "grad_norm": 0.16901570124704474,
      "kl": 0.0955810546875,
      "learning_rate": 4.98104491462166e-07,
      "loss": 0.0001,
      "reward": 1.8035714626312256,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 9186
    },
    {
      "completion_length": 244.08482933044434,
      "epoch": 1.540508822666499,
      "grad_norm": 0.21763164806003021,
      "kl": 0.0872344970703125,
      "learning_rate": 4.981030167921317e-07,
      "loss": 0.0001,
      "reward": 1.7553572058677673,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214671015739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9188
    },
    {
      "completion_length": 244.71876049041748,
      "epoch": 1.5408441259063665,
      "grad_norm": 0.26243133709106686,
      "kl": 0.0949249267578125,
      "learning_rate": 4.981015415508725e-07,
      "loss": 0.0001,
      "reward": 1.7107143625617027,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7107143141329288,
      "rewards/format_reward_func": 1.0,
      "step": 9190
    },
    {
      "completion_length": 242.61608028411865,
      "epoch": 1.5411794291462342,
      "grad_norm": 0.17520268835161978,
      "kl": 0.10418701171875,
      "learning_rate": 4.981000657383914e-07,
      "loss": 0.0001,
      "reward": 1.8000000268220901,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000417232513,
      "rewards/format_reward_func": 1.0,
      "step": 9192
    },
    {
      "completion_length": 239.6473331451416,
      "epoch": 1.5415147323861018,
      "grad_norm": 0.20087686035129643,
      "kl": 0.109222412109375,
      "learning_rate": 4.980985893546919e-07,
      "loss": 0.0001,
      "reward": 1.7428572475910187,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571730852127,
      "rewards/format_reward_func": 1.0,
      "step": 9194
    },
    {
      "completion_length": 244.56697463989258,
      "epoch": 1.5418500356259692,
      "grad_norm": 0.20778828417353182,
      "kl": 0.10235595703125,
      "learning_rate": 4.980971123997776e-07,
      "loss": 0.0001,
      "reward": 1.7303572297096252,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7348214574158192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9196
    },
    {
      "completion_length": 241.5357265472412,
      "epoch": 1.5421853388658366,
      "grad_norm": 0.28343013163751585,
      "kl": 0.094146728515625,
      "learning_rate": 4.980956348736516e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7875000275671482,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9198
    },
    {
      "completion_length": 236.90179920196533,
      "epoch": 1.5425206421057043,
      "grad_norm": 0.19766081655551698,
      "kl": 0.102569580078125,
      "learning_rate": 4.980941567763176e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 9200
    },
    {
      "completion_length": 236.74554538726807,
      "epoch": 1.542855945345572,
      "grad_norm": 0.22589139544291423,
      "kl": 0.1072540283203125,
      "learning_rate": 4.980926781077788e-07,
      "loss": 0.0001,
      "reward": 1.735714353621006,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143182307482,
      "rewards/format_reward_func": 1.0,
      "step": 9202
    },
    {
      "completion_length": 238.62500858306885,
      "epoch": 1.5431912485854395,
      "grad_norm": 0.2167072922018885,
      "kl": 0.0994873046875,
      "learning_rate": 4.980911988680386e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 9204
    },
    {
      "completion_length": 237.29018783569336,
      "epoch": 1.543526551825307,
      "grad_norm": 0.23633367374109066,
      "kl": 0.095428466796875,
      "learning_rate": 4.980897190571006e-07,
      "loss": 0.0001,
      "reward": 1.7321429401636124,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321428880095482,
      "rewards/format_reward_func": 1.0,
      "step": 9206
    },
    {
      "completion_length": 243.02233219146729,
      "epoch": 1.5438618550651746,
      "grad_norm": 0.2340838662723435,
      "kl": 0.11846923828125,
      "learning_rate": 4.980882386749681e-07,
      "loss": 0.0001,
      "reward": 1.7339286655187607,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7383928894996643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9208
    },
    {
      "completion_length": 236.89733219146729,
      "epoch": 1.544197158305042,
      "grad_norm": 0.2706127020814316,
      "kl": 0.097137451171875,
      "learning_rate": 4.980867577216444e-07,
      "loss": 0.0001,
      "reward": 1.7214286550879478,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7214285880327225,
      "rewards/format_reward_func": 1.0,
      "step": 9210
    },
    {
      "completion_length": 233.38393878936768,
      "epoch": 1.5445324615449096,
      "grad_norm": 0.20047158656273134,
      "kl": 0.105621337890625,
      "learning_rate": 4.98085276197133e-07,
      "loss": 0.0001,
      "reward": 1.707142911851406,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7071428932249546,
      "rewards/format_reward_func": 1.0,
      "step": 9212
    },
    {
      "completion_length": 244.8303680419922,
      "epoch": 1.5448677647847773,
      "grad_norm": 0.134787846932301,
      "kl": 0.119598388671875,
      "learning_rate": 4.980837941014374e-07,
      "loss": 0.0001,
      "reward": 1.6964286714792252,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6964286044239998,
      "rewards/format_reward_func": 1.0,
      "step": 9214
    },
    {
      "completion_length": 238.9776906967163,
      "epoch": 1.545203068024645,
      "grad_norm": 0.2149803153413945,
      "kl": 0.110198974609375,
      "learning_rate": 4.980823114345608e-07,
      "loss": 0.0001,
      "reward": 1.739285796880722,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857447266579,
      "rewards/format_reward_func": 1.0,
      "step": 9216
    },
    {
      "completion_length": 224.63393783569336,
      "epoch": 1.5455383712645123,
      "grad_norm": 0.24007473653044104,
      "kl": 0.105682373046875,
      "learning_rate": 4.980808281965068e-07,
      "loss": 0.0001,
      "reward": 1.7357143759727478,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143275439739,
      "rewards/format_reward_func": 1.0,
      "step": 9218
    },
    {
      "completion_length": 235.7634038925171,
      "epoch": 1.5458736745043797,
      "grad_norm": 0.1669212593409537,
      "kl": 0.110992431640625,
      "learning_rate": 4.980793443872788e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428842842579,
      "rewards/format_reward_func": 1.0,
      "step": 9220
    },
    {
      "completion_length": 226.36161613464355,
      "epoch": 1.5462089777442474,
      "grad_norm": 0.16524317745228337,
      "kl": 0.111541748046875,
      "learning_rate": 4.980778600068801e-07,
      "loss": 0.0001,
      "reward": 1.7214286401867867,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7214286103844643,
      "rewards/format_reward_func": 1.0,
      "step": 9222
    },
    {
      "completion_length": 219.8169765472412,
      "epoch": 1.546544280984115,
      "grad_norm": 0.2477187774885109,
      "kl": 0.091156005859375,
      "learning_rate": 4.980763750553142e-07,
      "loss": 0.0001,
      "reward": 1.8071429207921028,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8071428835391998,
      "rewards/format_reward_func": 1.0,
      "step": 9224
    },
    {
      "completion_length": 235.0000114440918,
      "epoch": 1.5468795842239826,
      "grad_norm": 0.11273552378741418,
      "kl": 0.117584228515625,
      "learning_rate": 4.980748895325845e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 9226
    },
    {
      "completion_length": 224.48215293884277,
      "epoch": 1.5472148874638503,
      "grad_norm": 0.2502290193023174,
      "kl": 0.1129608154296875,
      "learning_rate": 4.980734034386944e-07,
      "loss": 0.0001,
      "reward": 1.7964286133646965,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.796428594738245,
      "rewards/format_reward_func": 1.0,
      "step": 9228
    },
    {
      "completion_length": 222.09375953674316,
      "epoch": 1.5475501907037177,
      "grad_norm": 0.1613240611923338,
      "kl": 0.113433837890625,
      "learning_rate": 4.980719167736474e-07,
      "loss": 0.0001,
      "reward": 1.7839286476373672,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.788392897695303,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9230
    },
    {
      "completion_length": 226.97768688201904,
      "epoch": 1.547885493943585,
      "grad_norm": 0.6493610709705243,
      "kl": 0.1061248779296875,
      "learning_rate": 4.980704295374469e-07,
      "loss": 0.0001,
      "reward": 1.775000050663948,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 9232
    },
    {
      "completion_length": 222.69197463989258,
      "epoch": 1.5482207971834527,
      "grad_norm": 0.25523525723368407,
      "kl": 0.114501953125,
      "learning_rate": 4.980689417300963e-07,
      "loss": 0.0001,
      "reward": 1.76250009983778,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9234
    },
    {
      "completion_length": 223.65625858306885,
      "epoch": 1.5485561004233204,
      "grad_norm": 0.15555612635792165,
      "kl": 0.1201171875,
      "learning_rate": 4.980674533515989e-07,
      "loss": 0.0001,
      "reward": 1.8285714462399483,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8285714536905289,
      "rewards/format_reward_func": 1.0,
      "step": 9236
    },
    {
      "completion_length": 216.60268783569336,
      "epoch": 1.548891403663188,
      "grad_norm": 0.0735118992899377,
      "kl": 0.10296630859375,
      "learning_rate": 4.980659644019584e-07,
      "loss": 0.0001,
      "reward": 1.7500000596046448,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 9238
    },
    {
      "completion_length": 217.47322368621826,
      "epoch": 1.5492267069030554,
      "grad_norm": 0.25828737432664267,
      "kl": 0.104248046875,
      "learning_rate": 4.980644748811778e-07,
      "loss": 0.0001,
      "reward": 1.7464286610484123,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 9240
    },
    {
      "completion_length": 214.54911518096924,
      "epoch": 1.549562010142923,
      "grad_norm": 0.21234573043333244,
      "kl": 0.132171630859375,
      "learning_rate": 4.980629847892611e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571809083223,
      "rewards/format_reward_func": 1.0,
      "step": 9242
    },
    {
      "completion_length": 218.95982933044434,
      "epoch": 1.5498973133827905,
      "grad_norm": 0.15823434184499677,
      "kl": 0.100311279296875,
      "learning_rate": 4.980614941262113e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 9244
    },
    {
      "completion_length": 212.95536708831787,
      "epoch": 1.550232616622658,
      "grad_norm": 0.21603125601891104,
      "kl": 0.103302001953125,
      "learning_rate": 4.98060002892032e-07,
      "loss": 0.0001,
      "reward": 1.82857146859169,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8285714648663998,
      "rewards/format_reward_func": 1.0,
      "step": 9246
    },
    {
      "completion_length": 217.14733123779297,
      "epoch": 1.5505679198625257,
      "grad_norm": 0.2851890565606586,
      "kl": 0.114898681640625,
      "learning_rate": 4.980585110867265e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 9248
    },
    {
      "completion_length": 213.21875858306885,
      "epoch": 1.5509032231023934,
      "grad_norm": 0.40561699254233524,
      "kl": 0.11175537109375,
      "learning_rate": 4.980570187102985e-07,
      "loss": 0.0001,
      "reward": 1.7267857864499092,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7312500402331352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9250
    },
    {
      "completion_length": 214.15179634094238,
      "epoch": 1.5512385263422608,
      "grad_norm": 0.2224340282727774,
      "kl": 0.097808837890625,
      "learning_rate": 4.980555257627511e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 9252
    },
    {
      "completion_length": 210.20090198516846,
      "epoch": 1.5515738295821282,
      "grad_norm": 0.1431330448665486,
      "kl": 0.11029052734375,
      "learning_rate": 4.980540322440881e-07,
      "loss": 0.0001,
      "reward": 1.832142896950245,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8321428745985031,
      "rewards/format_reward_func": 1.0,
      "step": 9254
    },
    {
      "completion_length": 215.96429443359375,
      "epoch": 1.5519091328219958,
      "grad_norm": 0.42489945876210056,
      "kl": 0.1027374267578125,
      "learning_rate": 4.980525381543126e-07,
      "loss": 0.0001,
      "reward": 1.7607143446803093,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143260538578,
      "rewards/format_reward_func": 1.0,
      "step": 9256
    },
    {
      "completion_length": 219.38840198516846,
      "epoch": 1.5522444360618635,
      "grad_norm": 0.24199594531058644,
      "kl": 0.11187744140625,
      "learning_rate": 4.980510434934283e-07,
      "loss": 0.0001,
      "reward": 1.7035714983940125,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7035714685916901,
      "rewards/format_reward_func": 1.0,
      "step": 9258
    },
    {
      "completion_length": 205.69197463989258,
      "epoch": 1.552579739301731,
      "grad_norm": 0.2722525309108444,
      "kl": 0.092529296875,
      "learning_rate": 4.980495482614384e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143223285675,
      "rewards/format_reward_func": 1.0,
      "step": 9260
    },
    {
      "completion_length": 218.0134038925171,
      "epoch": 1.5529150425415985,
      "grad_norm": 0.16731072750023818,
      "kl": 0.0987548828125,
      "learning_rate": 4.980480524583465e-07,
      "loss": 0.0001,
      "reward": 1.7571429088711739,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7660714574158192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9262
    },
    {
      "completion_length": 210.11608123779297,
      "epoch": 1.5532503457814661,
      "grad_norm": 0.20944642746491426,
      "kl": 0.098785400390625,
      "learning_rate": 4.98046556084156e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857264727354,
      "rewards/format_reward_func": 1.0,
      "step": 9264
    },
    {
      "completion_length": 220.58483123779297,
      "epoch": 1.5535856490213336,
      "grad_norm": 0.39444762467548544,
      "kl": 0.1048736572265625,
      "learning_rate": 4.980450591388705e-07,
      "loss": 0.0001,
      "reward": 1.7178572192788124,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7267857398837805,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9266
    },
    {
      "completion_length": 219.38393688201904,
      "epoch": 1.5539209522612012,
      "grad_norm": 0.23693520520013314,
      "kl": 0.102447509765625,
      "learning_rate": 4.980435616224932e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 9268
    },
    {
      "completion_length": 218.48661613464355,
      "epoch": 1.5542562555010688,
      "grad_norm": 0.653163923404132,
      "kl": 0.10009765625,
      "learning_rate": 4.980420635350277e-07,
      "loss": 0.0001,
      "reward": 1.79464291036129,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7991071566939354,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9270
    },
    {
      "completion_length": 212.60715293884277,
      "epoch": 1.5545915587409365,
      "grad_norm": 0.6311999205666676,
      "kl": 0.097198486328125,
      "learning_rate": 4.980405648764773e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.785714304074645,
      "rewards/format_reward_func": 1.0,
      "step": 9272
    },
    {
      "completion_length": 221.80358219146729,
      "epoch": 1.5549268619808039,
      "grad_norm": 0.7841224772161673,
      "kl": 0.11199951171875,
      "learning_rate": 4.980390656468456e-07,
      "loss": 0.0001,
      "reward": 1.776785783469677,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7812500260770321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9274
    },
    {
      "completion_length": 212.97322368621826,
      "epoch": 1.5552621652206713,
      "grad_norm": 0.4696213408353257,
      "kl": 0.111358642578125,
      "learning_rate": 4.980375658461361e-07,
      "loss": 0.0001,
      "reward": 1.7696429044008255,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7741071730852127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9276
    },
    {
      "completion_length": 213.40179443359375,
      "epoch": 1.555597468460539,
      "grad_norm": 0.18546115989892328,
      "kl": 0.10791015625,
      "learning_rate": 4.980360654743521e-07,
      "loss": 0.0001,
      "reward": 1.7750000730156898,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 9278
    },
    {
      "completion_length": 210.76786708831787,
      "epoch": 1.5559327717004066,
      "grad_norm": 0.25145691695231076,
      "kl": 0.109344482421875,
      "learning_rate": 4.98034564531497e-07,
      "loss": 0.0001,
      "reward": 1.737500049173832,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643234461546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9280
    },
    {
      "completion_length": 213.77679443359375,
      "epoch": 1.5562680749402742,
      "grad_norm": 0.20233206188903105,
      "kl": 0.125946044921875,
      "learning_rate": 4.980330630175746e-07,
      "loss": 0.0001,
      "reward": 1.7526786401867867,
      "reward_std": 0.056821079924702644,
      "rewards/equation_reward_func": 0.7589286062866449,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 9282
    },
    {
      "completion_length": 215.31251049041748,
      "epoch": 1.5566033781801418,
      "grad_norm": 0.2149426775367761,
      "kl": 0.14520263671875,
      "learning_rate": 4.980315609325879e-07,
      "loss": 0.0001,
      "reward": 1.7571429088711739,
      "reward_std": 0.10101525206118822,
      "rewards/equation_reward_func": 0.7660714574158192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9284
    },
    {
      "completion_length": 222.44643878936768,
      "epoch": 1.5569386814200092,
      "grad_norm": 0.3231430975015793,
      "kl": 0.1226806640625,
      "learning_rate": 4.980300582765406e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7660714574158192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9286
    },
    {
      "completion_length": 218.65179634094238,
      "epoch": 1.5572739846598767,
      "grad_norm": 0.1417486913493435,
      "kl": 0.122344970703125,
      "learning_rate": 4.980285550494362e-07,
      "loss": 0.0001,
      "reward": 1.769642911851406,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.774107176810503,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9288
    },
    {
      "completion_length": 223.09376049041748,
      "epoch": 1.5576092878997443,
      "grad_norm": 0.4358009612428243,
      "kl": 0.121795654296875,
      "learning_rate": 4.980270512512782e-07,
      "loss": 0.0001,
      "reward": 1.7535714656114578,
      "reward_std": 0.10606601648032665,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 9290
    },
    {
      "completion_length": 208.5178680419922,
      "epoch": 1.557944591139612,
      "grad_norm": 0.276077930647653,
      "kl": 0.1182861328125,
      "learning_rate": 4.980255468820699e-07,
      "loss": 0.0001,
      "reward": 1.7750000357627869,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 9292
    },
    {
      "completion_length": 227.321439743042,
      "epoch": 1.5582798943794796,
      "grad_norm": 0.2746660299952982,
      "kl": 0.126617431640625,
      "learning_rate": 4.980240419418148e-07,
      "loss": 0.0001,
      "reward": 1.7285714745521545,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7285714577883482,
      "rewards/format_reward_func": 1.0,
      "step": 9294
    },
    {
      "completion_length": 205.32590103149414,
      "epoch": 1.558615197619347,
      "grad_norm": 0.251892460012717,
      "kl": 0.115447998046875,
      "learning_rate": 4.980225364305164e-07,
      "loss": 0.0001,
      "reward": 1.8071428760886192,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8071428947150707,
      "rewards/format_reward_func": 1.0,
      "step": 9296
    },
    {
      "completion_length": 221.91965293884277,
      "epoch": 1.5589505008592146,
      "grad_norm": 0.5319414903169392,
      "kl": 0.1319580078125,
      "learning_rate": 4.980210303481782e-07,
      "loss": 0.0001,
      "reward": 1.7446429133415222,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071671247482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9298
    },
    {
      "completion_length": 227.07590293884277,
      "epoch": 1.559285804099082,
      "grad_norm": 0.436605329446181,
      "kl": 0.13720703125,
      "learning_rate": 4.980195236948036e-07,
      "loss": 0.0001,
      "reward": 1.7053572311997414,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7098214644938707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9300
    },
    {
      "completion_length": 225.92858123779297,
      "epoch": 1.5596211073389497,
      "grad_norm": 0.2605545821872648,
      "kl": 0.1298828125,
      "learning_rate": 4.98018016470396e-07,
      "loss": 0.0001,
      "reward": 1.7750000804662704,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 9302
    },
    {
      "completion_length": 227.6607265472412,
      "epoch": 1.5599564105788173,
      "grad_norm": 0.4288816041377394,
      "kl": 0.13018798828125,
      "learning_rate": 4.980165086749592e-07,
      "loss": 0.0001,
      "reward": 1.7763393446803093,
      "reward_std": 0.07386740390211344,
      "rewards/equation_reward_func": 0.7839286029338837,
      "rewards/format_reward_func": 0.9924107193946838,
      "step": 9304
    },
    {
      "completion_length": 225.93750953674316,
      "epoch": 1.560291713818685,
      "grad_norm": 0.19834380017707282,
      "kl": 0.1331787109375,
      "learning_rate": 4.980150003084962e-07,
      "loss": 0.0001,
      "reward": 1.7553571984171867,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.759821455925703,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9306
    },
    {
      "completion_length": 225.0312614440918,
      "epoch": 1.5606270170585523,
      "grad_norm": 0.1229449240722239,
      "kl": 0.127410888671875,
      "learning_rate": 4.980134913710108e-07,
      "loss": 0.0001,
      "reward": 1.7232143580913544,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7276786081492901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9308
    },
    {
      "completion_length": 226.49554634094238,
      "epoch": 1.5609623202984197,
      "grad_norm": 0.5843174232684581,
      "kl": 0.125244140625,
      "learning_rate": 4.980119818625064e-07,
      "loss": 0.0001,
      "reward": 1.7660715132951736,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7705357484519482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9310
    },
    {
      "completion_length": 233.28125858306885,
      "epoch": 1.5612976235382874,
      "grad_norm": 0.12572644726899307,
      "kl": 0.1334228515625,
      "learning_rate": 4.980104717829865e-07,
      "loss": 0.0001,
      "reward": 1.7517857775092125,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7562500312924385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9312
    },
    {
      "completion_length": 228.2098331451416,
      "epoch": 1.561632926778155,
      "grad_norm": 0.20672609168897751,
      "kl": 0.13775634765625,
      "learning_rate": 4.980089611324545e-07,
      "loss": 0.0001,
      "reward": 1.7285715118050575,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714652389288,
      "rewards/format_reward_func": 1.0,
      "step": 9314
    },
    {
      "completion_length": 232.10268878936768,
      "epoch": 1.5619682300180227,
      "grad_norm": 0.27711563932045935,
      "kl": 0.13140869140625,
      "learning_rate": 4.980074499109139e-07,
      "loss": 0.0001,
      "reward": 1.7071429342031479,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7071428894996643,
      "rewards/format_reward_func": 1.0,
      "step": 9316
    },
    {
      "completion_length": 235.50000953674316,
      "epoch": 1.56230353325789,
      "grad_norm": 0.2513881781273045,
      "kl": 0.143341064453125,
      "learning_rate": 4.980059381183682e-07,
      "loss": 0.0001,
      "reward": 1.7321429401636124,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321428749710321,
      "rewards/format_reward_func": 1.0,
      "step": 9318
    },
    {
      "completion_length": 222.54018688201904,
      "epoch": 1.5626388364977577,
      "grad_norm": 0.20648914774796778,
      "kl": 0.136016845703125,
      "learning_rate": 4.980044257548209e-07,
      "loss": 0.0001,
      "reward": 1.7678572237491608,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 9320
    },
    {
      "completion_length": 235.7634048461914,
      "epoch": 1.5629741397376251,
      "grad_norm": 0.26204188469958983,
      "kl": 0.159210205078125,
      "learning_rate": 4.980029128202755e-07,
      "loss": 0.0002,
      "reward": 1.7553572058677673,
      "reward_std": 0.0833375845104456,
      "rewards/equation_reward_func": 0.7687500342726707,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 9322
    },
    {
      "completion_length": 215.79465293884277,
      "epoch": 1.5633094429774927,
      "grad_norm": 0.31719240838874485,
      "kl": 0.118438720703125,
      "learning_rate": 4.980013993147353e-07,
      "loss": 0.0001,
      "reward": 1.7928571626543999,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 9324
    },
    {
      "completion_length": 211.04911518096924,
      "epoch": 1.5636447462173604,
      "grad_norm": 0.14979368350307748,
      "kl": 0.1125640869140625,
      "learning_rate": 4.97999885238204e-07,
      "loss": 0.0001,
      "reward": 1.8232143446803093,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8276785835623741,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9326
    },
    {
      "completion_length": 223.22768878936768,
      "epoch": 1.563980049457228,
      "grad_norm": 0.31624646214512,
      "kl": 0.12713623046875,
      "learning_rate": 4.979983705906852e-07,
      "loss": 0.0001,
      "reward": 1.7500000968575478,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 9328
    },
    {
      "completion_length": 218.57590293884277,
      "epoch": 1.5643153526970954,
      "grad_norm": 0.395515407485876,
      "kl": 0.124420166015625,
      "learning_rate": 4.979968553721819e-07,
      "loss": 0.0001,
      "reward": 1.7660714983940125,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7705357484519482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9330
    },
    {
      "completion_length": 211.87054634094238,
      "epoch": 1.5646506559369628,
      "grad_norm": 0.2938857964215618,
      "kl": 0.10455322265625,
      "learning_rate": 4.97995339582698e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 9332
    },
    {
      "completion_length": 212.06250953674316,
      "epoch": 1.5649859591768305,
      "grad_norm": 0.2958914012499181,
      "kl": 0.135986328125,
      "learning_rate": 4.97993823222237e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 9334
    },
    {
      "completion_length": 216.01786708831787,
      "epoch": 1.5653212624166981,
      "grad_norm": 0.22586410780715277,
      "kl": 0.13189697265625,
      "learning_rate": 4.979923062908022e-07,
      "loss": 0.0001,
      "reward": 1.7267858013510704,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7312500402331352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9336
    },
    {
      "completion_length": 210.58036422729492,
      "epoch": 1.5656565656565657,
      "grad_norm": 0.31286413540600805,
      "kl": 0.1175537109375,
      "learning_rate": 4.979907887883971e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 9338
    },
    {
      "completion_length": 209.39286613464355,
      "epoch": 1.5659918688964332,
      "grad_norm": 0.004725622950303789,
      "kl": 0.134185791015625,
      "learning_rate": 4.979892707150253e-07,
      "loss": 0.0001,
      "reward": 1.8035714700818062,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 9340
    },
    {
      "completion_length": 210.72322368621826,
      "epoch": 1.5663271721363008,
      "grad_norm": 0.22215252473600736,
      "kl": 0.126220703125,
      "learning_rate": 4.979877520706902e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7785714510828257,
      "rewards/format_reward_func": 1.0,
      "step": 9342
    },
    {
      "completion_length": 211.70982837677002,
      "epoch": 1.5666624753761682,
      "grad_norm": 0.28259642807042173,
      "kl": 0.126251220703125,
      "learning_rate": 4.979862328553954e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7678571697324514,
      "rewards/format_reward_func": 1.0,
      "step": 9344
    },
    {
      "completion_length": 203.66072463989258,
      "epoch": 1.5669977786160358,
      "grad_norm": 0.11420190732138948,
      "kl": 0.1182861328125,
      "learning_rate": 4.979847130691442e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000081956387,
      "rewards/format_reward_func": 1.0,
      "step": 9346
    },
    {
      "completion_length": 204.75893878936768,
      "epoch": 1.5673330818559035,
      "grad_norm": 0.2543779500102239,
      "kl": 0.117279052734375,
      "learning_rate": 4.979831927119405e-07,
      "loss": 0.0001,
      "reward": 1.7035715132951736,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7035714611411095,
      "rewards/format_reward_func": 1.0,
      "step": 9348
    },
    {
      "completion_length": 214.9107265472412,
      "epoch": 1.567668385095771,
      "grad_norm": 0.2692869245101748,
      "kl": 0.16180419921875,
      "learning_rate": 4.979816717837874e-07,
      "loss": 0.0002,
      "reward": 1.7339286357164383,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7383928950875998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9350
    },
    {
      "completion_length": 200.07143783569336,
      "epoch": 1.5680036883356385,
      "grad_norm": 0.3329726441021534,
      "kl": 0.144622802734375,
      "learning_rate": 4.979801502846885e-07,
      "loss": 0.0001,
      "reward": 1.785714328289032,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.785714328289032,
      "rewards/format_reward_func": 1.0,
      "step": 9352
    },
    {
      "completion_length": 206.977689743042,
      "epoch": 1.568338991575506,
      "grad_norm": 0.18730600752968107,
      "kl": 0.13232421875,
      "learning_rate": 4.979786282146474e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143022119999,
      "rewards/format_reward_func": 1.0,
      "step": 9354
    },
    {
      "completion_length": 208.29018783569336,
      "epoch": 1.5686742948153736,
      "grad_norm": 0.13967090602322763,
      "kl": 0.225189208984375,
      "learning_rate": 4.979771055736677e-07,
      "loss": 0.0002,
      "reward": 1.807142898440361,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428909897804,
      "rewards/format_reward_func": 1.0,
      "step": 9356
    },
    {
      "completion_length": 207.01340293884277,
      "epoch": 1.5690095980552412,
      "grad_norm": 0.27378668060925143,
      "kl": 0.123443603515625,
      "learning_rate": 4.979755823617525e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857447266579,
      "rewards/format_reward_func": 1.0,
      "step": 9358
    },
    {
      "completion_length": 204.61607933044434,
      "epoch": 1.5693449012951088,
      "grad_norm": 0.1799442544954095,
      "kl": 0.1424560546875,
      "learning_rate": 4.979740585789057e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428701281548,
      "rewards/format_reward_func": 1.0,
      "step": 9360
    },
    {
      "completion_length": 218.508939743042,
      "epoch": 1.5696802045349765,
      "grad_norm": 0.10917451343746819,
      "kl": 0.1553955078125,
      "learning_rate": 4.979725342251307e-07,
      "loss": 0.0002,
      "reward": 1.7892857789993286,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857640981674,
      "rewards/format_reward_func": 1.0,
      "step": 9362
    },
    {
      "completion_length": 199.85268878936768,
      "epoch": 1.5700155077748439,
      "grad_norm": 0.34883051529480286,
      "kl": 0.137542724609375,
      "learning_rate": 4.979710093004311e-07,
      "loss": 0.0001,
      "reward": 1.764285795390606,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 9364
    },
    {
      "completion_length": 213.64286613464355,
      "epoch": 1.5703508110147113,
      "grad_norm": 0.23714127105446917,
      "kl": 0.195068359375,
      "learning_rate": 4.9796948380481e-07,
      "loss": 0.0002,
      "reward": 1.7000000923871994,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7000000439584255,
      "rewards/format_reward_func": 1.0,
      "step": 9366
    },
    {
      "completion_length": 204.15179634094238,
      "epoch": 1.570686114254579,
      "grad_norm": 0.31596825075507784,
      "kl": 0.106353759765625,
      "learning_rate": 4.979679577382714e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571730852127,
      "rewards/format_reward_func": 1.0,
      "step": 9368
    },
    {
      "completion_length": 200.46875858306885,
      "epoch": 1.5710214174944466,
      "grad_norm": 0.25781774042292127,
      "kl": 0.127532958984375,
      "learning_rate": 4.979664311008185e-07,
      "loss": 0.0001,
      "reward": 1.8107143342494965,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 9370
    },
    {
      "completion_length": 207.22768783569336,
      "epoch": 1.5713567207343142,
      "grad_norm": 0.2883900615043515,
      "kl": 0.1649169921875,
      "learning_rate": 4.979649038924551e-07,
      "loss": 0.0002,
      "reward": 1.7642857730388641,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 9372
    },
    {
      "completion_length": 208.43304443359375,
      "epoch": 1.5716920239741816,
      "grad_norm": 0.26372455175352644,
      "kl": 0.18939208984375,
      "learning_rate": 4.979633761131845e-07,
      "loss": 0.0002,
      "reward": 1.7464286461472511,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7464286126196384,
      "rewards/format_reward_func": 1.0,
      "step": 9374
    },
    {
      "completion_length": 209.62947273254395,
      "epoch": 1.5720273272140493,
      "grad_norm": 0.3049719918557979,
      "kl": 0.19256591796875,
      "learning_rate": 4.979618477630102e-07,
      "loss": 0.0002,
      "reward": 1.7714286297559738,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714286111295223,
      "rewards/format_reward_func": 1.0,
      "step": 9376
    },
    {
      "completion_length": 201.66072273254395,
      "epoch": 1.5723626304539167,
      "grad_norm": 0.16845659994063056,
      "kl": 0.12188720703125,
      "learning_rate": 4.979603188419358e-07,
      "loss": 0.0001,
      "reward": 1.7392858117818832,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 1.0,
      "step": 9378
    },
    {
      "completion_length": 205.45536518096924,
      "epoch": 1.5726979336937843,
      "grad_norm": 0.22718930926429257,
      "kl": 0.132293701171875,
      "learning_rate": 4.979587893499649e-07,
      "loss": 0.0001,
      "reward": 1.735714353621006,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 9380
    },
    {
      "completion_length": 209.20983028411865,
      "epoch": 1.573033236933652,
      "grad_norm": 0.2509327116323553,
      "kl": 0.1441650390625,
      "learning_rate": 4.979572592871009e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 9382
    },
    {
      "completion_length": 211.25000858306885,
      "epoch": 1.5733685401735196,
      "grad_norm": 0.08130233488984963,
      "kl": 0.2010498046875,
      "learning_rate": 4.979557286533473e-07,
      "loss": 0.0002,
      "reward": 1.800000049173832,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 9384
    },
    {
      "completion_length": 200.88840007781982,
      "epoch": 1.573703843413387,
      "grad_norm": 0.08435947022938288,
      "kl": 0.12811279296875,
      "learning_rate": 4.979541974487077e-07,
      "loss": 0.0001,
      "reward": 1.7450893595814705,
      "reward_std": 0.033461302518844604,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 9386
    },
    {
      "completion_length": 216.1919755935669,
      "epoch": 1.5740391466532544,
      "grad_norm": 0.3632452957424689,
      "kl": 0.2205810546875,
      "learning_rate": 4.979526656731856e-07,
      "loss": 0.0002,
      "reward": 1.7303572222590446,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7348214574158192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9388
    },
    {
      "completion_length": 219.18304538726807,
      "epoch": 1.574374449893122,
      "grad_norm": 0.30401658940048676,
      "kl": 0.13153076171875,
      "learning_rate": 4.979511333267845e-07,
      "loss": 0.0001,
      "reward": 1.7267857939004898,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7312500402331352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9390
    },
    {
      "completion_length": 215.1741180419922,
      "epoch": 1.5747097531329897,
      "grad_norm": 0.24728973081644434,
      "kl": 0.20745849609375,
      "learning_rate": 4.979496004095081e-07,
      "loss": 0.0002,
      "reward": 1.7375001013278961,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7419643066823483,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9392
    },
    {
      "completion_length": 207.53572273254395,
      "epoch": 1.5750450563728573,
      "grad_norm": 0.21061804157548694,
      "kl": 0.24725341796875,
      "learning_rate": 4.979480669213596e-07,
      "loss": 0.0002,
      "reward": 1.739285796880722,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857559025288,
      "rewards/format_reward_func": 1.0,
      "step": 9394
    },
    {
      "completion_length": 197.40179443359375,
      "epoch": 1.5753803596127247,
      "grad_norm": 0.26878445750989804,
      "kl": 0.260589599609375,
      "learning_rate": 4.979465328623428e-07,
      "loss": 0.0003,
      "reward": 1.782142922282219,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 9396
    },
    {
      "completion_length": 191.18750953674316,
      "epoch": 1.5757156628525923,
      "grad_norm": 0.26487623591318504,
      "kl": 0.123138427734375,
      "learning_rate": 4.979449982324612e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 9398
    },
    {
      "completion_length": 193.94643878936768,
      "epoch": 1.5760509660924598,
      "grad_norm": 0.3455811579684974,
      "kl": 0.138336181640625,
      "learning_rate": 4.979434630317181e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 9400
    },
    {
      "completion_length": 200.52233123779297,
      "epoch": 1.5763862693323274,
      "grad_norm": 0.1336782326343173,
      "kl": 0.133148193359375,
      "learning_rate": 4.979419272601174e-07,
      "loss": 0.0001,
      "reward": 1.7464286461472511,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 9402
    },
    {
      "completion_length": 192.94643783569336,
      "epoch": 1.576721572572195,
      "grad_norm": 0.23079795904077532,
      "kl": 0.14453125,
      "learning_rate": 4.979403909176625e-07,
      "loss": 0.0001,
      "reward": 1.7821428999304771,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428868919611,
      "rewards/format_reward_func": 1.0,
      "step": 9404
    },
    {
      "completion_length": 203.18750858306885,
      "epoch": 1.5770568758120627,
      "grad_norm": 0.20095549925177159,
      "kl": 0.288726806640625,
      "learning_rate": 4.979388540043568e-07,
      "loss": 0.0003,
      "reward": 1.7750000730156898,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 9406
    },
    {
      "completion_length": 191.08036613464355,
      "epoch": 1.57739217905193,
      "grad_norm": 0.20087382205830567,
      "kl": 0.220916748046875,
      "learning_rate": 4.979373165202039e-07,
      "loss": 0.0002,
      "reward": 1.7535715103149414,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 9408
    },
    {
      "completion_length": 191.23661708831787,
      "epoch": 1.5777274822917975,
      "grad_norm": 0.13166586092674154,
      "kl": 0.138763427734375,
      "learning_rate": 4.979357784652073e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.753571467474103,
      "rewards/format_reward_func": 1.0,
      "step": 9410
    },
    {
      "completion_length": 188.56697177886963,
      "epoch": 1.5780627855316651,
      "grad_norm": 0.13549451743155633,
      "kl": 0.115631103515625,
      "learning_rate": 4.979342398393707e-07,
      "loss": 0.0001,
      "reward": 1.7214286476373672,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.721428606659174,
      "rewards/format_reward_func": 1.0,
      "step": 9412
    },
    {
      "completion_length": 181.10268592834473,
      "epoch": 1.5783980887715328,
      "grad_norm": 0.20587388166420736,
      "kl": 0.23712158203125,
      "learning_rate": 4.979327006426975e-07,
      "loss": 0.0002,
      "reward": 1.7053572311997414,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7098214700818062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9414
    },
    {
      "completion_length": 179.82143688201904,
      "epoch": 1.5787333920114004,
      "grad_norm": 0.324524725201001,
      "kl": 0.12750244140625,
      "learning_rate": 4.979311608751915e-07,
      "loss": 0.0001,
      "reward": 1.7321429401636124,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321429029107094,
      "rewards/format_reward_func": 1.0,
      "step": 9416
    },
    {
      "completion_length": 181.51340103149414,
      "epoch": 1.579068695251268,
      "grad_norm": 0.363719970992371,
      "kl": 0.156158447265625,
      "learning_rate": 4.979296205368558e-07,
      "loss": 0.0002,
      "reward": 1.7821428999304771,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821429036557674,
      "rewards/format_reward_func": 1.0,
      "step": 9418
    },
    {
      "completion_length": 184.34375953674316,
      "epoch": 1.5794039984911354,
      "grad_norm": 0.2968390327039545,
      "kl": 0.14080810546875,
      "learning_rate": 4.979280796276943e-07,
      "loss": 0.0001,
      "reward": 1.74642863124609,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.746428607031703,
      "rewards/format_reward_func": 1.0,
      "step": 9420
    },
    {
      "completion_length": 184.85715103149414,
      "epoch": 1.5797393017310029,
      "grad_norm": 0.2681062341314168,
      "kl": 0.1266632080078125,
      "learning_rate": 4.979265381477104e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 9422
    },
    {
      "completion_length": 181.8125057220459,
      "epoch": 1.5800746049708705,
      "grad_norm": 0.18636219767722761,
      "kl": 0.105621337890625,
      "learning_rate": 4.979249960969077e-07,
      "loss": 0.0001,
      "reward": 1.7410715222358704,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.745535746216774,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9424
    },
    {
      "completion_length": 192.44643592834473,
      "epoch": 1.5804099082107381,
      "grad_norm": 0.2592943293661732,
      "kl": 0.173614501953125,
      "learning_rate": 4.979234534752898e-07,
      "loss": 0.0002,
      "reward": 1.778571493923664,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 9426
    },
    {
      "completion_length": 180.08929347991943,
      "epoch": 1.5807452114506058,
      "grad_norm": 0.018409488738189025,
      "kl": 0.177093505859375,
      "learning_rate": 4.979219102828601e-07,
      "loss": 0.0002,
      "reward": 1.7678571939468384,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571902215481,
      "rewards/format_reward_func": 1.0,
      "step": 9428
    },
    {
      "completion_length": 188.42857933044434,
      "epoch": 1.5810805146904732,
      "grad_norm": 0.19724579011986249,
      "kl": 0.108642578125,
      "learning_rate": 4.979203665196222e-07,
      "loss": 0.0001,
      "reward": 1.7142857983708382,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857424914837,
      "rewards/format_reward_func": 1.0,
      "step": 9430
    },
    {
      "completion_length": 188.49554347991943,
      "epoch": 1.5814158179303408,
      "grad_norm": 0.37446507304918436,
      "kl": 0.10906982421875,
      "learning_rate": 4.979188221855797e-07,
      "loss": 0.0001,
      "reward": 1.7285714894533157,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7285714596509933,
      "rewards/format_reward_func": 1.0,
      "step": 9432
    },
    {
      "completion_length": 188.35268688201904,
      "epoch": 1.5817511211702082,
      "grad_norm": 0.28298486688243313,
      "kl": 0.1163330078125,
      "learning_rate": 4.979172772807363e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 9434
    },
    {
      "completion_length": 180.41518592834473,
      "epoch": 1.5820864244100759,
      "grad_norm": 0.19784500966747084,
      "kl": 0.117919921875,
      "learning_rate": 4.979157318050953e-07,
      "loss": 0.0001,
      "reward": 1.7571429088711739,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 9436
    },
    {
      "completion_length": 182.25893783569336,
      "epoch": 1.5824217276499435,
      "grad_norm": 0.18481772827255594,
      "kl": 0.114227294921875,
      "learning_rate": 4.979141857586604e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.757142897695303,
      "rewards/format_reward_func": 1.0,
      "step": 9438
    },
    {
      "completion_length": 181.86607933044434,
      "epoch": 1.5827570308898111,
      "grad_norm": 0.40152965965342363,
      "kl": 0.295135498046875,
      "learning_rate": 4.979126391414352e-07,
      "loss": 0.0003,
      "reward": 1.7428572326898575,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 9440
    },
    {
      "completion_length": 188.20090293884277,
      "epoch": 1.5830923341296785,
      "grad_norm": 0.3072119809838924,
      "kl": 0.13543701171875,
      "learning_rate": 4.97911091953423e-07,
      "loss": 0.0001,
      "reward": 1.72857154160738,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7285714540630579,
      "rewards/format_reward_func": 1.0,
      "step": 9442
    },
    {
      "completion_length": 184.23661518096924,
      "epoch": 1.583427637369546,
      "grad_norm": 0.27136970912830893,
      "kl": 0.186431884765625,
      "learning_rate": 4.979095441946276e-07,
      "loss": 0.0002,
      "reward": 1.7785715013742447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 9444
    },
    {
      "completion_length": 184.88393688201904,
      "epoch": 1.5837629406094136,
      "grad_norm": 0.2625038055494905,
      "kl": 0.111175537109375,
      "learning_rate": 4.979079958650525e-07,
      "loss": 0.0001,
      "reward": 1.735714353621006,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7357143126428127,
      "rewards/format_reward_func": 1.0,
      "step": 9446
    },
    {
      "completion_length": 188.89733028411865,
      "epoch": 1.5840982438492812,
      "grad_norm": 0.20991182672532305,
      "kl": 0.110870361328125,
      "learning_rate": 4.979064469647014e-07,
      "loss": 0.0001,
      "reward": 1.7392857819795609,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 1.0,
      "step": 9448
    },
    {
      "completion_length": 187.84822273254395,
      "epoch": 1.5844335470891489,
      "grad_norm": 0.24590372324776322,
      "kl": 0.131500244140625,
      "learning_rate": 4.979048974935776e-07,
      "loss": 0.0001,
      "reward": 1.7642857655882835,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 9450
    },
    {
      "completion_length": 185.66072368621826,
      "epoch": 1.5847688503290163,
      "grad_norm": 0.15685101508432459,
      "kl": 0.204193115234375,
      "learning_rate": 4.97903347451685e-07,
      "loss": 0.0002,
      "reward": 1.7357143387198448,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.735714316368103,
      "rewards/format_reward_func": 1.0,
      "step": 9452
    },
    {
      "completion_length": 204.43750762939453,
      "epoch": 1.585104153568884,
      "grad_norm": 0.28556903563308955,
      "kl": 0.4090576171875,
      "learning_rate": 4.979017968390268e-07,
      "loss": 0.0004,
      "reward": 1.7250000983476639,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7250000182539225,
      "rewards/format_reward_func": 1.0,
      "step": 9454
    },
    {
      "completion_length": 192.05804538726807,
      "epoch": 1.5854394568087513,
      "grad_norm": 0.268773106820556,
      "kl": 0.1308441162109375,
      "learning_rate": 4.979002456556068e-07,
      "loss": 0.0001,
      "reward": 1.7625000774860382,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9456
    },
    {
      "completion_length": 199.78125858306885,
      "epoch": 1.585774760048619,
      "grad_norm": 0.24722104897199124,
      "kl": 0.10369873046875,
      "learning_rate": 4.978986939014285e-07,
      "loss": 0.0001,
      "reward": 1.732142947614193,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321428954601288,
      "rewards/format_reward_func": 1.0,
      "step": 9458
    },
    {
      "completion_length": 198.80804347991943,
      "epoch": 1.5861100632884866,
      "grad_norm": 0.20960824627267807,
      "kl": 0.140289306640625,
      "learning_rate": 4.978971415764955e-07,
      "loss": 0.0001,
      "reward": 1.7464286610484123,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 9460
    },
    {
      "completion_length": 203.90179443359375,
      "epoch": 1.5864453665283542,
      "grad_norm": 0.291625399876519,
      "kl": 0.148956298828125,
      "learning_rate": 4.978955886808114e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857544124126,
      "rewards/format_reward_func": 1.0,
      "step": 9462
    },
    {
      "completion_length": 196.58483123779297,
      "epoch": 1.5867806697682216,
      "grad_norm": 0.3246287484114839,
      "kl": 0.113006591796875,
      "learning_rate": 4.978940352143797e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 9464
    },
    {
      "completion_length": 186.13840198516846,
      "epoch": 1.587115973008089,
      "grad_norm": 0.4854994353952275,
      "kl": 0.113677978515625,
      "learning_rate": 4.97892481177204e-07,
      "loss": 0.0001,
      "reward": 1.721428669989109,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7214286103844643,
      "rewards/format_reward_func": 1.0,
      "step": 9466
    },
    {
      "completion_length": 197.19197368621826,
      "epoch": 1.5874512762479567,
      "grad_norm": 0.31571928329953813,
      "kl": 0.101104736328125,
      "learning_rate": 4.97890926569288e-07,
      "loss": 0.0001,
      "reward": 1.8000000715255737,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 9468
    },
    {
      "completion_length": 198.43750953674316,
      "epoch": 1.5877865794878243,
      "grad_norm": 0.18825773566152013,
      "kl": 0.14898681640625,
      "learning_rate": 4.978893713906351e-07,
      "loss": 0.0001,
      "reward": 1.7678572237491608,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571864962578,
      "rewards/format_reward_func": 1.0,
      "step": 9470
    },
    {
      "completion_length": 206.13840198516846,
      "epoch": 1.588121882727692,
      "grad_norm": 0.1629431048033093,
      "kl": 0.111083984375,
      "learning_rate": 4.97887815641249e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 9472
    },
    {
      "completion_length": 203.19197273254395,
      "epoch": 1.5884571859675594,
      "grad_norm": 0.33415035155337125,
      "kl": 0.104461669921875,
      "learning_rate": 4.978862593211331e-07,
      "loss": 0.0001,
      "reward": 1.7508929297327995,
      "reward_std": 0.08965103607624769,
      "rewards/equation_reward_func": 0.7526786141097546,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 9474
    },
    {
      "completion_length": 195.89286708831787,
      "epoch": 1.588792489207427,
      "grad_norm": 0.3242104161132054,
      "kl": 0.09625244140625,
      "learning_rate": 4.978847024302914e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000417232513,
      "rewards/format_reward_func": 1.0,
      "step": 9476
    },
    {
      "completion_length": 194.82590293884277,
      "epoch": 1.5891277924472944,
      "grad_norm": 0.2486906112229366,
      "kl": 0.0967254638671875,
      "learning_rate": 4.97883144968727e-07,
      "loss": 0.0001,
      "reward": 1.7392857819795609,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857540398836,
      "rewards/format_reward_func": 1.0,
      "step": 9478
    },
    {
      "completion_length": 196.92411422729492,
      "epoch": 1.589463095687162,
      "grad_norm": 0.19468741497020042,
      "kl": 0.10626220703125,
      "learning_rate": 4.978815869364437e-07,
      "loss": 0.0001,
      "reward": 1.7464286237955093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464286237955093,
      "rewards/format_reward_func": 1.0,
      "step": 9480
    },
    {
      "completion_length": 205.51786613464355,
      "epoch": 1.5897983989270297,
      "grad_norm": 0.21640894935551291,
      "kl": 0.12603759765625,
      "learning_rate": 4.978800283334451e-07,
      "loss": 0.0001,
      "reward": 1.7272322252392769,
      "reward_std": 0.04230013629421592,
      "rewards/equation_reward_func": 0.7285714596509933,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 9482
    },
    {
      "completion_length": 193.09375762939453,
      "epoch": 1.5901337021668973,
      "grad_norm": 0.15703601361285324,
      "kl": 0.100189208984375,
      "learning_rate": 4.978784691597347e-07,
      "loss": 0.0001,
      "reward": 1.7892857640981674,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7892857361584902,
      "rewards/format_reward_func": 1.0,
      "step": 9484
    },
    {
      "completion_length": 188.65179347991943,
      "epoch": 1.5904690054067647,
      "grad_norm": 0.2601317163988536,
      "kl": 0.136810302734375,
      "learning_rate": 4.978769094153163e-07,
      "loss": 0.0001,
      "reward": 1.7857143580913544,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143245637417,
      "rewards/format_reward_func": 1.0,
      "step": 9486
    },
    {
      "completion_length": 196.0580472946167,
      "epoch": 1.5908043086466321,
      "grad_norm": 0.2600653373287085,
      "kl": 0.117095947265625,
      "learning_rate": 4.978753491001933e-07,
      "loss": 0.0001,
      "reward": 1.7857143133878708,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143245637417,
      "rewards/format_reward_func": 1.0,
      "step": 9488
    },
    {
      "completion_length": 200.73215198516846,
      "epoch": 1.5911396118864998,
      "grad_norm": 0.27887034524746496,
      "kl": 0.18231201171875,
      "learning_rate": 4.978737882143693e-07,
      "loss": 0.0002,
      "reward": 1.8178572058677673,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.817857164889574,
      "rewards/format_reward_func": 1.0,
      "step": 9490
    },
    {
      "completion_length": 202.33036708831787,
      "epoch": 1.5914749151263674,
      "grad_norm": 0.3410612068150405,
      "kl": 0.125335693359375,
      "learning_rate": 4.97872226757848e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7678571734577417,
      "rewards/format_reward_func": 1.0,
      "step": 9492
    },
    {
      "completion_length": 203.17858123779297,
      "epoch": 1.591810218366235,
      "grad_norm": 0.18929719663093758,
      "kl": 0.266021728515625,
      "learning_rate": 4.978706647306329e-07,
      "loss": 0.0003,
      "reward": 1.717857226729393,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7178571727126837,
      "rewards/format_reward_func": 1.0,
      "step": 9494
    },
    {
      "completion_length": 194.48661708831787,
      "epoch": 1.5921455216061027,
      "grad_norm": 0.44137712767320597,
      "kl": 0.210601806640625,
      "learning_rate": 4.978691021327276e-07,
      "loss": 0.0002,
      "reward": 1.7553572058677673,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7598214596509933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9496
    },
    {
      "completion_length": 208.91965007781982,
      "epoch": 1.59248082484597,
      "grad_norm": 0.3316997283825163,
      "kl": 0.11297607421875,
      "learning_rate": 4.978675389641357e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 9498
    },
    {
      "completion_length": 200.73215103149414,
      "epoch": 1.5928161280858375,
      "grad_norm": 0.27617819723690057,
      "kl": 0.2326507568359375,
      "learning_rate": 4.978659752248608e-07,
      "loss": 0.0002,
      "reward": 1.7714286223053932,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7714286111295223,
      "rewards/format_reward_func": 1.0,
      "step": 9500
    },
    {
      "completion_length": 210.42411613464355,
      "epoch": 1.5931514313257051,
      "grad_norm": 0.23732779196931864,
      "kl": 0.19146728515625,
      "learning_rate": 4.978644109149066e-07,
      "loss": 0.0002,
      "reward": 1.778571479022503,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 9502
    },
    {
      "completion_length": 206.71875858306885,
      "epoch": 1.5934867345655728,
      "grad_norm": 0.1889520164049944,
      "kl": 0.13079833984375,
      "learning_rate": 4.978628460342766e-07,
      "loss": 0.0001,
      "reward": 1.7714286148548126,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 9504
    },
    {
      "completion_length": 209.49107837677002,
      "epoch": 1.5938220378054404,
      "grad_norm": 0.28097707245087605,
      "kl": 0.125396728515625,
      "learning_rate": 4.978612805829745e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7571428846567869,
      "rewards/format_reward_func": 1.0,
      "step": 9506
    },
    {
      "completion_length": 219.37947368621826,
      "epoch": 1.5941573410453078,
      "grad_norm": 0.1472375894145825,
      "kl": 0.1482086181640625,
      "learning_rate": 4.978597145610037e-07,
      "loss": 0.0001,
      "reward": 1.7035714983940125,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7125000394880772,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9508
    },
    {
      "completion_length": 194.54465103149414,
      "epoch": 1.5944926442851755,
      "grad_norm": 0.30787715412318145,
      "kl": 0.1240234375,
      "learning_rate": 4.97858147968368e-07,
      "loss": 0.0001,
      "reward": 1.7732143476605415,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7776785977184772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9510
    },
    {
      "completion_length": 205.63840198516846,
      "epoch": 1.5948279475250429,
      "grad_norm": 0.22203363715006483,
      "kl": 0.12939453125,
      "learning_rate": 4.97856580805071e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714693367481,
      "rewards/format_reward_func": 1.0,
      "step": 9512
    },
    {
      "completion_length": 218.54465293884277,
      "epoch": 1.5951632507649105,
      "grad_norm": 0.3115238153228958,
      "kl": 0.197662353515625,
      "learning_rate": 4.978550130711163e-07,
      "loss": 0.0002,
      "reward": 1.7607143819332123,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.760714303702116,
      "rewards/format_reward_func": 1.0,
      "step": 9514
    },
    {
      "completion_length": 210.4062614440918,
      "epoch": 1.5954985540047781,
      "grad_norm": 0.33087151315019003,
      "kl": 0.162689208984375,
      "learning_rate": 4.978534447665072e-07,
      "loss": 0.0002,
      "reward": 1.76071435213089,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 9516
    },
    {
      "completion_length": 219.63840198516846,
      "epoch": 1.5958338572446458,
      "grad_norm": 0.3291601561579799,
      "kl": 0.1533203125,
      "learning_rate": 4.978518758912478e-07,
      "loss": 0.0002,
      "reward": 1.712500087916851,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.716964328661561,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9518
    },
    {
      "completion_length": 217.66518688201904,
      "epoch": 1.5961691604845132,
      "grad_norm": 0.16251913228443396,
      "kl": 0.175567626953125,
      "learning_rate": 4.978503064453414e-07,
      "loss": 0.0002,
      "reward": 1.7875000536441803,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7919643074274063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9520
    },
    {
      "completion_length": 214.29911708831787,
      "epoch": 1.5965044637243806,
      "grad_norm": 0.275822424940639,
      "kl": 0.133331298828125,
      "learning_rate": 4.978487364287918e-07,
      "loss": 0.0001,
      "reward": 1.7589286491274834,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7633928880095482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9522
    },
    {
      "completion_length": 222.16965293884277,
      "epoch": 1.5968397669642482,
      "grad_norm": 0.28340261424616403,
      "kl": 0.2544708251953125,
      "learning_rate": 4.978471658416024e-07,
      "loss": 0.0003,
      "reward": 1.7392857819795609,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7482143118977547,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9524
    },
    {
      "completion_length": 219.68304634094238,
      "epoch": 1.5971750702041159,
      "grad_norm": 0.3308031864705601,
      "kl": 0.1259765625,
      "learning_rate": 4.978455946837769e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 9526
    },
    {
      "completion_length": 228.6205472946167,
      "epoch": 1.5975103734439835,
      "grad_norm": 0.23764646592147184,
      "kl": 0.1822509765625,
      "learning_rate": 4.978440229553191e-07,
      "loss": 0.0002,
      "reward": 1.7535714954137802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 9528
    },
    {
      "completion_length": 221.06697273254395,
      "epoch": 1.597845676683851,
      "grad_norm": 0.1900498694957338,
      "kl": 0.152618408203125,
      "learning_rate": 4.978424506562324e-07,
      "loss": 0.0002,
      "reward": 1.780357226729393,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7848214469850063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9530
    },
    {
      "completion_length": 238.60268783569336,
      "epoch": 1.5981809799237185,
      "grad_norm": 0.21449225533502828,
      "kl": 0.18243408203125,
      "learning_rate": 4.978408777865204e-07,
      "loss": 0.0002,
      "reward": 1.71428582072258,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7142857350409031,
      "rewards/format_reward_func": 1.0,
      "step": 9532
    },
    {
      "completion_length": 228.45536708831787,
      "epoch": 1.598516283163586,
      "grad_norm": 0.25287001539221815,
      "kl": 0.203948974609375,
      "learning_rate": 4.978393043461869e-07,
      "loss": 0.0002,
      "reward": 1.7785714864730835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 9534
    },
    {
      "completion_length": 251.9598331451416,
      "epoch": 1.5988515864034536,
      "grad_norm": 0.35490299267432196,
      "kl": 0.30194091796875,
      "learning_rate": 4.978377303352353e-07,
      "loss": 0.0003,
      "reward": 1.705357238650322,
      "reward_std": 0.13384521286934614,
      "rewards/equation_reward_func": 0.7276785857975483,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 9536
    },
    {
      "completion_length": 235.15179824829102,
      "epoch": 1.5991868896433212,
      "grad_norm": 0.2435142863699395,
      "kl": 0.14727783203125,
      "learning_rate": 4.978361557536696e-07,
      "loss": 0.0001,
      "reward": 1.7714286521077156,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7803571745753288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9538
    },
    {
      "completion_length": 230.4866180419922,
      "epoch": 1.5995221928831889,
      "grad_norm": 0.26605692272202913,
      "kl": 0.159637451171875,
      "learning_rate": 4.97834580601493e-07,
      "loss": 0.0002,
      "reward": 1.7589286342263222,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7633928917348385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9540
    },
    {
      "completion_length": 235.8571548461914,
      "epoch": 1.5998574961230563,
      "grad_norm": 0.18506222865561817,
      "kl": 0.141998291015625,
      "learning_rate": 4.978330048787092e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.8089286088943481,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9542
    },
    {
      "completion_length": 225.2857255935669,
      "epoch": 1.6001927993629237,
      "grad_norm": 0.24497391090326853,
      "kl": 0.1688232421875,
      "learning_rate": 4.978314285853222e-07,
      "loss": 0.0002,
      "reward": 1.8107143267989159,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8107143044471741,
      "rewards/format_reward_func": 1.0,
      "step": 9544
    },
    {
      "completion_length": 229.04465293884277,
      "epoch": 1.6005281026027913,
      "grad_norm": 0.16757262472399967,
      "kl": 0.142120361328125,
      "learning_rate": 4.978298517213352e-07,
      "loss": 0.0001,
      "reward": 1.7142857909202576,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7232143171131611,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9546
    },
    {
      "completion_length": 230.67858219146729,
      "epoch": 1.600863405842659,
      "grad_norm": 0.32121427829377835,
      "kl": 0.168212890625,
      "learning_rate": 4.97828274286752e-07,
      "loss": 0.0002,
      "reward": 1.7035714983940125,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7035714648663998,
      "rewards/format_reward_func": 1.0,
      "step": 9548
    },
    {
      "completion_length": 242.65626049041748,
      "epoch": 1.6011987090825266,
      "grad_norm": 0.3786336398381342,
      "kl": 0.359466552734375,
      "learning_rate": 4.978266962815763e-07,
      "loss": 0.0004,
      "reward": 1.6892857998609543,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.6982143260538578,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9550
    },
    {
      "completion_length": 218.7142972946167,
      "epoch": 1.6015340123223942,
      "grad_norm": 0.1542904347763142,
      "kl": 0.2550048828125,
      "learning_rate": 4.978251177058116e-07,
      "loss": 0.0003,
      "reward": 1.766071505844593,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7705357372760773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9552
    },
    {
      "completion_length": 220.08483219146729,
      "epoch": 1.6018693155622616,
      "grad_norm": 0.2862551303649256,
      "kl": 0.214569091796875,
      "learning_rate": 4.978235385594616e-07,
      "loss": 0.0002,
      "reward": 1.8107143640518188,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.8107142969965935,
      "rewards/format_reward_func": 1.0,
      "step": 9554
    },
    {
      "completion_length": 225.383939743042,
      "epoch": 1.602204618802129,
      "grad_norm": 0.24500226164542638,
      "kl": 0.258697509765625,
      "learning_rate": 4.9782195884253e-07,
      "loss": 0.0003,
      "reward": 1.7464286386966705,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286033064127,
      "rewards/format_reward_func": 1.0,
      "step": 9556
    },
    {
      "completion_length": 211.54018783569336,
      "epoch": 1.6025399220419967,
      "grad_norm": 0.2018758611186633,
      "kl": 0.230255126953125,
      "learning_rate": 4.978203785550203e-07,
      "loss": 0.0002,
      "reward": 1.8357143476605415,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8357143104076385,
      "rewards/format_reward_func": 1.0,
      "step": 9558
    },
    {
      "completion_length": 226.4553680419922,
      "epoch": 1.6028752252818643,
      "grad_norm": 0.22041226567845768,
      "kl": 0.21649169921875,
      "learning_rate": 4.978187976969361e-07,
      "loss": 0.0002,
      "reward": 1.7392857745289803,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.7482143305242062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9560
    },
    {
      "completion_length": 203.51340198516846,
      "epoch": 1.603210528521732,
      "grad_norm": 0.6848401584259823,
      "kl": 0.314666748046875,
      "learning_rate": 4.978172162682812e-07,
      "loss": 0.0003,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857581377029,
      "rewards/format_reward_func": 1.0,
      "step": 9562
    },
    {
      "completion_length": 208.80804443359375,
      "epoch": 1.6035458317615994,
      "grad_norm": 0.23038082418106143,
      "kl": 0.174072265625,
      "learning_rate": 4.978156342690593e-07,
      "loss": 0.0002,
      "reward": 1.7946429252624512,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7991071678698063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9564
    },
    {
      "completion_length": 210.25893783569336,
      "epoch": 1.603881135001467,
      "grad_norm": 0.26692083769371217,
      "kl": 0.13800048828125,
      "learning_rate": 4.978140516992738e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 9566
    },
    {
      "completion_length": 209.26786613464355,
      "epoch": 1.6042164382413344,
      "grad_norm": 0.6875469688497401,
      "kl": 0.182403564453125,
      "learning_rate": 4.978124685589286e-07,
      "loss": 0.0002,
      "reward": 1.7714286595582962,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 9568
    },
    {
      "completion_length": 211.85268688201904,
      "epoch": 1.604551741481202,
      "grad_norm": 0.10928835035140934,
      "kl": 0.168670654296875,
      "learning_rate": 4.978108848480271e-07,
      "loss": 0.0002,
      "reward": 1.716071479022503,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7205357421189547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9570
    },
    {
      "completion_length": 206.12500953674316,
      "epoch": 1.6048870447210697,
      "grad_norm": 0.35867449197804246,
      "kl": 0.158905029296875,
      "learning_rate": 4.97809300566573e-07,
      "loss": 0.0002,
      "reward": 1.825000062584877,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8250000327825546,
      "rewards/format_reward_func": 1.0,
      "step": 9572
    },
    {
      "completion_length": 208.10268878936768,
      "epoch": 1.6052223479609373,
      "grad_norm": 0.3136847225169866,
      "kl": 0.259307861328125,
      "learning_rate": 4.978077157145702e-07,
      "loss": 0.0003,
      "reward": 1.7446429133415222,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7491071820259094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9574
    },
    {
      "completion_length": 194.73661422729492,
      "epoch": 1.6055576512008047,
      "grad_norm": 0.1617781791123457,
      "kl": 0.131927490234375,
      "learning_rate": 4.97806130292022e-07,
      "loss": 0.0001,
      "reward": 1.7089286595582962,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7133928909897804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9576
    },
    {
      "completion_length": 200.80358028411865,
      "epoch": 1.6058929544406721,
      "grad_norm": 0.5766524950768712,
      "kl": 0.162017822265625,
      "learning_rate": 4.978045442989323e-07,
      "loss": 0.0002,
      "reward": 1.775000087916851,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 9578
    },
    {
      "completion_length": 193.70090198516846,
      "epoch": 1.6062282576805398,
      "grad_norm": 0.33900780695246113,
      "kl": 0.133544921875,
      "learning_rate": 4.978029577353047e-07,
      "loss": 0.0001,
      "reward": 1.7714286521077156,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285925030708,
      "rewards/format_reward_func": 1.0,
      "step": 9580
    },
    {
      "completion_length": 198.50893688201904,
      "epoch": 1.6065635609204074,
      "grad_norm": 0.31181010593599673,
      "kl": 0.1903076171875,
      "learning_rate": 4.978013706011427e-07,
      "loss": 0.0002,
      "reward": 1.7500000596046448,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000279396772,
      "rewards/format_reward_func": 1.0,
      "step": 9582
    },
    {
      "completion_length": 201.63840293884277,
      "epoch": 1.606898864160275,
      "grad_norm": 0.2769393664102841,
      "kl": 0.160369873046875,
      "learning_rate": 4.977997828964501e-07,
      "loss": 0.0002,
      "reward": 1.7928571850061417,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 9584
    },
    {
      "completion_length": 192.71875858306885,
      "epoch": 1.6072341674001425,
      "grad_norm": 0.3967558757175834,
      "kl": 0.1456298828125,
      "learning_rate": 4.977981946212305e-07,
      "loss": 0.0001,
      "reward": 1.8035715073347092,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.8035714440047741,
      "rewards/format_reward_func": 1.0,
      "step": 9586
    },
    {
      "completion_length": 209.50893878936768,
      "epoch": 1.60756947064001,
      "grad_norm": 0.08294927589433111,
      "kl": 0.18817138671875,
      "learning_rate": 4.977966057754877e-07,
      "loss": 0.0002,
      "reward": 1.733928643167019,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7383928876370192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9588
    },
    {
      "completion_length": 208.34375953674316,
      "epoch": 1.6079047738798775,
      "grad_norm": 0.21126790987528263,
      "kl": 0.141693115234375,
      "learning_rate": 4.977950163592251e-07,
      "loss": 0.0001,
      "reward": 1.7214286401867867,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7214286215603352,
      "rewards/format_reward_func": 1.0,
      "step": 9590
    },
    {
      "completion_length": 222.32590198516846,
      "epoch": 1.6082400771197451,
      "grad_norm": 0.2547810669547315,
      "kl": 0.19598388671875,
      "learning_rate": 4.977934263724466e-07,
      "loss": 0.0002,
      "reward": 1.7285715118050575,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7285714671015739,
      "rewards/format_reward_func": 1.0,
      "step": 9592
    },
    {
      "completion_length": 208.25000858306885,
      "epoch": 1.6085753803596128,
      "grad_norm": 0.1419821972171379,
      "kl": 0.1297607421875,
      "learning_rate": 4.977918358151557e-07,
      "loss": 0.0001,
      "reward": 1.764285795390606,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 9594
    },
    {
      "completion_length": 199.52233123779297,
      "epoch": 1.6089106835994804,
      "grad_norm": 0.17384380850416808,
      "kl": 0.115386962890625,
      "learning_rate": 4.977902446873561e-07,
      "loss": 0.0001,
      "reward": 1.776785746216774,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7812500260770321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9596
    },
    {
      "completion_length": 201.48661613464355,
      "epoch": 1.6092459868393478,
      "grad_norm": 0.36774919060599237,
      "kl": 0.175018310546875,
      "learning_rate": 4.977886529890515e-07,
      "loss": 0.0002,
      "reward": 1.8000000417232513,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.800000037997961,
      "rewards/format_reward_func": 1.0,
      "step": 9598
    },
    {
      "completion_length": 213.88393878936768,
      "epoch": 1.6095812900792152,
      "grad_norm": 0.20051143266263496,
      "kl": 0.131591796875,
      "learning_rate": 4.977870607202456e-07,
      "loss": 0.0001,
      "reward": 1.7357143759727478,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143089175224,
      "rewards/format_reward_func": 1.0,
      "step": 9600
    },
    {
      "completion_length": 216.59822368621826,
      "epoch": 1.6099165933190829,
      "grad_norm": 0.29833393306666617,
      "kl": 0.210479736328125,
      "learning_rate": 4.977854678809419e-07,
      "loss": 0.0002,
      "reward": 1.7821429297327995,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428999304771,
      "rewards/format_reward_func": 1.0,
      "step": 9602
    },
    {
      "completion_length": 225.0178689956665,
      "epoch": 1.6102518965589505,
      "grad_norm": 0.2855325199172955,
      "kl": 0.134735107421875,
      "learning_rate": 4.977838744711443e-07,
      "loss": 0.0001,
      "reward": 1.7214286476373672,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.721428606659174,
      "rewards/format_reward_func": 1.0,
      "step": 9604
    },
    {
      "completion_length": 219.96429538726807,
      "epoch": 1.6105871997988181,
      "grad_norm": 0.1848297056822976,
      "kl": 0.1536865234375,
      "learning_rate": 4.977822804908562e-07,
      "loss": 0.0002,
      "reward": 1.7950893342494965,
      "reward_std": 0.02714784862473607,
      "rewards/equation_reward_func": 0.7964285928755999,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 9606
    },
    {
      "completion_length": 218.5178680419922,
      "epoch": 1.6109225030386856,
      "grad_norm": 0.12919435157945253,
      "kl": 0.184478759765625,
      "learning_rate": 4.977806859400816e-07,
      "loss": 0.0002,
      "reward": 1.8000000566244125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 9608
    },
    {
      "completion_length": 238.8705472946167,
      "epoch": 1.6112578062785532,
      "grad_norm": 0.6828934653371714,
      "kl": 0.240020751953125,
      "learning_rate": 4.977790908188239e-07,
      "loss": 0.0002,
      "reward": 1.7178572118282318,
      "reward_std": 0.09596449136734009,
      "rewards/equation_reward_func": 0.7357143089175224,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 9610
    },
    {
      "completion_length": 222.44197463989258,
      "epoch": 1.6115931095184206,
      "grad_norm": 0.3959098496736238,
      "kl": 0.192840576171875,
      "learning_rate": 4.977774951270869e-07,
      "loss": 0.0002,
      "reward": 1.739285796880722,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857559025288,
      "rewards/format_reward_func": 1.0,
      "step": 9612
    },
    {
      "completion_length": 225.5491180419922,
      "epoch": 1.6119284127582882,
      "grad_norm": 0.13520841983062826,
      "kl": 0.23187255859375,
      "learning_rate": 4.977758988648742e-07,
      "loss": 0.0002,
      "reward": 1.7321429178118706,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7321428786963224,
      "rewards/format_reward_func": 1.0,
      "step": 9614
    },
    {
      "completion_length": 219.13840293884277,
      "epoch": 1.6122637159981559,
      "grad_norm": 0.8712622346968198,
      "kl": 0.259063720703125,
      "learning_rate": 4.977743020321896e-07,
      "loss": 0.0003,
      "reward": 1.789285771548748,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 9616
    },
    {
      "completion_length": 221.70983219146729,
      "epoch": 1.6125990192380235,
      "grad_norm": 0.3392027587240292,
      "kl": 0.131103515625,
      "learning_rate": 4.977727046290365e-07,
      "loss": 0.0001,
      "reward": 1.7375000640749931,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643141329288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9618
    },
    {
      "completion_length": 222.52679634094238,
      "epoch": 1.612934322477891,
      "grad_norm": 0.29242676314966715,
      "kl": 0.13372802734375,
      "learning_rate": 4.977711066554189e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000447034836,
      "rewards/format_reward_func": 1.0,
      "step": 9620
    },
    {
      "completion_length": 231.80804634094238,
      "epoch": 1.6132696257177583,
      "grad_norm": 0.22772919308178813,
      "kl": 0.1455078125,
      "learning_rate": 4.977695081113403e-07,
      "loss": 0.0001,
      "reward": 1.7803572043776512,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7848214618861675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9622
    },
    {
      "completion_length": 215.53125953674316,
      "epoch": 1.613604928957626,
      "grad_norm": 0.24217956537165358,
      "kl": 0.1041412353515625,
      "learning_rate": 4.977679089968044e-07,
      "loss": 0.0001,
      "reward": 1.7678571864962578,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.767857164144516,
      "rewards/format_reward_func": 1.0,
      "step": 9624
    },
    {
      "completion_length": 227.67411613464355,
      "epoch": 1.6139402321974936,
      "grad_norm": 0.2160774633134719,
      "kl": 0.239593505859375,
      "learning_rate": 4.977663093118151e-07,
      "loss": 0.0002,
      "reward": 1.782142922282219,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 9626
    },
    {
      "completion_length": 230.6919755935669,
      "epoch": 1.6142755354373612,
      "grad_norm": 0.1998150112790896,
      "kl": 0.160888671875,
      "learning_rate": 4.977647090563757e-07,
      "loss": 0.0002,
      "reward": 1.7517857775092125,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500070780516,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9628
    },
    {
      "completion_length": 228.01786613464355,
      "epoch": 1.6146108386772289,
      "grad_norm": 0.22424239037497734,
      "kl": 0.160491943359375,
      "learning_rate": 4.977631082304901e-07,
      "loss": 0.0002,
      "reward": 1.7714286372065544,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 9630
    },
    {
      "completion_length": 232.89733409881592,
      "epoch": 1.6149461419170963,
      "grad_norm": 0.3348732755377922,
      "kl": 0.160919189453125,
      "learning_rate": 4.97761506834162e-07,
      "loss": 0.0002,
      "reward": 1.7446429058909416,
      "reward_std": 0.08838834706693888,
      "rewards/equation_reward_func": 0.7580357380211353,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 9632
    },
    {
      "completion_length": 234.50447368621826,
      "epoch": 1.6152814451569637,
      "grad_norm": 0.1766412362968214,
      "kl": 0.18402099609375,
      "learning_rate": 4.97759904867395e-07,
      "loss": 0.0002,
      "reward": 1.7321429252624512,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428917348385,
      "rewards/format_reward_func": 1.0,
      "step": 9634
    },
    {
      "completion_length": 230.77233123779297,
      "epoch": 1.6156167483968313,
      "grad_norm": 0.2087278214446919,
      "kl": 0.19451904296875,
      "learning_rate": 4.977583023301929e-07,
      "loss": 0.0002,
      "reward": 1.7035715132951736,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7035714648663998,
      "rewards/format_reward_func": 1.0,
      "step": 9636
    },
    {
      "completion_length": 227.9375123977661,
      "epoch": 1.615952051636699,
      "grad_norm": 0.7674398765859399,
      "kl": 0.294891357421875,
      "learning_rate": 4.977566992225594e-07,
      "loss": 0.0003,
      "reward": 1.816071480512619,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.820535734295845,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9638
    },
    {
      "completion_length": 231.5312614440918,
      "epoch": 1.6162873548765666,
      "grad_norm": 0.24075346358740746,
      "kl": 0.1507568359375,
      "learning_rate": 4.97755095544498e-07,
      "loss": 0.0002,
      "reward": 1.7892857789993286,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857305705547,
      "rewards/format_reward_func": 1.0,
      "step": 9640
    },
    {
      "completion_length": 237.66072463989258,
      "epoch": 1.616622658116434,
      "grad_norm": 0.2443153487966591,
      "kl": 0.151885986328125,
      "learning_rate": 4.977534912960124e-07,
      "loss": 0.0002,
      "reward": 1.698214367032051,
      "reward_std": 0.06313453428447247,
      "rewards/equation_reward_func": 0.7116071619093418,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 9642
    },
    {
      "completion_length": 236.14733028411865,
      "epoch": 1.6169579613563017,
      "grad_norm": 0.33337972372874886,
      "kl": 0.476715087890625,
      "learning_rate": 4.977518864771065e-07,
      "loss": 0.0005,
      "reward": 1.7375000938773155,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.741964302957058,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9644
    },
    {
      "completion_length": 229.7812614440918,
      "epoch": 1.617293264596169,
      "grad_norm": 0.16053063780435226,
      "kl": 0.09991455078125,
      "learning_rate": 4.97750281087784e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 9646
    },
    {
      "completion_length": 234.77679920196533,
      "epoch": 1.6176285678360367,
      "grad_norm": 0.21157833209685314,
      "kl": 0.130218505859375,
      "learning_rate": 4.977486751280484e-07,
      "loss": 0.0001,
      "reward": 1.8035714700818062,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 9648
    },
    {
      "completion_length": 234.44643878936768,
      "epoch": 1.6179638710759043,
      "grad_norm": 0.3738183027599362,
      "kl": 0.736419677734375,
      "learning_rate": 4.977470685979035e-07,
      "loss": 0.0007,
      "reward": 1.8035714700818062,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.8125000223517418,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9650
    },
    {
      "completion_length": 231.8482265472412,
      "epoch": 1.618299174315772,
      "grad_norm": 0.1965934925777346,
      "kl": 0.094573974609375,
      "learning_rate": 4.97745461497353e-07,
      "loss": 0.0001,
      "reward": 1.841071493923664,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8455357328057289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9652
    },
    {
      "completion_length": 231.94643783569336,
      "epoch": 1.6186344775556394,
      "grad_norm": 0.3680231256765474,
      "kl": 0.1709747314453125,
      "learning_rate": 4.977438538264006e-07,
      "loss": 0.0002,
      "reward": 1.8571428954601288,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8571428693830967,
      "rewards/format_reward_func": 1.0,
      "step": 9654
    },
    {
      "completion_length": 224.1741180419922,
      "epoch": 1.6189697807955068,
      "grad_norm": 0.32522419030659977,
      "kl": 0.321319580078125,
      "learning_rate": 4.9774224558505e-07,
      "loss": 0.0003,
      "reward": 1.7875000461935997,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7919643297791481,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9656
    },
    {
      "completion_length": 227.69643783569336,
      "epoch": 1.6193050840353744,
      "grad_norm": 0.27102926902278984,
      "kl": 0.31005859375,
      "learning_rate": 4.977406367733049e-07,
      "loss": 0.0003,
      "reward": 1.7321429327130318,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.732142886146903,
      "rewards/format_reward_func": 1.0,
      "step": 9658
    },
    {
      "completion_length": 236.2634048461914,
      "epoch": 1.619640387275242,
      "grad_norm": 0.16385341685718283,
      "kl": 1.131195068359375,
      "learning_rate": 4.977390273911689e-07,
      "loss": 0.0011,
      "reward": 1.7357143759727478,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 9660
    },
    {
      "completion_length": 234.70983123779297,
      "epoch": 1.6199756905151097,
      "grad_norm": 0.28091750128202797,
      "kl": 1.291107177734375,
      "learning_rate": 4.97737417438646e-07,
      "loss": 0.0013,
      "reward": 1.6535715013742447,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.6625000461935997,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9662
    },
    {
      "completion_length": 220.4017972946167,
      "epoch": 1.6203109937549771,
      "grad_norm": 0.33978137394630953,
      "kl": 0.361083984375,
      "learning_rate": 4.977358069157395e-07,
      "loss": 0.0004,
      "reward": 1.7714286372065544,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 9664
    },
    {
      "completion_length": 229.77233219146729,
      "epoch": 1.6206462969948447,
      "grad_norm": 0.6532627671050593,
      "kl": 0.758575439453125,
      "learning_rate": 4.977341958224535e-07,
      "loss": 0.0008,
      "reward": 1.725000075995922,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7339286003261805,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9666
    },
    {
      "completion_length": 228.43751049041748,
      "epoch": 1.6209816002347122,
      "grad_norm": 0.18385989770126193,
      "kl": 0.35162353515625,
      "learning_rate": 4.977325841587914e-07,
      "loss": 0.0004,
      "reward": 1.7808036357164383,
      "reward_std": 0.037249373737722635,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 9668
    },
    {
      "completion_length": 237.67411518096924,
      "epoch": 1.6213169034745798,
      "grad_norm": 0.24482785614371022,
      "kl": 0.98797607421875,
      "learning_rate": 4.97730971924757e-07,
      "loss": 0.001,
      "reward": 1.7500000447034836,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7589286230504513,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9670
    },
    {
      "completion_length": 224.8750114440918,
      "epoch": 1.6216522067144474,
      "grad_norm": 0.4871330431732284,
      "kl": 0.494415283203125,
      "learning_rate": 4.977293591203542e-07,
      "loss": 0.0005,
      "reward": 1.80892863124609,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.813392885029316,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9672
    },
    {
      "completion_length": 222.71875858306885,
      "epoch": 1.621987509954315,
      "grad_norm": 0.14262834413088796,
      "kl": 0.99700927734375,
      "learning_rate": 4.977277457455865e-07,
      "loss": 0.001,
      "reward": 1.725000075995922,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7250000424683094,
      "rewards/format_reward_func": 1.0,
      "step": 9674
    },
    {
      "completion_length": 223.46876049041748,
      "epoch": 1.6223228131941825,
      "grad_norm": 0.34430265186672526,
      "kl": 0.10821533203125,
      "learning_rate": 4.977261318004576e-07,
      "loss": 0.0001,
      "reward": 1.7214286476373672,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7214286103844643,
      "rewards/format_reward_func": 1.0,
      "step": 9676
    },
    {
      "completion_length": 232.05804634094238,
      "epoch": 1.62265811643405,
      "grad_norm": 0.20829203027319398,
      "kl": 0.208831787109375,
      "learning_rate": 4.977245172849714e-07,
      "loss": 0.0002,
      "reward": 1.764285795390606,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 9678
    },
    {
      "completion_length": 227.8616180419922,
      "epoch": 1.6229934196739175,
      "grad_norm": 0.2671142585139483,
      "kl": 0.1011962890625,
      "learning_rate": 4.977229021991315e-07,
      "loss": 0.0001,
      "reward": 1.7625000551342964,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643051922321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9680
    },
    {
      "completion_length": 230.9955472946167,
      "epoch": 1.6233287229137852,
      "grad_norm": 0.2495018380150175,
      "kl": 0.2869873046875,
      "learning_rate": 4.977212865429416e-07,
      "loss": 0.0003,
      "reward": 1.778571493923664,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 9682
    },
    {
      "completion_length": 219.17411613464355,
      "epoch": 1.6236640261536528,
      "grad_norm": 0.11882449677412509,
      "kl": 0.189483642578125,
      "learning_rate": 4.977196703164055e-07,
      "loss": 0.0002,
      "reward": 1.7446429282426834,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7491071633994579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9684
    },
    {
      "completion_length": 220.46429443359375,
      "epoch": 1.6239993293935204,
      "grad_norm": 0.1870383835898375,
      "kl": 0.11187744140625,
      "learning_rate": 4.977180535195268e-07,
      "loss": 0.0001,
      "reward": 1.7392857745289803,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7392857447266579,
      "rewards/format_reward_func": 1.0,
      "step": 9686
    },
    {
      "completion_length": 225.55358219146729,
      "epoch": 1.6243346326333878,
      "grad_norm": 0.19740042234481134,
      "kl": 0.11981201171875,
      "learning_rate": 4.977164361523093e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 9688
    },
    {
      "completion_length": 224.74554538726807,
      "epoch": 1.6246699358732553,
      "grad_norm": 0.2529255736949469,
      "kl": 0.167327880859375,
      "learning_rate": 4.977148182147567e-07,
      "loss": 0.0002,
      "reward": 1.789285771548748,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857268452644,
      "rewards/format_reward_func": 1.0,
      "step": 9690
    },
    {
      "completion_length": 231.6696538925171,
      "epoch": 1.625005239113123,
      "grad_norm": 0.1329825256542032,
      "kl": 0.29461669921875,
      "learning_rate": 4.977131997068729e-07,
      "loss": 0.0003,
      "reward": 1.7714286297559738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 9692
    },
    {
      "completion_length": 218.3125114440918,
      "epoch": 1.6253405423529905,
      "grad_norm": 0.0972662980687718,
      "kl": 0.149688720703125,
      "learning_rate": 4.977115806286613e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 9694
    },
    {
      "completion_length": 217.821439743042,
      "epoch": 1.6256758455928582,
      "grad_norm": 0.3084129310617149,
      "kl": 0.125244140625,
      "learning_rate": 4.977099609801259e-07,
      "loss": 0.0001,
      "reward": 1.776785783469677,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.781250037252903,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9696
    },
    {
      "completion_length": 221.05358123779297,
      "epoch": 1.6260111488327256,
      "grad_norm": 0.2154034072950012,
      "kl": 0.092742919921875,
      "learning_rate": 4.977083407612702e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 9698
    },
    {
      "completion_length": 222.59822368621826,
      "epoch": 1.626346452072593,
      "grad_norm": 0.1480951218735697,
      "kl": 0.094696044921875,
      "learning_rate": 4.977067199720981e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 9700
    },
    {
      "completion_length": 225.29465293884277,
      "epoch": 1.6266817553124606,
      "grad_norm": 0.25106638423487115,
      "kl": 0.103851318359375,
      "learning_rate": 4.977050986126134e-07,
      "loss": 0.0001,
      "reward": 1.764285795390606,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 9702
    },
    {
      "completion_length": 220.85268783569336,
      "epoch": 1.6270170585523283,
      "grad_norm": 0.2751909383121962,
      "kl": 0.1004638671875,
      "learning_rate": 4.977034766828195e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 9704
    },
    {
      "completion_length": 222.95983028411865,
      "epoch": 1.627352361792196,
      "grad_norm": 0.34442896985833393,
      "kl": 0.0963592529296875,
      "learning_rate": 4.977018541827206e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 1.0,
      "step": 9706
    },
    {
      "completion_length": 232.1071538925171,
      "epoch": 1.6276876650320635,
      "grad_norm": 0.1972980182956902,
      "kl": 0.1473541259765625,
      "learning_rate": 4.977002311123201e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 9708
    },
    {
      "completion_length": 232.39286613464355,
      "epoch": 1.628022968271931,
      "grad_norm": 0.18802488475716975,
      "kl": 0.1038055419921875,
      "learning_rate": 4.976986074716217e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 9710
    },
    {
      "completion_length": 227.821439743042,
      "epoch": 1.6283582715117983,
      "grad_norm": 0.14353477227332534,
      "kl": 0.1986083984375,
      "learning_rate": 4.976969832606295e-07,
      "loss": 0.0002,
      "reward": 1.776785783469677,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7812500186264515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9712
    },
    {
      "completion_length": 223.3616180419922,
      "epoch": 1.628693574751666,
      "grad_norm": 0.002946186467856951,
      "kl": 0.0805816650390625,
      "learning_rate": 4.976953584793469e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 9714
    },
    {
      "completion_length": 221.93750858306885,
      "epoch": 1.6290288779915336,
      "grad_norm": 0.2969509883521366,
      "kl": 0.1015625,
      "learning_rate": 4.976937331277777e-07,
      "loss": 0.0001,
      "reward": 1.8107143267989159,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8107143007218838,
      "rewards/format_reward_func": 1.0,
      "step": 9716
    },
    {
      "completion_length": 228.8884038925171,
      "epoch": 1.6293641812314013,
      "grad_norm": 0.18499841231582298,
      "kl": 0.1175689697265625,
      "learning_rate": 4.976921072059256e-07,
      "loss": 0.0001,
      "reward": 1.8000000417232513,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.800000037997961,
      "rewards/format_reward_func": 1.0,
      "step": 9718
    },
    {
      "completion_length": 235.00001049041748,
      "epoch": 1.6296994844712687,
      "grad_norm": 0.14542382371891566,
      "kl": 0.104278564453125,
      "learning_rate": 4.976904807137947e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 1.0,
      "step": 9720
    },
    {
      "completion_length": 235.32143878936768,
      "epoch": 1.6300347877111363,
      "grad_norm": 0.23501990256504454,
      "kl": 0.1071014404296875,
      "learning_rate": 4.976888536513883e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 9722
    },
    {
      "completion_length": 230.77232933044434,
      "epoch": 1.6303700909510037,
      "grad_norm": 0.22814759709441632,
      "kl": 0.094635009765625,
      "learning_rate": 4.976872260187104e-07,
      "loss": 0.0001,
      "reward": 1.7428572326898575,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 9724
    },
    {
      "completion_length": 228.00447463989258,
      "epoch": 1.6307053941908713,
      "grad_norm": 0.31272582817111516,
      "kl": 0.10003662109375,
      "learning_rate": 4.976855978157646e-07,
      "loss": 0.0001,
      "reward": 1.741964340209961,
      "reward_std": 0.09217641782015562,
      "rewards/equation_reward_func": 0.7482143230736256,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 9726
    },
    {
      "completion_length": 225.5000114440918,
      "epoch": 1.631040697430739,
      "grad_norm": 0.23990043104025616,
      "kl": 0.333709716796875,
      "learning_rate": 4.976839690425547e-07,
      "loss": 0.0003,
      "reward": 1.725000075995922,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000387430191,
      "rewards/format_reward_func": 1.0,
      "step": 9728
    },
    {
      "completion_length": 226.41072368621826,
      "epoch": 1.6313760006706066,
      "grad_norm": 0.20449317919171378,
      "kl": 0.1821441650390625,
      "learning_rate": 4.976823396990845e-07,
      "loss": 0.0002,
      "reward": 1.7839286178350449,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7883928790688515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9730
    },
    {
      "completion_length": 224.50447273254395,
      "epoch": 1.631711303910474,
      "grad_norm": 0.13556935221852442,
      "kl": 0.160125732421875,
      "learning_rate": 4.976807097853577e-07,
      "loss": 0.0002,
      "reward": 1.7892857789993286,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 9732
    },
    {
      "completion_length": 229.99554634094238,
      "epoch": 1.6320466071503414,
      "grad_norm": 0.24666051030834293,
      "kl": 0.097808837890625,
      "learning_rate": 4.976790793013781e-07,
      "loss": 0.0001,
      "reward": 1.7339286357164383,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.738392885774374,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9734
    },
    {
      "completion_length": 221.28572463989258,
      "epoch": 1.632381910390209,
      "grad_norm": 0.2695799750218758,
      "kl": 0.0812835693359375,
      "learning_rate": 4.976774482471494e-07,
      "loss": 0.0001,
      "reward": 1.7696429342031479,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9736
    },
    {
      "completion_length": 226.5357255935669,
      "epoch": 1.6327172136300767,
      "grad_norm": 0.301846569929245,
      "kl": 0.2186279296875,
      "learning_rate": 4.976758166226755e-07,
      "loss": 0.0002,
      "reward": 1.741071492433548,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7455357499420643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9738
    },
    {
      "completion_length": 222.57590198516846,
      "epoch": 1.6330525168699443,
      "grad_norm": 0.17720910829776573,
      "kl": 0.168731689453125,
      "learning_rate": 4.9767418442796e-07,
      "loss": 0.0002,
      "reward": 1.7464286386966705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 9740
    },
    {
      "completion_length": 227.6696548461914,
      "epoch": 1.6333878201098118,
      "grad_norm": 0.23504752245686839,
      "kl": 0.344696044921875,
      "learning_rate": 4.976725516630065e-07,
      "loss": 0.0003,
      "reward": 1.8250000551342964,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8250000216066837,
      "rewards/format_reward_func": 1.0,
      "step": 9742
    },
    {
      "completion_length": 219.54465293884277,
      "epoch": 1.6337231233496794,
      "grad_norm": 0.26755522942864296,
      "kl": 0.0870819091796875,
      "learning_rate": 4.976709183278192e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 9744
    },
    {
      "completion_length": 223.22768878936768,
      "epoch": 1.6340584265895468,
      "grad_norm": 0.3647382535741721,
      "kl": 0.13238525390625,
      "learning_rate": 4.976692844224014e-07,
      "loss": 0.0001,
      "reward": 1.7178572416305542,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7178571820259094,
      "rewards/format_reward_func": 1.0,
      "step": 9746
    },
    {
      "completion_length": 222.90179538726807,
      "epoch": 1.6343937298294144,
      "grad_norm": 0.23603635934814576,
      "kl": 0.084197998046875,
      "learning_rate": 4.976676499467573e-07,
      "loss": 0.0001,
      "reward": 1.7125000730156898,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7169642969965935,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9748
    },
    {
      "completion_length": 223.4776906967163,
      "epoch": 1.634729033069282,
      "grad_norm": 0.17505501443511048,
      "kl": 0.0858306884765625,
      "learning_rate": 4.976660149008903e-07,
      "loss": 0.0001,
      "reward": 1.8107143342494965,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 9750
    },
    {
      "completion_length": 226.9910831451416,
      "epoch": 1.6350643363091497,
      "grad_norm": 0.16053011603624132,
      "kl": 0.1113739013671875,
      "learning_rate": 4.976643792848043e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143245637417,
      "rewards/format_reward_func": 1.0,
      "step": 9752
    },
    {
      "completion_length": 229.87054634094238,
      "epoch": 1.6353996395490171,
      "grad_norm": 0.20286619977289153,
      "kl": 0.1265869140625,
      "learning_rate": 4.976627430985031e-07,
      "loss": 0.0001,
      "reward": 1.7232143357396126,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7276786062866449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9754
    },
    {
      "completion_length": 229.2366180419922,
      "epoch": 1.6357349427888845,
      "grad_norm": 0.2432832648509416,
      "kl": 0.539398193359375,
      "learning_rate": 4.976611063419906e-07,
      "loss": 0.0005,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 9756
    },
    {
      "completion_length": 225.43304538726807,
      "epoch": 1.6360702460287522,
      "grad_norm": 0.23562147245432455,
      "kl": 0.093414306640625,
      "learning_rate": 4.976594690152702e-07,
      "loss": 0.0001,
      "reward": 1.742857187986374,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7517857514321804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9758
    },
    {
      "completion_length": 226.06697368621826,
      "epoch": 1.6364055492686198,
      "grad_norm": 0.2776689340708705,
      "kl": 0.14593505859375,
      "learning_rate": 4.976578311183459e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 9760
    },
    {
      "completion_length": 229.14732933044434,
      "epoch": 1.6367408525084874,
      "grad_norm": 0.2487283953788348,
      "kl": 0.4114837646484375,
      "learning_rate": 4.976561926512215e-07,
      "loss": 0.0004,
      "reward": 1.7625000476837158,
      "reward_std": 0.09343910776078701,
      "rewards/equation_reward_func": 0.7669643219560385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9762
    },
    {
      "completion_length": 231.17411708831787,
      "epoch": 1.637076155748355,
      "grad_norm": 0.2730592489396775,
      "kl": 0.090362548828125,
      "learning_rate": 4.976545536139007e-07,
      "loss": 0.0001,
      "reward": 1.7250000685453415,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7250000387430191,
      "rewards/format_reward_func": 1.0,
      "step": 9764
    },
    {
      "completion_length": 224.81250858306885,
      "epoch": 1.6374114589882225,
      "grad_norm": 0.2095179396768771,
      "kl": 0.0867919921875,
      "learning_rate": 4.976529140063874e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.785714328289032,
      "rewards/format_reward_func": 1.0,
      "step": 9766
    },
    {
      "completion_length": 239.69197463989258,
      "epoch": 1.63774676222809,
      "grad_norm": 0.4736853723651829,
      "kl": 0.27679443359375,
      "learning_rate": 4.976512738286851e-07,
      "loss": 0.0003,
      "reward": 1.7535715028643608,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.7625000402331352,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9768
    },
    {
      "completion_length": 231.71429634094238,
      "epoch": 1.6380820654679575,
      "grad_norm": 0.12591280961738255,
      "kl": 0.321502685546875,
      "learning_rate": 4.976496330807978e-07,
      "loss": 0.0003,
      "reward": 1.8214286118745804,
      "reward_std": 0.03030457627028227,
      "rewards/equation_reward_func": 0.8303571566939354,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9770
    },
    {
      "completion_length": 231.53126049041748,
      "epoch": 1.6384173687078252,
      "grad_norm": 0.23487225655647087,
      "kl": 0.55450439453125,
      "learning_rate": 4.976479917627292e-07,
      "loss": 0.0006,
      "reward": 1.7683036401867867,
      "reward_std": 0.05492704268544912,
      "rewards/equation_reward_func": 0.7741071730852127,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 9772
    },
    {
      "completion_length": 226.21429634094238,
      "epoch": 1.6387526719476928,
      "grad_norm": 0.26357185990527765,
      "kl": 0.21807861328125,
      "learning_rate": 4.976463498744832e-07,
      "loss": 0.0002,
      "reward": 1.7232143729925156,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7276786081492901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9774
    },
    {
      "completion_length": 231.37501335144043,
      "epoch": 1.6390879751875602,
      "grad_norm": 0.2148319851533725,
      "kl": 0.1714019775390625,
      "learning_rate": 4.976447074160634e-07,
      "loss": 0.0002,
      "reward": 1.7946429327130318,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7991071604192257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9776
    },
    {
      "completion_length": 234.99108123779297,
      "epoch": 1.6394232784274279,
      "grad_norm": 0.0844546691876832,
      "kl": 0.27728271484375,
      "learning_rate": 4.976430643874737e-07,
      "loss": 0.0003,
      "reward": 1.7178572192788124,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7178571745753288,
      "rewards/format_reward_func": 1.0,
      "step": 9778
    },
    {
      "completion_length": 239.3303689956665,
      "epoch": 1.6397585816672953,
      "grad_norm": 0.35980099021948647,
      "kl": 0.845733642578125,
      "learning_rate": 4.976414207887178e-07,
      "loss": 0.0008,
      "reward": 1.7589286416769028,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7633928991854191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9780
    },
    {
      "completion_length": 231.68304634094238,
      "epoch": 1.640093884907163,
      "grad_norm": 0.16832908265496788,
      "kl": 0.0909423828125,
      "learning_rate": 4.976397766197996e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714440047741,
      "rewards/format_reward_func": 1.0,
      "step": 9782
    },
    {
      "completion_length": 231.8839406967163,
      "epoch": 1.6404291881470305,
      "grad_norm": 0.2077517323854166,
      "kl": 0.097900390625,
      "learning_rate": 4.976381318807228e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714286111295223,
      "rewards/format_reward_func": 1.0,
      "step": 9784
    },
    {
      "completion_length": 235.29019165039062,
      "epoch": 1.6407644913868982,
      "grad_norm": 0.1573480194081336,
      "kl": 0.0998077392578125,
      "learning_rate": 4.976364865714911e-07,
      "loss": 0.0001,
      "reward": 1.8017857670783997,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8062500394880772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9786
    },
    {
      "completion_length": 247.6250114440918,
      "epoch": 1.6410997946267656,
      "grad_norm": 0.24239820048925825,
      "kl": 0.5035400390625,
      "learning_rate": 4.976348406921085e-07,
      "loss": 0.0005,
      "reward": 1.7767857685685158,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7812500298023224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9788
    },
    {
      "completion_length": 246.48215579986572,
      "epoch": 1.641435097866633,
      "grad_norm": 0.2469812162663503,
      "kl": 0.15966796875,
      "learning_rate": 4.976331942425786e-07,
      "loss": 0.0002,
      "reward": 1.7553572207689285,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7598214596509933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9790
    },
    {
      "completion_length": 236.31251049041748,
      "epoch": 1.6417704011065006,
      "grad_norm": 0.18019076628164507,
      "kl": 0.4852447509765625,
      "learning_rate": 4.976315472229054e-07,
      "loss": 0.0005,
      "reward": 1.7321429327130318,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321428880095482,
      "rewards/format_reward_func": 1.0,
      "step": 9792
    },
    {
      "completion_length": 232.93304634094238,
      "epoch": 1.6421057043463683,
      "grad_norm": 0.17304451156624187,
      "kl": 0.187408447265625,
      "learning_rate": 4.976298996330925e-07,
      "loss": 0.0002,
      "reward": 1.7767858058214188,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7812500335276127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9794
    },
    {
      "completion_length": 238.790189743042,
      "epoch": 1.642441007586236,
      "grad_norm": 0.7010478912700674,
      "kl": 1.2734222412109375,
      "learning_rate": 4.976282514731437e-07,
      "loss": 0.0013,
      "reward": 1.6839286386966705,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.6883929129689932,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9796
    },
    {
      "completion_length": 254.2901906967163,
      "epoch": 1.6427763108261033,
      "grad_norm": 0.16518823323916085,
      "kl": 0.342803955078125,
      "learning_rate": 4.976266027430629e-07,
      "loss": 0.0003,
      "reward": 1.7464286386966705,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.746428607031703,
      "rewards/format_reward_func": 1.0,
      "step": 9798
    },
    {
      "completion_length": 248.65625953674316,
      "epoch": 1.643111614065971,
      "grad_norm": 0.2258722497635101,
      "kl": 0.151885986328125,
      "learning_rate": 4.976249534428539e-07,
      "loss": 0.0002,
      "reward": 1.7642857804894447,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7732143141329288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9800
    },
    {
      "completion_length": 241.86161994934082,
      "epoch": 1.6434469173058384,
      "grad_norm": 0.21150592179344968,
      "kl": 0.124969482421875,
      "learning_rate": 4.976233035725203e-07,
      "loss": 0.0001,
      "reward": 1.7767857983708382,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7812500186264515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9802
    },
    {
      "completion_length": 239.51786613464355,
      "epoch": 1.643782220545706,
      "grad_norm": 0.23342208363613637,
      "kl": 0.11627197265625,
      "learning_rate": 4.976216531320662e-07,
      "loss": 0.0001,
      "reward": 1.7035715207457542,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.712500024586916,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9804
    },
    {
      "completion_length": 243.0803689956665,
      "epoch": 1.6441175237855736,
      "grad_norm": 0.14752362469685548,
      "kl": 0.207550048828125,
      "learning_rate": 4.976200021214952e-07,
      "loss": 0.0002,
      "reward": 1.7678572088479996,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7767857424914837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9806
    },
    {
      "completion_length": 238.94197368621826,
      "epoch": 1.6444528270254413,
      "grad_norm": 0.23430527663124873,
      "kl": 0.1428680419921875,
      "learning_rate": 4.97618350540811e-07,
      "loss": 0.0001,
      "reward": 1.7625000774860382,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643051922321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9808
    },
    {
      "completion_length": 239.4062614440918,
      "epoch": 1.6447881302653087,
      "grad_norm": 2.2863623227606498,
      "kl": 0.110076904296875,
      "learning_rate": 4.976166983900177e-07,
      "loss": 0.0001,
      "reward": 1.7339286655187607,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7383928783237934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9810
    },
    {
      "completion_length": 236.11608123779297,
      "epoch": 1.645123433505176,
      "grad_norm": 0.1577864150084241,
      "kl": 0.1080169677734375,
      "learning_rate": 4.976150456691189e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.035355339758098125,
      "rewards/equation_reward_func": 0.7696428932249546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9812
    },
    {
      "completion_length": 229.29465293884277,
      "epoch": 1.6454587367450437,
      "grad_norm": 0.23221787032821553,
      "kl": 0.15777587890625,
      "learning_rate": 4.976133923781186e-07,
      "loss": 0.0002,
      "reward": 1.7785714864730835,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7875000163912773,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9814
    },
    {
      "completion_length": 246.70090579986572,
      "epoch": 1.6457940399849114,
      "grad_norm": 0.3261517870491476,
      "kl": 0.225372314453125,
      "learning_rate": 4.976117385170204e-07,
      "loss": 0.0002,
      "reward": 1.7928571924567223,
      "reward_std": 0.06060915347188711,
      "rewards/equation_reward_func": 0.8017857521772385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9816
    },
    {
      "completion_length": 224.31250858306885,
      "epoch": 1.646129343224779,
      "grad_norm": 0.4089467330222451,
      "kl": 0.14581298828125,
      "learning_rate": 4.97610084085828e-07,
      "loss": 0.0001,
      "reward": 1.8500000834465027,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8500000052154064,
      "rewards/format_reward_func": 1.0,
      "step": 9818
    },
    {
      "completion_length": 231.48661994934082,
      "epoch": 1.6464646464646466,
      "grad_norm": 0.24769592529913995,
      "kl": 0.213653564453125,
      "learning_rate": 4.976084290845455e-07,
      "loss": 0.0002,
      "reward": 1.7982143461704254,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.8116071671247482,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 9820
    },
    {
      "completion_length": 229.10715198516846,
      "epoch": 1.646799949704514,
      "grad_norm": 0.23369410087790052,
      "kl": 0.16387939453125,
      "learning_rate": 4.976067735131766e-07,
      "loss": 0.0002,
      "reward": 1.7517857626080513,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7562500312924385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9822
    },
    {
      "completion_length": 237.2142972946167,
      "epoch": 1.6471352529443815,
      "grad_norm": 0.16716198794533854,
      "kl": 0.13714599609375,
      "learning_rate": 4.976051173717251e-07,
      "loss": 0.0001,
      "reward": 1.7696429193019867,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.774107176810503,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9824
    },
    {
      "completion_length": 235.5580472946167,
      "epoch": 1.647470556184249,
      "grad_norm": 0.26337107233516976,
      "kl": 0.1463623046875,
      "learning_rate": 4.976034606601948e-07,
      "loss": 0.0001,
      "reward": 1.8160714581608772,
      "reward_std": 0.07828682195395231,
      "rewards/equation_reward_func": 0.8294643089175224,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 9826
    },
    {
      "completion_length": 233.915189743042,
      "epoch": 1.6478058594241167,
      "grad_norm": 0.5915121442371221,
      "kl": 0.158599853515625,
      "learning_rate": 4.976018033785895e-07,
      "loss": 0.0002,
      "reward": 1.7642857879400253,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 9828
    },
    {
      "completion_length": 244.946439743042,
      "epoch": 1.6481411626639844,
      "grad_norm": 0.22913292433146903,
      "kl": 0.179351806640625,
      "learning_rate": 4.976001455269129e-07,
      "loss": 0.0002,
      "reward": 1.7357143685221672,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7446429021656513,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9830
    },
    {
      "completion_length": 234.47322368621826,
      "epoch": 1.6484764659038518,
      "grad_norm": 0.20752086023371275,
      "kl": 0.12939453125,
      "learning_rate": 4.975984871051692e-07,
      "loss": 0.0001,
      "reward": 1.7803571969270706,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7848214469850063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9832
    },
    {
      "completion_length": 224.13393688201904,
      "epoch": 1.6488117691437192,
      "grad_norm": 0.19172764728090289,
      "kl": 0.1249542236328125,
      "learning_rate": 4.97596828113362e-07,
      "loss": 0.0001,
      "reward": 1.764285758137703,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7821428868919611,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 9834
    },
    {
      "completion_length": 231.31697463989258,
      "epoch": 1.6491470723835868,
      "grad_norm": 0.2749263834024471,
      "kl": 0.183258056640625,
      "learning_rate": 4.97595168551495e-07,
      "loss": 0.0002,
      "reward": 1.7928572073578835,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.8017857410013676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9836
    },
    {
      "completion_length": 238.1696548461914,
      "epoch": 1.6494823756234545,
      "grad_norm": 0.1818458948457541,
      "kl": 0.223663330078125,
      "learning_rate": 4.975935084195721e-07,
      "loss": 0.0002,
      "reward": 1.757142923772335,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7660714574158192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9838
    },
    {
      "completion_length": 240.446439743042,
      "epoch": 1.649817678863322,
      "grad_norm": 0.2215274701890657,
      "kl": 0.308563232421875,
      "learning_rate": 4.975918477175972e-07,
      "loss": 0.0003,
      "reward": 1.7482143342494965,
      "reward_std": 0.07828682195395231,
      "rewards/equation_reward_func": 0.7705357447266579,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 9840
    },
    {
      "completion_length": 229.79911613464355,
      "epoch": 1.6501529821031897,
      "grad_norm": 0.162230038132875,
      "kl": 0.106109619140625,
      "learning_rate": 4.975901864455739e-07,
      "loss": 0.0001,
      "reward": 1.7946429178118706,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7991071790456772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9842
    },
    {
      "completion_length": 248.53126430511475,
      "epoch": 1.6504882853430571,
      "grad_norm": 0.32201021323163814,
      "kl": 0.21990966796875,
      "learning_rate": 4.975885246035064e-07,
      "loss": 0.0002,
      "reward": 1.7267857640981674,
      "reward_std": 0.10354063473641872,
      "rewards/equation_reward_func": 0.7491071671247482,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 9844
    },
    {
      "completion_length": 246.05804824829102,
      "epoch": 1.6508235885829246,
      "grad_norm": 0.3693541503025461,
      "kl": 0.27056884765625,
      "learning_rate": 4.975868621913983e-07,
      "loss": 0.0003,
      "reward": 1.7410714775323868,
      "reward_std": 0.08333758544176817,
      "rewards/equation_reward_func": 0.7633928917348385,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 9846
    },
    {
      "completion_length": 246.50447845458984,
      "epoch": 1.6511588918227922,
      "grad_norm": 0.7884187970758376,
      "kl": 0.19110107421875,
      "learning_rate": 4.975851992092533e-07,
      "loss": 0.0002,
      "reward": 1.6910715103149414,
      "reward_std": 0.08838834706693888,
      "rewards/equation_reward_func": 0.7223214618861675,
      "rewards/format_reward_func": 0.9687500074505806,
      "step": 9848
    },
    {
      "completion_length": 238.7410831451416,
      "epoch": 1.6514941950626598,
      "grad_norm": 0.256588565457127,
      "kl": 0.242462158203125,
      "learning_rate": 4.975835356570755e-07,
      "loss": 0.0002,
      "reward": 1.78035718947649,
      "reward_std": 0.07828682195395231,
      "rewards/equation_reward_func": 0.8026785962283611,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 9850
    },
    {
      "completion_length": 237.477689743042,
      "epoch": 1.6518294983025275,
      "grad_norm": 0.22375889051073486,
      "kl": 0.1295928955078125,
      "learning_rate": 4.975818715348686e-07,
      "loss": 0.0001,
      "reward": 1.71785718947649,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7357143145054579,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 9852
    },
    {
      "completion_length": 221.727689743042,
      "epoch": 1.6521648015423949,
      "grad_norm": 0.2291120045991947,
      "kl": 0.1092376708984375,
      "learning_rate": 4.975802068426364e-07,
      "loss": 0.0001,
      "reward": 1.7625000774860382,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643312692642,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9854
    },
    {
      "completion_length": 223.55804443359375,
      "epoch": 1.6525001047822625,
      "grad_norm": 0.26357104191843117,
      "kl": 0.112152099609375,
      "learning_rate": 4.975785415803828e-07,
      "loss": 0.0001,
      "reward": 1.7625000551342964,
      "reward_std": 0.0833375845104456,
      "rewards/equation_reward_func": 0.7758928947150707,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 9856
    },
    {
      "completion_length": 221.83483219146729,
      "epoch": 1.65283540802213,
      "grad_norm": 0.5731160641229195,
      "kl": 0.19775390625,
      "learning_rate": 4.975768757481115e-07,
      "loss": 0.0002,
      "reward": 1.7892857491970062,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.798214316368103,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9858
    },
    {
      "completion_length": 229.1562623977661,
      "epoch": 1.6531707112619975,
      "grad_norm": 0.2752841675439393,
      "kl": 0.270050048828125,
      "learning_rate": 4.975752093458266e-07,
      "loss": 0.0003,
      "reward": 1.7553572058677673,
      "reward_std": 0.10354063659906387,
      "rewards/equation_reward_func": 0.7687500193715096,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 9860
    },
    {
      "completion_length": 213.55358123779297,
      "epoch": 1.6535060145018652,
      "grad_norm": 0.3404765282706294,
      "kl": 0.1291351318359375,
      "learning_rate": 4.975735423735316e-07,
      "loss": 0.0001,
      "reward": 1.801785759627819,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8062500320374966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9862
    },
    {
      "completion_length": 221.33483028411865,
      "epoch": 1.6538413177417328,
      "grad_norm": 0.19012431000183802,
      "kl": 0.151336669921875,
      "learning_rate": 4.975718748312306e-07,
      "loss": 0.0002,
      "reward": 1.7892857268452644,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7982143238186836,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9864
    },
    {
      "completion_length": 222.90179634094238,
      "epoch": 1.6541766209816002,
      "grad_norm": 0.004469077198342189,
      "kl": 0.235565185546875,
      "learning_rate": 4.975702067189274e-07,
      "loss": 0.0002,
      "reward": 1.7500000596046448,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000447034836,
      "rewards/format_reward_func": 1.0,
      "step": 9866
    },
    {
      "completion_length": 221.87054252624512,
      "epoch": 1.6545119242214676,
      "grad_norm": 0.5435280581221478,
      "kl": 0.3485107421875,
      "learning_rate": 4.975685380366257e-07,
      "loss": 0.0003,
      "reward": 1.739285796880722,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7392857447266579,
      "rewards/format_reward_func": 1.0,
      "step": 9868
    },
    {
      "completion_length": 224.46429538726807,
      "epoch": 1.6548472274613353,
      "grad_norm": 0.20865656624230186,
      "kl": 0.588104248046875,
      "learning_rate": 4.975668687843295e-07,
      "loss": 0.0006,
      "reward": 1.7625000551342964,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9870
    },
    {
      "completion_length": 219.540189743042,
      "epoch": 1.655182530701203,
      "grad_norm": 0.308684592760428,
      "kl": 0.479217529296875,
      "learning_rate": 4.975651989620425e-07,
      "loss": 0.0005,
      "reward": 1.7250000461935997,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7250000350177288,
      "rewards/format_reward_func": 1.0,
      "step": 9872
    },
    {
      "completion_length": 219.31697463989258,
      "epoch": 1.6555178339410705,
      "grad_norm": 0.21206860929059224,
      "kl": 0.5569000244140625,
      "learning_rate": 4.975635285697687e-07,
      "loss": 0.0006,
      "reward": 1.853571467101574,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8535714522004128,
      "rewards/format_reward_func": 1.0,
      "step": 9874
    },
    {
      "completion_length": 227.2321538925171,
      "epoch": 1.655853137180938,
      "grad_norm": 0.24630201716195602,
      "kl": 0.1265411376953125,
      "learning_rate": 4.975618576075119e-07,
      "loss": 0.0001,
      "reward": 1.7267858013510704,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7312500216066837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9876
    },
    {
      "completion_length": 219.62947368621826,
      "epoch": 1.6561884404208056,
      "grad_norm": 0.23565190769001568,
      "kl": 0.22161865234375,
      "learning_rate": 4.975601860752758e-07,
      "loss": 0.0002,
      "reward": 1.778571493923664,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.778571454808116,
      "rewards/format_reward_func": 1.0,
      "step": 9878
    },
    {
      "completion_length": 220.07590198516846,
      "epoch": 1.656523743660673,
      "grad_norm": 0.20121385950706266,
      "kl": 0.21588134765625,
      "learning_rate": 4.975585139730644e-07,
      "loss": 0.0002,
      "reward": 1.7892857640981674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 9880
    },
    {
      "completion_length": 229.6696538925171,
      "epoch": 1.6568590469005406,
      "grad_norm": 0.30524107494236097,
      "kl": 0.0986175537109375,
      "learning_rate": 4.975568413008816e-07,
      "loss": 0.0001,
      "reward": 1.7892857566475868,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7892857491970062,
      "rewards/format_reward_func": 1.0,
      "step": 9882
    },
    {
      "completion_length": 224.22322273254395,
      "epoch": 1.6571943501404083,
      "grad_norm": 0.3024153219313286,
      "kl": 0.0926361083984375,
      "learning_rate": 4.97555168058731e-07,
      "loss": 0.0001,
      "reward": 1.7375000640749931,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7419643364846706,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9884
    },
    {
      "completion_length": 218.5178689956665,
      "epoch": 1.657529653380276,
      "grad_norm": 0.2697534224150527,
      "kl": 0.094818115234375,
      "learning_rate": 4.975534942466168e-07,
      "loss": 0.0001,
      "reward": 1.7892857640981674,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857491970062,
      "rewards/format_reward_func": 1.0,
      "step": 9886
    },
    {
      "completion_length": 223.04911613464355,
      "epoch": 1.6578649566201433,
      "grad_norm": 0.1835474696257643,
      "kl": 0.1168365478515625,
      "learning_rate": 4.975518198645425e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 9888
    },
    {
      "completion_length": 222.79018878936768,
      "epoch": 1.6582002598600107,
      "grad_norm": 0.1563539322514873,
      "kl": 0.186553955078125,
      "learning_rate": 4.975501449125122e-07,
      "loss": 0.0002,
      "reward": 1.8107143193483353,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8107143193483353,
      "rewards/format_reward_func": 1.0,
      "step": 9890
    },
    {
      "completion_length": 224.33483219146729,
      "epoch": 1.6585355630998784,
      "grad_norm": 0.29091891567336037,
      "kl": 0.1243438720703125,
      "learning_rate": 4.975484693905298e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 9892
    },
    {
      "completion_length": 225.27233123779297,
      "epoch": 1.658870866339746,
      "grad_norm": 0.37451253496480214,
      "kl": 0.140167236328125,
      "learning_rate": 4.975467932985989e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7678571827709675,
      "rewards/format_reward_func": 1.0,
      "step": 9894
    },
    {
      "completion_length": 237.70536708831787,
      "epoch": 1.6592061695796136,
      "grad_norm": 0.268947938153285,
      "kl": 0.9615478515625,
      "learning_rate": 4.975451166367235e-07,
      "loss": 0.001,
      "reward": 1.8000000566244125,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 9896
    },
    {
      "completion_length": 233.5982255935669,
      "epoch": 1.6595414728194813,
      "grad_norm": 0.2700866753770015,
      "kl": 0.112548828125,
      "learning_rate": 4.975434394049075e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 9898
    },
    {
      "completion_length": 233.76340198516846,
      "epoch": 1.6598767760593487,
      "grad_norm": 0.2781299469560648,
      "kl": 0.2715911865234375,
      "learning_rate": 4.975417616031547e-07,
      "loss": 0.0003,
      "reward": 1.7678572088479996,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 9900
    },
    {
      "completion_length": 245.7500114440918,
      "epoch": 1.660212079299216,
      "grad_norm": 0.2604433695384531,
      "kl": 0.42327880859375,
      "learning_rate": 4.97540083231469e-07,
      "loss": 0.0004,
      "reward": 1.8285714760422707,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.8285714574158192,
      "rewards/format_reward_func": 1.0,
      "step": 9902
    },
    {
      "completion_length": 246.0937623977661,
      "epoch": 1.6605473825390837,
      "grad_norm": 0.523587038459449,
      "kl": 0.1160888671875,
      "learning_rate": 4.975384042898542e-07,
      "loss": 0.0001,
      "reward": 1.7660714983940125,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7705357484519482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9904
    },
    {
      "completion_length": 240.57590579986572,
      "epoch": 1.6608826857789514,
      "grad_norm": 0.2091887597400727,
      "kl": 0.1318206787109375,
      "learning_rate": 4.975367247783144e-07,
      "loss": 0.0001,
      "reward": 1.81428574770689,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857328057289,
      "rewards/format_reward_func": 1.0,
      "step": 9906
    },
    {
      "completion_length": 251.88840579986572,
      "epoch": 1.661217989018819,
      "grad_norm": 0.38583211577731064,
      "kl": 0.101470947265625,
      "learning_rate": 4.97535044696853e-07,
      "loss": 0.0001,
      "reward": 1.7196429446339607,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7241071797907352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9908
    },
    {
      "completion_length": 251.37947750091553,
      "epoch": 1.6615532922586864,
      "grad_norm": 0.3002960218342867,
      "kl": 0.4559783935546875,
      "learning_rate": 4.975333640454743e-07,
      "loss": 0.0005,
      "reward": 1.7357143610715866,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7357143238186836,
      "rewards/format_reward_func": 1.0,
      "step": 9910
    },
    {
      "completion_length": 240.1785831451416,
      "epoch": 1.661888595498554,
      "grad_norm": 0.23452481171864847,
      "kl": 0.275421142578125,
      "learning_rate": 4.975316828241821e-07,
      "loss": 0.0003,
      "reward": 1.8107143267989159,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8107143193483353,
      "rewards/format_reward_func": 1.0,
      "step": 9912
    },
    {
      "completion_length": 238.3839406967163,
      "epoch": 1.6622238987384215,
      "grad_norm": 0.2104834790867371,
      "kl": 0.110626220703125,
      "learning_rate": 4.9753000103298e-07,
      "loss": 0.0001,
      "reward": 1.7625000551342964,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643275439739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9914
    },
    {
      "completion_length": 252.7946548461914,
      "epoch": 1.662559201978289,
      "grad_norm": 0.2198794519534018,
      "kl": 0.5018157958984375,
      "learning_rate": 4.975283186718722e-07,
      "loss": 0.0005,
      "reward": 1.7125000730156898,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7169643174856901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9916
    },
    {
      "completion_length": 253.2544755935669,
      "epoch": 1.6628945052181567,
      "grad_norm": 0.6512800667382872,
      "kl": 1.954864501953125,
      "learning_rate": 4.975266357408623e-07,
      "loss": 0.002,
      "reward": 1.7464286163449287,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7553571872413158,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9918
    },
    {
      "completion_length": 264.151798248291,
      "epoch": 1.6632298084580244,
      "grad_norm": 0.19630697068622924,
      "kl": 0.124298095703125,
      "learning_rate": 4.975249522399544e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714786499739,
      "rewards/format_reward_func": 1.0,
      "step": 9920
    },
    {
      "completion_length": 243.2455472946167,
      "epoch": 1.6635651116978918,
      "grad_norm": 0.3169098579351197,
      "kl": 0.1141357421875,
      "learning_rate": 4.975232681691523e-07,
      "loss": 0.0001,
      "reward": 1.7446429207921028,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071708500385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9922
    },
    {
      "completion_length": 247.95983219146729,
      "epoch": 1.6639004149377592,
      "grad_norm": 0.24182742790034367,
      "kl": 0.21661376953125,
      "learning_rate": 4.975215835284598e-07,
      "loss": 0.0002,
      "reward": 1.8000000640749931,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 9924
    },
    {
      "completion_length": 239.7009048461914,
      "epoch": 1.6642357181776268,
      "grad_norm": 0.007444658415270113,
      "kl": 0.1461181640625,
      "learning_rate": 4.975198983178808e-07,
      "loss": 0.0001,
      "reward": 1.7339286133646965,
      "reward_std": 0.0328299580141902,
      "rewards/equation_reward_func": 0.7383928932249546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9926
    },
    {
      "completion_length": 245.08036708831787,
      "epoch": 1.6645710214174945,
      "grad_norm": 0.18782957922663335,
      "kl": 0.115631103515625,
      "learning_rate": 4.975182125374193e-07,
      "loss": 0.0001,
      "reward": 1.7928571924567223,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571812808514,
      "rewards/format_reward_func": 1.0,
      "step": 9928
    },
    {
      "completion_length": 257.1517972946167,
      "epoch": 1.664906324657362,
      "grad_norm": 0.3034523369429072,
      "kl": 0.4335174560546875,
      "learning_rate": 4.975165261870791e-07,
      "loss": 0.0004,
      "reward": 1.778571479022503,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7785714417695999,
      "rewards/format_reward_func": 1.0,
      "step": 9930
    },
    {
      "completion_length": 257.20537281036377,
      "epoch": 1.6652416278972295,
      "grad_norm": 0.2824380825578585,
      "kl": 0.2369384765625,
      "learning_rate": 4.975148392668641e-07,
      "loss": 0.0002,
      "reward": 1.750000074505806,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.758928619325161,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9932
    },
    {
      "completion_length": 245.7232265472412,
      "epoch": 1.6655769311370971,
      "grad_norm": 0.20963993568996167,
      "kl": 0.15716552734375,
      "learning_rate": 4.975131517767782e-07,
      "loss": 0.0002,
      "reward": 1.7785714864730835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 9934
    },
    {
      "completion_length": 254.8437623977661,
      "epoch": 1.6659122343769646,
      "grad_norm": 0.1708738271349833,
      "kl": 0.327911376953125,
      "learning_rate": 4.975114637168252e-07,
      "loss": 0.0003,
      "reward": 1.7428572103381157,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 9936
    },
    {
      "completion_length": 251.62054634094238,
      "epoch": 1.6662475376168322,
      "grad_norm": 0.15899310756923962,
      "kl": 0.11993408203125,
      "learning_rate": 4.97509775087009e-07,
      "loss": 0.0001,
      "reward": 1.742857187986374,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 9938
    },
    {
      "completion_length": 259.2500114440918,
      "epoch": 1.6665828408566998,
      "grad_norm": 0.25007766641668766,
      "kl": 0.113739013671875,
      "learning_rate": 4.975080858873336e-07,
      "loss": 0.0001,
      "reward": 1.744642935693264,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7491071671247482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9940
    },
    {
      "completion_length": 253.57144260406494,
      "epoch": 1.6669181440965675,
      "grad_norm": 0.29124021885748613,
      "kl": 0.125030517578125,
      "learning_rate": 4.975063961178027e-07,
      "loss": 0.0001,
      "reward": 1.7982143685221672,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8026785887777805,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9942
    },
    {
      "completion_length": 250.4375114440918,
      "epoch": 1.6672534473364349,
      "grad_norm": 0.36358150917938814,
      "kl": 0.12762451171875,
      "learning_rate": 4.975047057784204e-07,
      "loss": 0.0001,
      "reward": 1.7303572222590446,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7348214611411095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9944
    },
    {
      "completion_length": 264.16965675354004,
      "epoch": 1.6675887505763023,
      "grad_norm": 0.27750091888726475,
      "kl": 0.11468505859375,
      "learning_rate": 4.975030148691905e-07,
      "loss": 0.0001,
      "reward": 1.7321429401636124,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7410714589059353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9946
    },
    {
      "completion_length": 264.3125104904175,
      "epoch": 1.66792405381617,
      "grad_norm": 0.26333116652019245,
      "kl": 0.13232421875,
      "learning_rate": 4.975013233901169e-07,
      "loss": 0.0001,
      "reward": 1.7464286610484123,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 9948
    },
    {
      "completion_length": 251.47769260406494,
      "epoch": 1.6682593570560376,
      "grad_norm": 0.21635846864664374,
      "kl": 0.112518310546875,
      "learning_rate": 4.974996313412034e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.810714315623045,
      "rewards/format_reward_func": 1.0,
      "step": 9950
    },
    {
      "completion_length": 252.16072368621826,
      "epoch": 1.6685946602959052,
      "grad_norm": 0.1487337396106638,
      "kl": 0.1123046875,
      "learning_rate": 4.974979387224541e-07,
      "loss": 0.0001,
      "reward": 1.7178571969270706,
      "reward_std": 0.08586296625435352,
      "rewards/equation_reward_func": 0.7357143275439739,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 9952
    },
    {
      "completion_length": 262.29465675354004,
      "epoch": 1.6689299635357726,
      "grad_norm": 0.13238518388734732,
      "kl": 0.10601806640625,
      "learning_rate": 4.974962455338728e-07,
      "loss": 0.0001,
      "reward": 1.7946429029107094,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7991071715950966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9954
    },
    {
      "completion_length": 266.8928689956665,
      "epoch": 1.6692652667756402,
      "grad_norm": 0.1476093870816,
      "kl": 0.1008758544921875,
      "learning_rate": 4.974945517754633e-07,
      "loss": 0.0001,
      "reward": 1.7375000715255737,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7508928906172514,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 9956
    },
    {
      "completion_length": 252.73661994934082,
      "epoch": 1.6696005700155077,
      "grad_norm": 0.6375593922999553,
      "kl": 0.1016998291015625,
      "learning_rate": 4.974928574472296e-07,
      "loss": 0.0001,
      "reward": 1.7839286178350449,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7883928753435612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9958
    },
    {
      "completion_length": 262.15626335144043,
      "epoch": 1.6699358732553753,
      "grad_norm": 0.1283348204099573,
      "kl": 0.1146392822265625,
      "learning_rate": 4.974911625491755e-07,
      "loss": 0.0001,
      "reward": 1.7410714849829674,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7455357629805803,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9960
    },
    {
      "completion_length": 264.4419775009155,
      "epoch": 1.670271176495243,
      "grad_norm": 0.15103147121960364,
      "kl": 0.107666015625,
      "learning_rate": 4.974894670813051e-07,
      "loss": 0.0001,
      "reward": 1.7785714715719223,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 9962
    },
    {
      "completion_length": 255.56697273254395,
      "epoch": 1.6706064797351106,
      "grad_norm": 0.4432214758021507,
      "kl": 0.110931396484375,
      "learning_rate": 4.974877710436222e-07,
      "loss": 0.0001,
      "reward": 1.7142857909202576,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7321428917348385,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 9964
    },
    {
      "completion_length": 253.93751049041748,
      "epoch": 1.670941782974978,
      "grad_norm": 0.1251880983482693,
      "kl": 0.114776611328125,
      "learning_rate": 4.974860744361306e-07,
      "loss": 0.0001,
      "reward": 1.7428572177886963,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571619093418,
      "rewards/format_reward_func": 1.0,
      "step": 9966
    },
    {
      "completion_length": 252.72769451141357,
      "epoch": 1.6712770862148454,
      "grad_norm": 0.21358964680499068,
      "kl": 0.11474609375,
      "learning_rate": 4.974843772588343e-07,
      "loss": 0.0001,
      "reward": 1.7910714894533157,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7955357432365417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9968
    },
    {
      "completion_length": 250.03572177886963,
      "epoch": 1.671612389454713,
      "grad_norm": 0.17397521385257672,
      "kl": 0.102630615234375,
      "learning_rate": 4.974826795117371e-07,
      "loss": 0.0001,
      "reward": 1.733928605914116,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7383928969502449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9970
    },
    {
      "completion_length": 239.4910831451416,
      "epoch": 1.6719476926945807,
      "grad_norm": 0.29886153992095205,
      "kl": 0.106719970703125,
      "learning_rate": 4.974809811948432e-07,
      "loss": 0.0001,
      "reward": 1.7107143700122833,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7107143215835094,
      "rewards/format_reward_func": 1.0,
      "step": 9972
    },
    {
      "completion_length": 244.91072463989258,
      "epoch": 1.6722829959344483,
      "grad_norm": 0.2621951499898351,
      "kl": 0.11273193359375,
      "learning_rate": 4.974792823081563e-07,
      "loss": 0.0001,
      "reward": 1.785714365541935,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 9974
    },
    {
      "completion_length": 235.9375123977661,
      "epoch": 1.672618299174316,
      "grad_norm": 0.1754438312521652,
      "kl": 0.0966033935546875,
      "learning_rate": 4.974775828516803e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 9976
    },
    {
      "completion_length": 252.69644355773926,
      "epoch": 1.6729536024141833,
      "grad_norm": 0.19395800186925502,
      "kl": 0.168914794921875,
      "learning_rate": 4.974758828254192e-07,
      "loss": 0.0002,
      "reward": 1.8214286118745804,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8214285895228386,
      "rewards/format_reward_func": 1.0,
      "step": 9978
    },
    {
      "completion_length": 246.8973331451416,
      "epoch": 1.6732889056540508,
      "grad_norm": 0.2813224848583,
      "kl": 0.138519287109375,
      "learning_rate": 4.974741822293768e-07,
      "loss": 0.0001,
      "reward": 1.7750000804662704,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7839285880327225,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 9980
    },
    {
      "completion_length": 253.12054824829102,
      "epoch": 1.6736242088939184,
      "grad_norm": 0.14022046326230286,
      "kl": 0.106109619140625,
      "learning_rate": 4.97472481063557e-07,
      "loss": 0.0001,
      "reward": 1.746428668498993,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 9982
    },
    {
      "completion_length": 247.61608409881592,
      "epoch": 1.673959512133786,
      "grad_norm": 0.1908507372333823,
      "kl": 0.1064453125,
      "learning_rate": 4.974707793279638e-07,
      "loss": 0.0001,
      "reward": 1.7785714715719223,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 9984
    },
    {
      "completion_length": 258.3660831451416,
      "epoch": 1.6742948153736537,
      "grad_norm": 0.19304693534597053,
      "kl": 0.1163177490234375,
      "learning_rate": 4.974690770226012e-07,
      "loss": 0.0001,
      "reward": 1.6821429282426834,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6821428928524256,
      "rewards/format_reward_func": 1.0,
      "step": 9986
    },
    {
      "completion_length": 246.45090579986572,
      "epoch": 1.674630118613521,
      "grad_norm": 0.1695020044930438,
      "kl": 0.12274169921875,
      "learning_rate": 4.974673741474731e-07,
      "loss": 0.0001,
      "reward": 1.7482143491506577,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7526785992085934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9988
    },
    {
      "completion_length": 241.7544765472412,
      "epoch": 1.6749654218533887,
      "grad_norm": 0.00708512969713439,
      "kl": 0.116058349609375,
      "learning_rate": 4.974656707025832e-07,
      "loss": 0.0001,
      "reward": 1.821428619325161,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.8214285932481289,
      "rewards/format_reward_func": 1.0,
      "step": 9990
    },
    {
      "completion_length": 252.5491189956665,
      "epoch": 1.6753007250932561,
      "grad_norm": 0.09447739873956,
      "kl": 0.15313720703125,
      "learning_rate": 4.974639666879356e-07,
      "loss": 0.0002,
      "reward": 1.7267857789993286,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7312500439584255,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9992
    },
    {
      "completion_length": 239.4419765472412,
      "epoch": 1.6756360283331238,
      "grad_norm": 0.3231878362488016,
      "kl": 0.1316986083984375,
      "learning_rate": 4.974622621035342e-07,
      "loss": 0.0001,
      "reward": 1.7446429282426834,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7491071708500385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 9994
    },
    {
      "completion_length": 241.5759038925171,
      "epoch": 1.6759713315729914,
      "grad_norm": 0.14058138538907927,
      "kl": 0.0970916748046875,
      "learning_rate": 4.97460556949383e-07,
      "loss": 0.0001,
      "reward": 1.74642863124609,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286163449287,
      "rewards/format_reward_func": 1.0,
      "step": 9996
    },
    {
      "completion_length": 253.74108505249023,
      "epoch": 1.676306634812859,
      "grad_norm": 0.22176562412261744,
      "kl": 0.138458251953125,
      "learning_rate": 4.974588512254858e-07,
      "loss": 0.0001,
      "reward": 1.8107143715023994,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 9998
    },
    {
      "completion_length": 248.26340675354004,
      "epoch": 1.6766419380527264,
      "grad_norm": 0.2757893418813273,
      "kl": 0.1244964599609375,
      "learning_rate": 4.974571449318465e-07,
      "loss": 0.0001,
      "reward": 1.7642858028411865,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 10000
    },
    {
      "completion_length": 257.0134029388428,
      "epoch": 1.6769772412925938,
      "grad_norm": 0.686820231753818,
      "kl": 0.2299652099609375,
      "learning_rate": 4.974554380684692e-07,
      "loss": 0.0002,
      "reward": 1.7446429431438446,
      "reward_std": 0.09848987217992544,
      "rewards/equation_reward_func": 0.7580357417464256,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10002
    },
    {
      "completion_length": 254.4687623977661,
      "epoch": 1.6773125445324615,
      "grad_norm": 0.3286740146450279,
      "kl": 0.186676025390625,
      "learning_rate": 4.974537306353577e-07,
      "loss": 0.0002,
      "reward": 1.798214353621006,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8026785887777805,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10004
    },
    {
      "completion_length": 251.5714406967163,
      "epoch": 1.6776478477723291,
      "grad_norm": 0.13520624126222636,
      "kl": 0.3016204833984375,
      "learning_rate": 4.97452022632516e-07,
      "loss": 0.0003,
      "reward": 1.7482143491506577,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7526786103844643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10006
    },
    {
      "completion_length": 258.53125953674316,
      "epoch": 1.6779831510121967,
      "grad_norm": 0.15517144788447346,
      "kl": 0.12713623046875,
      "learning_rate": 4.97450314059948e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 10008
    },
    {
      "completion_length": 249.79465579986572,
      "epoch": 1.6783184542520642,
      "grad_norm": 0.25555380520299875,
      "kl": 0.12371826171875,
      "learning_rate": 4.974486049176575e-07,
      "loss": 0.0001,
      "reward": 1.7071429342031479,
      "reward_std": 0.06060915347188711,
      "rewards/equation_reward_func": 0.7160714641213417,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10010
    },
    {
      "completion_length": 256.10268688201904,
      "epoch": 1.6786537574919318,
      "grad_norm": 0.11627685203612302,
      "kl": 0.11810302734375,
      "learning_rate": 4.974468952056487e-07,
      "loss": 0.0001,
      "reward": 1.7196429446339607,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7241071686148643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10012
    },
    {
      "completion_length": 253.6428680419922,
      "epoch": 1.6789890607317992,
      "grad_norm": 0.20586557717180662,
      "kl": 0.1288604736328125,
      "learning_rate": 4.974451849239253e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7589285895228386,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10014
    },
    {
      "completion_length": 250.83483123779297,
      "epoch": 1.6793243639716668,
      "grad_norm": 0.1705164777603785,
      "kl": 0.10638427734375,
      "learning_rate": 4.974434740724915e-07,
      "loss": 0.0001,
      "reward": 1.751785784959793,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7562500331550837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10016
    },
    {
      "completion_length": 258.96875953674316,
      "epoch": 1.6796596672115345,
      "grad_norm": 0.12622978400779727,
      "kl": 0.180267333984375,
      "learning_rate": 4.974417626513509e-07,
      "loss": 0.0002,
      "reward": 1.8214286118745804,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8214286006987095,
      "rewards/format_reward_func": 1.0,
      "step": 10018
    },
    {
      "completion_length": 262.65626525878906,
      "epoch": 1.6799949704514021,
      "grad_norm": 0.260643559753684,
      "kl": 0.163970947265625,
      "learning_rate": 4.974400506605077e-07,
      "loss": 0.0002,
      "reward": 1.7375000938773155,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7419643178582191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10020
    },
    {
      "completion_length": 264.2321538925171,
      "epoch": 1.6803302736912695,
      "grad_norm": 0.22700129756442697,
      "kl": 0.22357177734375,
      "learning_rate": 4.974383380999657e-07,
      "loss": 0.0002,
      "reward": 1.7482143640518188,
      "reward_std": 0.042931484058499336,
      "rewards/equation_reward_func": 0.7616071701049805,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10022
    },
    {
      "completion_length": 260.0178699493408,
      "epoch": 1.680665576931137,
      "grad_norm": 0.6076334386049436,
      "kl": 0.208984375,
      "learning_rate": 4.97436624969729e-07,
      "loss": 0.0002,
      "reward": 1.7589286342263222,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7723214589059353,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10024
    },
    {
      "completion_length": 272.1696557998657,
      "epoch": 1.6810008801710046,
      "grad_norm": 0.1626425044654749,
      "kl": 0.129058837890625,
      "learning_rate": 4.974349112698014e-07,
      "loss": 0.0001,
      "reward": 1.7910714745521545,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.8044643066823483,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10026
    },
    {
      "completion_length": 258.67412090301514,
      "epoch": 1.6813361834108722,
      "grad_norm": 0.2425983207701506,
      "kl": 0.142913818359375,
      "learning_rate": 4.974331970001869e-07,
      "loss": 0.0001,
      "reward": 1.791071504354477,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7955357395112514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10028
    },
    {
      "completion_length": 257.92412090301514,
      "epoch": 1.6816714866507398,
      "grad_norm": 0.21702457725067278,
      "kl": 0.1026611328125,
      "learning_rate": 4.974314821608894e-07,
      "loss": 0.0001,
      "reward": 1.7571429163217545,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7660714462399483,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10030
    },
    {
      "completion_length": 248.46429634094238,
      "epoch": 1.6820067898906075,
      "grad_norm": 0.30085577116212653,
      "kl": 0.1477813720703125,
      "learning_rate": 4.974297667519129e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7625000290572643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10032
    },
    {
      "completion_length": 257.2589416503906,
      "epoch": 1.682342093130475,
      "grad_norm": 0.06563080695637653,
      "kl": 0.148345947265625,
      "learning_rate": 4.974280507732613e-07,
      "loss": 0.0001,
      "reward": 1.787500061094761,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7919643148779869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10034
    },
    {
      "completion_length": 261.8884048461914,
      "epoch": 1.6826773963703423,
      "grad_norm": 0.3108712427672666,
      "kl": 0.203338623046875,
      "learning_rate": 4.974263342249387e-07,
      "loss": 0.0002,
      "reward": 1.7535714879631996,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7625000160187483,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10036
    },
    {
      "completion_length": 266.7187623977661,
      "epoch": 1.68301269961021,
      "grad_norm": 0.39533845320926486,
      "kl": 0.125885009765625,
      "learning_rate": 4.974246171069489e-07,
      "loss": 0.0001,
      "reward": 1.7000000700354576,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7089286036789417,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10038
    },
    {
      "completion_length": 246.85715198516846,
      "epoch": 1.6833480028500776,
      "grad_norm": 0.24514446508009619,
      "kl": 0.11993408203125,
      "learning_rate": 4.974228994192959e-07,
      "loss": 0.0001,
      "reward": 1.7732143327593803,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7776786014437675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10040
    },
    {
      "completion_length": 266.3794765472412,
      "epoch": 1.6836833060899452,
      "grad_norm": 0.21864629994207177,
      "kl": 0.13507080078125,
      "learning_rate": 4.974211811619836e-07,
      "loss": 0.0001,
      "reward": 1.814285770058632,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857365310192,
      "rewards/format_reward_func": 1.0,
      "step": 10042
    },
    {
      "completion_length": 262.80358123779297,
      "epoch": 1.6840186093298126,
      "grad_norm": 0.17710018125664703,
      "kl": 0.136474609375,
      "learning_rate": 4.97419462335016e-07,
      "loss": 0.0001,
      "reward": 1.7357143461704254,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7446428909897804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10044
    },
    {
      "completion_length": 249.85715293884277,
      "epoch": 1.6843539125696803,
      "grad_norm": 0.32386004346591574,
      "kl": 0.110443115234375,
      "learning_rate": 4.974177429383971e-07,
      "loss": 0.0001,
      "reward": 1.800000049173832,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 10046
    },
    {
      "completion_length": 260.33483505249023,
      "epoch": 1.6846892158095477,
      "grad_norm": 0.2952786587131419,
      "kl": 0.142913818359375,
      "learning_rate": 4.974160229721308e-07,
      "loss": 0.0001,
      "reward": 1.7303572073578835,
      "reward_std": 0.09848987124860287,
      "rewards/equation_reward_func": 0.7437500320374966,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10048
    },
    {
      "completion_length": 264.16519260406494,
      "epoch": 1.6850245190494153,
      "grad_norm": 0.4340353985467339,
      "kl": 0.12249755859375,
      "learning_rate": 4.974143024362211e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 10050
    },
    {
      "completion_length": 245.21875953674316,
      "epoch": 1.685359822289283,
      "grad_norm": 0.41833085967800654,
      "kl": 0.1136474609375,
      "learning_rate": 4.974125813306719e-07,
      "loss": 0.0001,
      "reward": 1.6964286416769028,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7142857499420643,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 10052
    },
    {
      "completion_length": 252.24554634094238,
      "epoch": 1.6856951255291506,
      "grad_norm": 0.3466336130381244,
      "kl": 0.116119384765625,
      "learning_rate": 4.974108596554872e-07,
      "loss": 0.0001,
      "reward": 1.8142857775092125,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857514321804,
      "rewards/format_reward_func": 1.0,
      "step": 10054
    },
    {
      "completion_length": 252.09822368621826,
      "epoch": 1.686030428769018,
      "grad_norm": 0.2296487564634485,
      "kl": 0.110504150390625,
      "learning_rate": 4.97409137410671e-07,
      "loss": 0.0001,
      "reward": 1.7446429207921028,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7491071727126837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10056
    },
    {
      "completion_length": 244.94643783569336,
      "epoch": 1.6863657320088854,
      "grad_norm": 0.15632409057849708,
      "kl": 0.120025634765625,
      "learning_rate": 4.974074145962272e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 10058
    },
    {
      "completion_length": 243.4330472946167,
      "epoch": 1.686701035248753,
      "grad_norm": 0.2494711951314461,
      "kl": 0.131256103515625,
      "learning_rate": 4.974056912121599e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 10060
    },
    {
      "completion_length": 251.37054634094238,
      "epoch": 1.6870363384886207,
      "grad_norm": 0.23723346999630718,
      "kl": 0.111907958984375,
      "learning_rate": 4.974039672584729e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.757142897695303,
      "rewards/format_reward_func": 1.0,
      "step": 10062
    },
    {
      "completion_length": 263.2991189956665,
      "epoch": 1.6873716417284883,
      "grad_norm": 0.24109290815217851,
      "kl": 0.156463623046875,
      "learning_rate": 4.974022427351703e-07,
      "loss": 0.0002,
      "reward": 1.7267857864499092,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7312500402331352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10064
    },
    {
      "completion_length": 248.97322463989258,
      "epoch": 1.6877069449683557,
      "grad_norm": 0.24947769782804471,
      "kl": 0.12640380859375,
      "learning_rate": 4.974005176422559e-07,
      "loss": 0.0001,
      "reward": 1.7607143446803093,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 10066
    },
    {
      "completion_length": 245.2946538925171,
      "epoch": 1.6880422482082234,
      "grad_norm": 0.40610948081296316,
      "kl": 0.125946044921875,
      "learning_rate": 4.973987919797337e-07,
      "loss": 0.0001,
      "reward": 1.8107143640518188,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.8196428716182709,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10068
    },
    {
      "completion_length": 257.37947273254395,
      "epoch": 1.6883775514480908,
      "grad_norm": 0.12776086661506092,
      "kl": 0.151214599609375,
      "learning_rate": 4.973970657476079e-07,
      "loss": 0.0002,
      "reward": 1.7875000685453415,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7919643074274063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10070
    },
    {
      "completion_length": 231.50893878936768,
      "epoch": 1.6887128546879584,
      "grad_norm": 0.1613835712573395,
      "kl": 0.0987548828125,
      "learning_rate": 4.973953389458824e-07,
      "loss": 0.0001,
      "reward": 1.8392857760190964,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.839285746216774,
      "rewards/format_reward_func": 1.0,
      "step": 10072
    },
    {
      "completion_length": 250.8750114440918,
      "epoch": 1.689048157927826,
      "grad_norm": 0.21526186107713036,
      "kl": 0.144287109375,
      "learning_rate": 4.97393611574561e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 10074
    },
    {
      "completion_length": 242.64733505249023,
      "epoch": 1.6893834611676937,
      "grad_norm": 0.14668393885929962,
      "kl": 0.16693115234375,
      "learning_rate": 4.973918836336478e-07,
      "loss": 0.0002,
      "reward": 1.7892857566475868,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857305705547,
      "rewards/format_reward_func": 1.0,
      "step": 10076
    },
    {
      "completion_length": 251.165189743042,
      "epoch": 1.689718764407561,
      "grad_norm": 0.6243773833866528,
      "kl": 0.200439453125,
      "learning_rate": 4.973901551231467e-07,
      "loss": 0.0002,
      "reward": 1.7625000774860382,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10078
    },
    {
      "completion_length": 243.98215293884277,
      "epoch": 1.6900540676474285,
      "grad_norm": 0.158549076583998,
      "kl": 0.1374359130859375,
      "learning_rate": 4.973884260430617e-07,
      "loss": 0.0001,
      "reward": 1.7625000551342964,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643200933933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10080
    },
    {
      "completion_length": 250.665189743042,
      "epoch": 1.6903893708872961,
      "grad_norm": 0.17691845376375445,
      "kl": 0.2376861572265625,
      "learning_rate": 4.97386696393397e-07,
      "loss": 0.0002,
      "reward": 1.8000000715255737,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000156462193,
      "rewards/format_reward_func": 1.0,
      "step": 10082
    },
    {
      "completion_length": 248.63840293884277,
      "epoch": 1.6907246741271638,
      "grad_norm": 0.20282061468046836,
      "kl": 0.1807861328125,
      "learning_rate": 4.973849661741563e-07,
      "loss": 0.0002,
      "reward": 1.7732143700122833,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7776786014437675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10084
    },
    {
      "completion_length": 244.196439743042,
      "epoch": 1.6910599773670314,
      "grad_norm": 0.28045689010751607,
      "kl": 0.71612548828125,
      "learning_rate": 4.973832353853436e-07,
      "loss": 0.0007,
      "reward": 1.7857143431901932,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143003493547,
      "rewards/format_reward_func": 1.0,
      "step": 10086
    },
    {
      "completion_length": 242.47322750091553,
      "epoch": 1.6913952806068988,
      "grad_norm": 0.3193025602899345,
      "kl": 0.498504638671875,
      "learning_rate": 4.973815040269631e-07,
      "loss": 0.0005,
      "reward": 1.7500000670552254,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 10088
    },
    {
      "completion_length": 246.7142972946167,
      "epoch": 1.6917305838467664,
      "grad_norm": 0.23974640616671034,
      "kl": 0.428253173828125,
      "learning_rate": 4.973797720990186e-07,
      "loss": 0.0004,
      "reward": 1.7857143506407738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143059372902,
      "rewards/format_reward_func": 1.0,
      "step": 10090
    },
    {
      "completion_length": 264.4419775009155,
      "epoch": 1.6920658870866339,
      "grad_norm": 0.4536671668732646,
      "kl": 0.81256103515625,
      "learning_rate": 4.973780396015142e-07,
      "loss": 0.0008,
      "reward": 1.7321429178118706,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7321429029107094,
      "rewards/format_reward_func": 1.0,
      "step": 10092
    },
    {
      "completion_length": 245.4776906967163,
      "epoch": 1.6924011903265015,
      "grad_norm": 0.17562989852246272,
      "kl": 0.906951904296875,
      "learning_rate": 4.973763065344538e-07,
      "loss": 0.0009,
      "reward": 1.7232143729925156,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7276786081492901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10094
    },
    {
      "completion_length": 254.79018688201904,
      "epoch": 1.6927364935663691,
      "grad_norm": 0.07665879961083437,
      "kl": 0.70660400390625,
      "learning_rate": 4.973745728978413e-07,
      "loss": 0.0007,
      "reward": 1.7892857566475868,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 10096
    },
    {
      "completion_length": 245.40626049041748,
      "epoch": 1.6930717968062368,
      "grad_norm": 0.2503366730466342,
      "kl": 0.56756591796875,
      "learning_rate": 4.97372838691681e-07,
      "loss": 0.0006,
      "reward": 1.7589286491274834,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7633928805589676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10098
    },
    {
      "completion_length": 253.65626049041748,
      "epoch": 1.6934071000461042,
      "grad_norm": 0.6320229532136868,
      "kl": 1.1220703125,
      "learning_rate": 4.973711039159765e-07,
      "loss": 0.0011,
      "reward": 1.7776786237955093,
      "reward_std": 0.05177031829953194,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 0.9848214387893677,
      "step": 10100
    },
    {
      "completion_length": 245.08483219146729,
      "epoch": 1.6937424032859716,
      "grad_norm": 0.1919645569093751,
      "kl": 0.560791015625,
      "learning_rate": 4.973693685707322e-07,
      "loss": 0.0006,
      "reward": 1.7500000521540642,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 10102
    },
    {
      "completion_length": 240.29911518096924,
      "epoch": 1.6940777065258392,
      "grad_norm": 0.1526633101158587,
      "kl": 0.2593841552734375,
      "learning_rate": 4.973676326559518e-07,
      "loss": 0.0003,
      "reward": 1.7964286282658577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964285872876644,
      "rewards/format_reward_func": 1.0,
      "step": 10104
    },
    {
      "completion_length": 243.66518878936768,
      "epoch": 1.6944130097657069,
      "grad_norm": 0.19763174283560406,
      "kl": 0.2373809814453125,
      "learning_rate": 4.973658961716394e-07,
      "loss": 0.0002,
      "reward": 1.7839286476373672,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7883928753435612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10106
    },
    {
      "completion_length": 242.6651906967163,
      "epoch": 1.6947483130055745,
      "grad_norm": 0.18538824341398405,
      "kl": 0.131744384765625,
      "learning_rate": 4.973641591177991e-07,
      "loss": 0.0001,
      "reward": 1.8178572058677673,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.817857164889574,
      "rewards/format_reward_func": 1.0,
      "step": 10108
    },
    {
      "completion_length": 242.34822273254395,
      "epoch": 1.6950836162454421,
      "grad_norm": 0.21636593330112164,
      "kl": 0.161346435546875,
      "learning_rate": 4.973624214944347e-07,
      "loss": 0.0002,
      "reward": 1.7607143595814705,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7607143167406321,
      "rewards/format_reward_func": 1.0,
      "step": 10110
    },
    {
      "completion_length": 243.72322273254395,
      "epoch": 1.6954189194853095,
      "grad_norm": 0.4051968109433958,
      "kl": 0.2559814453125,
      "learning_rate": 4.973606833015503e-07,
      "loss": 0.0003,
      "reward": 1.7660714983940125,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7705357391387224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10112
    },
    {
      "completion_length": 250.1696548461914,
      "epoch": 1.695754222725177,
      "grad_norm": 0.1942327326972562,
      "kl": 0.23480224609375,
      "learning_rate": 4.973589445391497e-07,
      "loss": 0.0002,
      "reward": 1.778571493923664,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 10114
    },
    {
      "completion_length": 231.977689743042,
      "epoch": 1.6960895259650446,
      "grad_norm": 0.2900110471204609,
      "kl": 0.150970458984375,
      "learning_rate": 4.973572052072374e-07,
      "loss": 0.0002,
      "reward": 1.7642857655882835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857525497675,
      "rewards/format_reward_func": 1.0,
      "step": 10116
    },
    {
      "completion_length": 244.3303680419922,
      "epoch": 1.6964248292049122,
      "grad_norm": 0.23325186627488984,
      "kl": 0.206512451171875,
      "learning_rate": 4.973554653058169e-07,
      "loss": 0.0002,
      "reward": 1.7250000908970833,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000424683094,
      "rewards/format_reward_func": 1.0,
      "step": 10118
    },
    {
      "completion_length": 239.58483219146729,
      "epoch": 1.6967601324447799,
      "grad_norm": 0.1860929377205112,
      "kl": 0.1572265625,
      "learning_rate": 4.973537248348925e-07,
      "loss": 0.0002,
      "reward": 1.753571480512619,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714656114578,
      "rewards/format_reward_func": 1.0,
      "step": 10120
    },
    {
      "completion_length": 240.98661613464355,
      "epoch": 1.6970954356846473,
      "grad_norm": 0.8941528688423361,
      "kl": 0.22747802734375,
      "learning_rate": 4.973519837944681e-07,
      "loss": 0.0002,
      "reward": 1.7821429073810577,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821429036557674,
      "rewards/format_reward_func": 1.0,
      "step": 10122
    },
    {
      "completion_length": 253.20983123779297,
      "epoch": 1.697430738924515,
      "grad_norm": 0.46380905705084785,
      "kl": 0.25750732421875,
      "learning_rate": 4.973502421845476e-07,
      "loss": 0.0003,
      "reward": 1.710714377462864,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7196428775787354,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10124
    },
    {
      "completion_length": 249.33037185668945,
      "epoch": 1.6977660421643823,
      "grad_norm": 0.15391532324689755,
      "kl": 0.18988037109375,
      "learning_rate": 4.973485000051354e-07,
      "loss": 0.0002,
      "reward": 1.769642911851406,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10126
    },
    {
      "completion_length": 252.89287090301514,
      "epoch": 1.69810134540425,
      "grad_norm": 0.49648863518351344,
      "kl": 0.2744140625,
      "learning_rate": 4.973467572562351e-07,
      "loss": 0.0003,
      "reward": 1.7339286282658577,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7383929006755352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10128
    },
    {
      "completion_length": 260.56251335144043,
      "epoch": 1.6984366486441176,
      "grad_norm": 0.23758751631952763,
      "kl": 0.212677001953125,
      "learning_rate": 4.973450139378508e-07,
      "loss": 0.0002,
      "reward": 1.7392857819795609,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7482143230736256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10130
    },
    {
      "completion_length": 246.54465675354004,
      "epoch": 1.6987719518839852,
      "grad_norm": 0.008302879011098514,
      "kl": 0.142120361328125,
      "learning_rate": 4.973432700499866e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 10132
    },
    {
      "completion_length": 239.49107837677002,
      "epoch": 1.6991072551238526,
      "grad_norm": 0.1169948662184085,
      "kl": 0.385284423828125,
      "learning_rate": 4.973415255926466e-07,
      "loss": 0.0004,
      "reward": 1.7857143357396126,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143115252256,
      "rewards/format_reward_func": 1.0,
      "step": 10134
    },
    {
      "completion_length": 250.99108505249023,
      "epoch": 1.69944255836372,
      "grad_norm": 0.43755620630028064,
      "kl": 0.3654937744140625,
      "learning_rate": 4.973397805658345e-07,
      "loss": 0.0004,
      "reward": 1.7625000849366188,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669642977416515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10136
    },
    {
      "completion_length": 241.7544755935669,
      "epoch": 1.6997778616035877,
      "grad_norm": 0.1611913201639629,
      "kl": 0.258697509765625,
      "learning_rate": 4.973380349695547e-07,
      "loss": 0.0003,
      "reward": 1.7321429401636124,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321428824216127,
      "rewards/format_reward_func": 1.0,
      "step": 10138
    },
    {
      "completion_length": 248.00447845458984,
      "epoch": 1.7001131648434553,
      "grad_norm": 0.3664824885281984,
      "kl": 0.603790283203125,
      "learning_rate": 4.973362888038109e-07,
      "loss": 0.0006,
      "reward": 1.7535714879631996,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714488476515,
      "rewards/format_reward_func": 1.0,
      "step": 10140
    },
    {
      "completion_length": 251.26787090301514,
      "epoch": 1.700448468083323,
      "grad_norm": 0.40636458209251985,
      "kl": 0.312957763671875,
      "learning_rate": 4.973345420686073e-07,
      "loss": 0.0003,
      "reward": 1.6928572207689285,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.6928571797907352,
      "rewards/format_reward_func": 1.0,
      "step": 10142
    },
    {
      "completion_length": 248.34376049041748,
      "epoch": 1.7007837713231904,
      "grad_norm": 0.33955868228708397,
      "kl": 1.279510498046875,
      "learning_rate": 4.973327947639478e-07,
      "loss": 0.0013,
      "reward": 1.7035714983940125,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7035714648663998,
      "rewards/format_reward_func": 1.0,
      "step": 10144
    },
    {
      "completion_length": 250.28572845458984,
      "epoch": 1.701119074563058,
      "grad_norm": 0.16777483507884186,
      "kl": 0.157379150390625,
      "learning_rate": 4.973310468898366e-07,
      "loss": 0.0002,
      "reward": 1.7642857804894447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 10146
    },
    {
      "completion_length": 253.16965675354004,
      "epoch": 1.7014543778029254,
      "grad_norm": 0.19270567991823492,
      "kl": 0.84698486328125,
      "learning_rate": 4.973292984462777e-07,
      "loss": 0.0008,
      "reward": 1.7446429133415222,
      "reward_std": 0.06060915347188711,
      "rewards/equation_reward_func": 0.7571428958326578,
      "rewards/format_reward_func": 0.987500011920929,
      "step": 10148
    },
    {
      "completion_length": 246.55804443359375,
      "epoch": 1.701789681042793,
      "grad_norm": 0.24648732219059558,
      "kl": 0.365814208984375,
      "learning_rate": 4.973275494332749e-07,
      "loss": 0.0004,
      "reward": 1.7750000730156898,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 10150
    },
    {
      "completion_length": 262.4017972946167,
      "epoch": 1.7021249842826607,
      "grad_norm": 0.36717976358266147,
      "kl": 0.6148681640625,
      "learning_rate": 4.973257998508325e-07,
      "loss": 0.0006,
      "reward": 1.8000000566244125,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000156462193,
      "rewards/format_reward_func": 1.0,
      "step": 10152
    },
    {
      "completion_length": 241.30804634094238,
      "epoch": 1.7024602875225283,
      "grad_norm": 0.13942221880283598,
      "kl": 0.142333984375,
      "learning_rate": 4.973240496989543e-07,
      "loss": 0.0001,
      "reward": 1.7803571820259094,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7848214656114578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10154
    },
    {
      "completion_length": 239.52232933044434,
      "epoch": 1.7027955907623957,
      "grad_norm": 0.505066602090983,
      "kl": 0.241119384765625,
      "learning_rate": 4.973222989776446e-07,
      "loss": 0.0002,
      "reward": 1.7678572162985802,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571864962578,
      "rewards/format_reward_func": 1.0,
      "step": 10156
    },
    {
      "completion_length": 239.7009038925171,
      "epoch": 1.7031308940022631,
      "grad_norm": 0.2516769436746728,
      "kl": 0.176483154296875,
      "learning_rate": 4.973205476869072e-07,
      "loss": 0.0002,
      "reward": 1.7625000551342964,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10158
    },
    {
      "completion_length": 249.70983028411865,
      "epoch": 1.7034661972421308,
      "grad_norm": 0.1202381463499803,
      "kl": 0.1517333984375,
      "learning_rate": 4.973187958267461e-07,
      "loss": 0.0002,
      "reward": 1.76071435213089,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7607143335044384,
      "rewards/format_reward_func": 1.0,
      "step": 10160
    },
    {
      "completion_length": 237.6026906967163,
      "epoch": 1.7038015004819984,
      "grad_norm": 0.15176271405999694,
      "kl": 0.1248779296875,
      "learning_rate": 4.973170433971655e-07,
      "loss": 0.0001,
      "reward": 1.789285771548748,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 10162
    },
    {
      "completion_length": 246.08483219146729,
      "epoch": 1.704136803721866,
      "grad_norm": 0.3686108597554467,
      "kl": 0.121490478515625,
      "learning_rate": 4.973152903981693e-07,
      "loss": 0.0001,
      "reward": 1.803571492433548,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 10164
    },
    {
      "completion_length": 245.67858219146729,
      "epoch": 1.7044721069617337,
      "grad_norm": 0.22525795074098146,
      "kl": 0.11285400390625,
      "learning_rate": 4.973135368297617e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.8035714477300644,
      "rewards/format_reward_func": 1.0,
      "step": 10166
    },
    {
      "completion_length": 238.79911613464355,
      "epoch": 1.704807410201601,
      "grad_norm": 0.1792667048858632,
      "kl": 0.13275146484375,
      "learning_rate": 4.973117826919467e-07,
      "loss": 0.0001,
      "reward": 1.8142857551574707,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857439815998,
      "rewards/format_reward_func": 1.0,
      "step": 10168
    },
    {
      "completion_length": 237.5134038925171,
      "epoch": 1.7051427134414685,
      "grad_norm": 0.002530607373799404,
      "kl": 0.0972442626953125,
      "learning_rate": 4.973100279847281e-07,
      "loss": 0.0001,
      "reward": 1.8071428909897804,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.807142898440361,
      "rewards/format_reward_func": 1.0,
      "step": 10170
    },
    {
      "completion_length": 246.03126335144043,
      "epoch": 1.7054780166813361,
      "grad_norm": 0.3430750903754556,
      "kl": 0.111419677734375,
      "learning_rate": 4.973082727081103e-07,
      "loss": 0.0001,
      "reward": 1.7053572237491608,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7098214738070965,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10172
    },
    {
      "completion_length": 245.3437623977661,
      "epoch": 1.7058133199212038,
      "grad_norm": 0.30599076791001545,
      "kl": 0.1102294921875,
      "learning_rate": 4.97306516862097e-07,
      "loss": 0.0001,
      "reward": 1.832142911851406,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8321428783237934,
      "rewards/format_reward_func": 1.0,
      "step": 10174
    },
    {
      "completion_length": 243.66518878936768,
      "epoch": 1.7061486231610714,
      "grad_norm": 0.3329256720540618,
      "kl": 0.143310546875,
      "learning_rate": 4.973047604466925e-07,
      "loss": 0.0001,
      "reward": 1.7482143640518188,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526785954833031,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10176
    },
    {
      "completion_length": 233.24554443359375,
      "epoch": 1.7064839264009388,
      "grad_norm": 0.2505718550948968,
      "kl": 0.121307373046875,
      "learning_rate": 4.973030034619007e-07,
      "loss": 0.0001,
      "reward": 1.7214286550879478,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7214286122471094,
      "rewards/format_reward_func": 1.0,
      "step": 10178
    },
    {
      "completion_length": 240.8259048461914,
      "epoch": 1.7068192296408065,
      "grad_norm": 0.3175999733034826,
      "kl": 0.1077423095703125,
      "learning_rate": 4.973012459077257e-07,
      "loss": 0.0001,
      "reward": 1.70000009983778,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7000000346451998,
      "rewards/format_reward_func": 1.0,
      "step": 10180
    },
    {
      "completion_length": 238.7589406967163,
      "epoch": 1.7071545328806739,
      "grad_norm": 0.8609321651175816,
      "kl": 0.1385498046875,
      "learning_rate": 4.972994877841715e-07,
      "loss": 0.0001,
      "reward": 1.7767857611179352,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7812500409781933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10182
    },
    {
      "completion_length": 242.0625114440918,
      "epoch": 1.7074898361205415,
      "grad_norm": 0.3140538824392083,
      "kl": 0.120025634765625,
      "learning_rate": 4.972977290912423e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 10184
    },
    {
      "completion_length": 240.6696538925171,
      "epoch": 1.7078251393604091,
      "grad_norm": 0.16024782471908663,
      "kl": 0.1128692626953125,
      "learning_rate": 4.97295969828942e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714440047741,
      "rewards/format_reward_func": 1.0,
      "step": 10186
    },
    {
      "completion_length": 237.9241189956665,
      "epoch": 1.7081604426002768,
      "grad_norm": 0.20369444105406395,
      "kl": 0.1204833984375,
      "learning_rate": 4.972942099972746e-07,
      "loss": 0.0001,
      "reward": 1.825000062584877,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.825000025331974,
      "rewards/format_reward_func": 1.0,
      "step": 10188
    },
    {
      "completion_length": 237.3214406967163,
      "epoch": 1.7084957458401442,
      "grad_norm": 0.46196819682538787,
      "kl": 0.1105804443359375,
      "learning_rate": 4.972924495962443e-07,
      "loss": 0.0001,
      "reward": 1.8250000476837158,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8250000216066837,
      "rewards/format_reward_func": 1.0,
      "step": 10190
    },
    {
      "completion_length": 247.79465293884277,
      "epoch": 1.7088310490800116,
      "grad_norm": 0.17530326999013046,
      "kl": 0.09844970703125,
      "learning_rate": 4.97290688625855e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7446428835391998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10192
    },
    {
      "completion_length": 239.915189743042,
      "epoch": 1.7091663523198792,
      "grad_norm": 0.17669363274091412,
      "kl": 0.10247802734375,
      "learning_rate": 4.97288927086111e-07,
      "loss": 0.0001,
      "reward": 1.8000000268220901,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 10194
    },
    {
      "completion_length": 251.86608219146729,
      "epoch": 1.7095016555597469,
      "grad_norm": 0.1184864862329785,
      "kl": 0.111236572265625,
      "learning_rate": 4.972871649770162e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 10196
    },
    {
      "completion_length": 249.87947845458984,
      "epoch": 1.7098369587996145,
      "grad_norm": 0.3240525754200581,
      "kl": 0.1515350341796875,
      "learning_rate": 4.972854022985746e-07,
      "loss": 0.0002,
      "reward": 1.753571517765522,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714507102966,
      "rewards/format_reward_func": 1.0,
      "step": 10198
    },
    {
      "completion_length": 258.5848331451416,
      "epoch": 1.710172262039482,
      "grad_norm": 0.13925645977761708,
      "kl": 0.135101318359375,
      "learning_rate": 4.972836390507902e-07,
      "loss": 0.0001,
      "reward": 1.7285715267062187,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7375000268220901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10200
    },
    {
      "completion_length": 255.58483219146729,
      "epoch": 1.7105075652793496,
      "grad_norm": 0.3032562982546393,
      "kl": 0.102783203125,
      "learning_rate": 4.972818752336674e-07,
      "loss": 0.0001,
      "reward": 1.7928571924567223,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 10202
    },
    {
      "completion_length": 257.7321557998657,
      "epoch": 1.710842868519217,
      "grad_norm": 0.3941826817499621,
      "kl": 0.14019775390625,
      "learning_rate": 4.972801108472099e-07,
      "loss": 0.0001,
      "reward": 1.7321429252624512,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7410714626312256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10204
    },
    {
      "completion_length": 265.51340675354004,
      "epoch": 1.7111781717590846,
      "grad_norm": 0.25116486602831917,
      "kl": 0.133148193359375,
      "learning_rate": 4.97278345891422e-07,
      "loss": 0.0001,
      "reward": 1.798214316368103,
      "reward_std": 0.06313453149050474,
      "rewards/equation_reward_func": 0.8026785962283611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10206
    },
    {
      "completion_length": 265.3839416503906,
      "epoch": 1.7115134749989522,
      "grad_norm": 0.14768036461804573,
      "kl": 0.12890625,
      "learning_rate": 4.972765803663076e-07,
      "loss": 0.0001,
      "reward": 1.7267857939004898,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7312500290572643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10208
    },
    {
      "completion_length": 254.28572845458984,
      "epoch": 1.7118487782388199,
      "grad_norm": 0.18517748549085836,
      "kl": 0.101348876953125,
      "learning_rate": 4.972748142718708e-07,
      "loss": 0.0001,
      "reward": 1.789285771548748,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857305705547,
      "rewards/format_reward_func": 1.0,
      "step": 10210
    },
    {
      "completion_length": 248.27233219146729,
      "epoch": 1.7121840814786873,
      "grad_norm": 0.16112355740952278,
      "kl": 0.117645263671875,
      "learning_rate": 4.972730476081157e-07,
      "loss": 0.0001,
      "reward": 1.7571429461240768,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 10212
    },
    {
      "completion_length": 249.39733600616455,
      "epoch": 1.7125193847185547,
      "grad_norm": 0.27793530860374993,
      "kl": 0.2154541015625,
      "learning_rate": 4.972712803750464e-07,
      "loss": 0.0002,
      "reward": 1.8196429163217545,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8241071775555611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10214
    },
    {
      "completion_length": 255.77233409881592,
      "epoch": 1.7128546879584223,
      "grad_norm": 0.17317650979675972,
      "kl": 0.1054840087890625,
      "learning_rate": 4.972695125726669e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 10216
    },
    {
      "completion_length": 253.52233123779297,
      "epoch": 1.71318999119829,
      "grad_norm": 0.22910347888793958,
      "kl": 0.133819580078125,
      "learning_rate": 4.972677442009813e-07,
      "loss": 0.0001,
      "reward": 1.7785715088248253,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 10218
    },
    {
      "completion_length": 256.03126335144043,
      "epoch": 1.7135252944381576,
      "grad_norm": 0.1251190478169731,
      "kl": 0.1388702392578125,
      "learning_rate": 4.972659752599937e-07,
      "loss": 0.0001,
      "reward": 1.8482143357396126,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.852678582072258,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10220
    },
    {
      "completion_length": 263.6428699493408,
      "epoch": 1.713860597678025,
      "grad_norm": 0.1527141001062361,
      "kl": 0.1422882080078125,
      "learning_rate": 4.972642057497082e-07,
      "loss": 0.0001,
      "reward": 1.7696429193019867,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7741071656346321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10222
    },
    {
      "completion_length": 267.70983123779297,
      "epoch": 1.7141959009178926,
      "grad_norm": 0.2617393189987282,
      "kl": 0.106292724609375,
      "learning_rate": 4.972624356701287e-07,
      "loss": 0.0001,
      "reward": 1.733928643167019,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7383928932249546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10224
    },
    {
      "completion_length": 259.7410831451416,
      "epoch": 1.71453120415776,
      "grad_norm": 0.2660914466620459,
      "kl": 0.13519287109375,
      "learning_rate": 4.972606650212595e-07,
      "loss": 0.0001,
      "reward": 1.7767857760190964,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7812500149011612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10226
    },
    {
      "completion_length": 262.5357255935669,
      "epoch": 1.7148665073976277,
      "grad_norm": 0.09373059801058264,
      "kl": 0.147003173828125,
      "learning_rate": 4.972588938031045e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 10228
    },
    {
      "completion_length": 269.6651906967163,
      "epoch": 1.7152018106374953,
      "grad_norm": 0.09178558498879472,
      "kl": 0.278045654296875,
      "learning_rate": 4.972571220156679e-07,
      "loss": 0.0003,
      "reward": 1.79464291036129,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7991071529686451,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10230
    },
    {
      "completion_length": 273.5357255935669,
      "epoch": 1.715537113877363,
      "grad_norm": 0.20699127352252952,
      "kl": 0.10955810546875,
      "learning_rate": 4.972553496589536e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000242143869,
      "rewards/format_reward_func": 1.0,
      "step": 10232
    },
    {
      "completion_length": 255.79911994934082,
      "epoch": 1.7158724171172304,
      "grad_norm": 0.16328593761963217,
      "kl": 0.16827392578125,
      "learning_rate": 4.97253576732966e-07,
      "loss": 0.0002,
      "reward": 1.7714286521077156,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714286111295223,
      "rewards/format_reward_func": 1.0,
      "step": 10234
    },
    {
      "completion_length": 251.2812623977661,
      "epoch": 1.7162077203570978,
      "grad_norm": 0.15051227767423914,
      "kl": 0.2393951416015625,
      "learning_rate": 4.972518032377088e-07,
      "loss": 0.0002,
      "reward": 1.7500000819563866,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 10236
    },
    {
      "completion_length": 266.82143783569336,
      "epoch": 1.7165430235969654,
      "grad_norm": 0.5221781685923255,
      "kl": 0.2209320068359375,
      "learning_rate": 4.972500291731865e-07,
      "loss": 0.0002,
      "reward": 1.7973214834928513,
      "reward_std": 0.07449874933809042,
      "rewards/equation_reward_func": 0.8080357369035482,
      "rewards/format_reward_func": 0.9892857223749161,
      "step": 10238
    },
    {
      "completion_length": 256.92858505249023,
      "epoch": 1.716878326836833,
      "grad_norm": 0.1457588564956292,
      "kl": 0.137115478515625,
      "learning_rate": 4.972482545394028e-07,
      "loss": 0.0001,
      "reward": 1.7678572237491608,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 10240
    },
    {
      "completion_length": 261.8526916503906,
      "epoch": 1.7172136300767007,
      "grad_norm": 0.1387693804521752,
      "kl": 0.1348876953125,
      "learning_rate": 4.972464793363619e-07,
      "loss": 0.0001,
      "reward": 1.760714314877987,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.760714340955019,
      "rewards/format_reward_func": 1.0,
      "step": 10242
    },
    {
      "completion_length": 267.48662281036377,
      "epoch": 1.7175489333165683,
      "grad_norm": 0.45388286887329304,
      "kl": 0.276153564453125,
      "learning_rate": 4.972447035640681e-07,
      "loss": 0.0003,
      "reward": 1.7482143864035606,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10244
    },
    {
      "completion_length": 256.8303689956665,
      "epoch": 1.7178842365564357,
      "grad_norm": 0.26309543690645204,
      "kl": 0.109283447265625,
      "learning_rate": 4.972429272225252e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 10246
    },
    {
      "completion_length": 256.8884048461914,
      "epoch": 1.7182195397963032,
      "grad_norm": 0.5253921470745806,
      "kl": 0.2630615234375,
      "learning_rate": 4.972411503117374e-07,
      "loss": 0.0003,
      "reward": 1.7035714983940125,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7125000301748514,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10248
    },
    {
      "completion_length": 261.7857255935669,
      "epoch": 1.7185548430361708,
      "grad_norm": 0.1953770623697014,
      "kl": 0.12847900390625,
      "learning_rate": 4.972393728317089e-07,
      "loss": 0.0001,
      "reward": 1.7410714998841286,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357518047094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10250
    },
    {
      "completion_length": 259.8348321914673,
      "epoch": 1.7188901462760384,
      "grad_norm": 0.1585139958072636,
      "kl": 0.1113128662109375,
      "learning_rate": 4.972375947824437e-07,
      "loss": 0.0001,
      "reward": 1.7642857506871223,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857600003481,
      "rewards/format_reward_func": 1.0,
      "step": 10252
    },
    {
      "completion_length": 251.85269260406494,
      "epoch": 1.719225449515906,
      "grad_norm": 0.21025823635972254,
      "kl": 0.142669677734375,
      "learning_rate": 4.972358161639458e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286111295223,
      "rewards/format_reward_func": 1.0,
      "step": 10254
    },
    {
      "completion_length": 246.71429538726807,
      "epoch": 1.7195607527557735,
      "grad_norm": 0.21935861333026402,
      "kl": 0.0960845947265625,
      "learning_rate": 4.972340369762193e-07,
      "loss": 0.0001,
      "reward": 1.7821428999304771,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 10256
    },
    {
      "completion_length": 249.4509038925171,
      "epoch": 1.719896055995641,
      "grad_norm": 0.19696143606941816,
      "kl": 0.1159515380859375,
      "learning_rate": 4.972322572192686e-07,
      "loss": 0.0001,
      "reward": 1.785714328289032,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 10258
    },
    {
      "completion_length": 253.4955472946167,
      "epoch": 1.7202313592355085,
      "grad_norm": 0.2697506218673313,
      "kl": 0.1621856689453125,
      "learning_rate": 4.972304768930973e-07,
      "loss": 0.0002,
      "reward": 1.7750000581145287,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 10260
    },
    {
      "completion_length": 248.9821548461914,
      "epoch": 1.7205666624753762,
      "grad_norm": 0.15170022565279304,
      "kl": 0.102203369140625,
      "learning_rate": 4.9722869599771e-07,
      "loss": 0.0001,
      "reward": 1.8250000402331352,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8250000290572643,
      "rewards/format_reward_func": 1.0,
      "step": 10262
    },
    {
      "completion_length": 241.9017972946167,
      "epoch": 1.7209019657152438,
      "grad_norm": 0.19211303932289808,
      "kl": 0.105621337890625,
      "learning_rate": 4.972269145331106e-07,
      "loss": 0.0001,
      "reward": 1.7982143387198448,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8026785925030708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10264
    },
    {
      "completion_length": 254.39733505249023,
      "epoch": 1.7212372689551114,
      "grad_norm": 0.013160530116678611,
      "kl": 0.108795166015625,
      "learning_rate": 4.972251324993031e-07,
      "loss": 0.0001,
      "reward": 1.7946429252624512,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7991071790456772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10266
    },
    {
      "completion_length": 242.98661708831787,
      "epoch": 1.7215725721949788,
      "grad_norm": 0.15062515796873527,
      "kl": 0.085784912109375,
      "learning_rate": 4.972233498962917e-07,
      "loss": 0.0001,
      "reward": 1.7785714715719223,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 10268
    },
    {
      "completion_length": 252.28572463989258,
      "epoch": 1.7219078754348462,
      "grad_norm": 0.3019032739127625,
      "kl": 0.0868988037109375,
      "learning_rate": 4.972215667240805e-07,
      "loss": 0.0001,
      "reward": 1.7357143685221672,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.735714316368103,
      "rewards/format_reward_func": 1.0,
      "step": 10270
    },
    {
      "completion_length": 248.76340675354004,
      "epoch": 1.7222431786747139,
      "grad_norm": 0.17839336544905793,
      "kl": 0.0845184326171875,
      "learning_rate": 4.972197829826735e-07,
      "loss": 0.0001,
      "reward": 1.8053571581840515,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8098214566707611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10272
    },
    {
      "completion_length": 250.40179824829102,
      "epoch": 1.7225784819145815,
      "grad_norm": 0.4455969725900186,
      "kl": 0.11224365234375,
      "learning_rate": 4.97217998672075e-07,
      "loss": 0.0001,
      "reward": 1.758928619325161,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7633928935974836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10274
    },
    {
      "completion_length": 251.383939743042,
      "epoch": 1.7229137851544492,
      "grad_norm": 0.16785836779638325,
      "kl": 0.0938873291015625,
      "learning_rate": 4.972162137922888e-07,
      "loss": 0.0001,
      "reward": 1.7946429029107094,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7991071660071611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10276
    },
    {
      "completion_length": 243.18304538726807,
      "epoch": 1.7232490883943166,
      "grad_norm": 0.23542771845399904,
      "kl": 0.0963134765625,
      "learning_rate": 4.972144283433194e-07,
      "loss": 0.0001,
      "reward": 1.7785715162754059,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 10278
    },
    {
      "completion_length": 248.352689743042,
      "epoch": 1.7235843916341842,
      "grad_norm": 0.23453639466499993,
      "kl": 0.09661865234375,
      "learning_rate": 4.972126423251708e-07,
      "loss": 0.0001,
      "reward": 1.7825893387198448,
      "reward_std": 0.06502856989391148,
      "rewards/equation_reward_func": 0.7883929014205933,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 10280
    },
    {
      "completion_length": 249.73215675354004,
      "epoch": 1.7239196948740516,
      "grad_norm": 0.1695213510329949,
      "kl": 0.114288330078125,
      "learning_rate": 4.972108557378469e-07,
      "loss": 0.0001,
      "reward": 1.7901786267757416,
      "reward_std": 0.044194172602146864,
      "rewards/equation_reward_func": 0.7919643223285675,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 10282
    },
    {
      "completion_length": 247.0312623977661,
      "epoch": 1.7242549981139192,
      "grad_norm": 0.20108784450563666,
      "kl": 0.107269287109375,
      "learning_rate": 4.972090685813519e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 10284
    },
    {
      "completion_length": 248.40626525878906,
      "epoch": 1.7245903013537869,
      "grad_norm": 0.17376718166315555,
      "kl": 0.098052978515625,
      "learning_rate": 4.972072808556901e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 10286
    },
    {
      "completion_length": 253.43304920196533,
      "epoch": 1.7249256045936545,
      "grad_norm": 0.20889097778803672,
      "kl": 0.1026458740234375,
      "learning_rate": 4.972054925608654e-07,
      "loss": 0.0001,
      "reward": 1.7678571864962578,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571790456772,
      "rewards/format_reward_func": 1.0,
      "step": 10288
    },
    {
      "completion_length": 251.602689743042,
      "epoch": 1.725260907833522,
      "grad_norm": 0.1601895959215239,
      "kl": 0.0942535400390625,
      "learning_rate": 4.972037036968821e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 10290
    },
    {
      "completion_length": 252.18304634094238,
      "epoch": 1.7255962110733893,
      "grad_norm": 0.18190231757421382,
      "kl": 0.1245269775390625,
      "learning_rate": 4.972019142637442e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 10292
    },
    {
      "completion_length": 249.75894260406494,
      "epoch": 1.725931514313257,
      "grad_norm": 0.17078338524343523,
      "kl": 0.0865631103515625,
      "learning_rate": 4.972001242614558e-07,
      "loss": 0.0001,
      "reward": 1.773214340209961,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7776785884052515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10294
    },
    {
      "completion_length": 244.34375953674316,
      "epoch": 1.7262668175531246,
      "grad_norm": 0.2378874275442031,
      "kl": 0.1673583984375,
      "learning_rate": 4.97198333690021e-07,
      "loss": 0.0002,
      "reward": 1.769642896950245,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7741071879863739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10296
    },
    {
      "completion_length": 249.7098331451416,
      "epoch": 1.7266021207929922,
      "grad_norm": 0.21160423128606037,
      "kl": 0.120391845703125,
      "learning_rate": 4.971965425494439e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 10298
    },
    {
      "completion_length": 253.62947177886963,
      "epoch": 1.7269374240328599,
      "grad_norm": 0.20583500077418201,
      "kl": 0.18536376953125,
      "learning_rate": 4.97194750839729e-07,
      "loss": 0.0002,
      "reward": 1.7464286461472511,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286126196384,
      "rewards/format_reward_func": 1.0,
      "step": 10300
    },
    {
      "completion_length": 235.7901906967163,
      "epoch": 1.7272727272727273,
      "grad_norm": 0.3967885953207802,
      "kl": 0.1246337890625,
      "learning_rate": 4.971929585608799e-07,
      "loss": 0.0001,
      "reward": 1.7928571924567223,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571812808514,
      "rewards/format_reward_func": 1.0,
      "step": 10302
    },
    {
      "completion_length": 240.97768688201904,
      "epoch": 1.7276080305125947,
      "grad_norm": 0.46283659163232105,
      "kl": 0.155731201171875,
      "learning_rate": 4.97191165712901e-07,
      "loss": 0.0002,
      "reward": 1.742857240140438,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571693599224,
      "rewards/format_reward_func": 1.0,
      "step": 10304
    },
    {
      "completion_length": 241.07590675354004,
      "epoch": 1.7279433337524623,
      "grad_norm": 0.2614746911479949,
      "kl": 0.0865325927734375,
      "learning_rate": 4.971893722957964e-07,
      "loss": 0.0001,
      "reward": 1.8017857521772385,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.8062500171363354,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10306
    },
    {
      "completion_length": 230.19197463989258,
      "epoch": 1.72827863699233,
      "grad_norm": 0.18608370494522042,
      "kl": 0.1041259765625,
      "learning_rate": 4.971875783095702e-07,
      "loss": 0.0001,
      "reward": 1.803571492433548,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 10308
    },
    {
      "completion_length": 240.24108505249023,
      "epoch": 1.7286139402321976,
      "grad_norm": 0.26414739512639407,
      "kl": 0.112152099609375,
      "learning_rate": 4.971857837542266e-07,
      "loss": 0.0001,
      "reward": 1.8035714849829674,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8035714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 10310
    },
    {
      "completion_length": 241.75893878936768,
      "epoch": 1.728949243472065,
      "grad_norm": 0.2030729173493549,
      "kl": 0.12939453125,
      "learning_rate": 4.971839886297697e-07,
      "loss": 0.0001,
      "reward": 1.7785714715719223,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.778571454808116,
      "rewards/format_reward_func": 1.0,
      "step": 10312
    },
    {
      "completion_length": 229.4955472946167,
      "epoch": 1.7292845467119327,
      "grad_norm": 0.2002807067146816,
      "kl": 0.106903076171875,
      "learning_rate": 4.971821929362035e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 10314
    },
    {
      "completion_length": 241.39733409881592,
      "epoch": 1.7296198499518,
      "grad_norm": 0.22584604105987863,
      "kl": 0.0989837646484375,
      "learning_rate": 4.971803966735322e-07,
      "loss": 0.0001,
      "reward": 1.739285796880722,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 1.0,
      "step": 10316
    },
    {
      "completion_length": 245.0357265472412,
      "epoch": 1.7299551531916677,
      "grad_norm": 0.16363201642424538,
      "kl": 0.09918212890625,
      "learning_rate": 4.971785998417601e-07,
      "loss": 0.0001,
      "reward": 1.7732143327593803,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7776786014437675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10318
    },
    {
      "completion_length": 241.36608219146729,
      "epoch": 1.7302904564315353,
      "grad_norm": 0.11868823441642207,
      "kl": 0.119659423828125,
      "learning_rate": 4.971768024408912e-07,
      "loss": 0.0001,
      "reward": 1.7589286342263222,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7633928842842579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10320
    },
    {
      "completion_length": 246.12947368621826,
      "epoch": 1.730625759671403,
      "grad_norm": 0.4974318341941413,
      "kl": 0.11785888671875,
      "learning_rate": 4.971750044709296e-07,
      "loss": 0.0001,
      "reward": 1.7357143312692642,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 10322
    },
    {
      "completion_length": 243.7142972946167,
      "epoch": 1.7309610629112704,
      "grad_norm": 0.22373018764174843,
      "kl": 0.10406494140625,
      "learning_rate": 4.971732059318796e-07,
      "loss": 0.0001,
      "reward": 1.7678571939468384,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 10324
    },
    {
      "completion_length": 244.84822750091553,
      "epoch": 1.7312963661511378,
      "grad_norm": 0.3080666523543961,
      "kl": 0.157073974609375,
      "learning_rate": 4.971714068237452e-07,
      "loss": 0.0002,
      "reward": 1.7892857566475868,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 10326
    },
    {
      "completion_length": 247.28572750091553,
      "epoch": 1.7316316693910054,
      "grad_norm": 0.194548889979958,
      "kl": 0.18048095703125,
      "learning_rate": 4.971696071465305e-07,
      "loss": 0.0002,
      "reward": 1.7517857775092125,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500387430191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10328
    },
    {
      "completion_length": 255.8348331451416,
      "epoch": 1.731966972630873,
      "grad_norm": 0.10122060964439665,
      "kl": 0.1008148193359375,
      "learning_rate": 4.971678069002398e-07,
      "loss": 0.0001,
      "reward": 1.8250000476837158,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.825000025331974,
      "rewards/format_reward_func": 1.0,
      "step": 10330
    },
    {
      "completion_length": 258.7410821914673,
      "epoch": 1.7323022758707407,
      "grad_norm": 0.6716880283698804,
      "kl": 0.2233734130859375,
      "learning_rate": 4.971660060848772e-07,
      "loss": 0.0002,
      "reward": 1.8125000521540642,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8169643171131611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10332
    },
    {
      "completion_length": 260.14733505249023,
      "epoch": 1.7326375791106081,
      "grad_norm": 0.10600836730768504,
      "kl": 0.189788818359375,
      "learning_rate": 4.971642047004466e-07,
      "loss": 0.0002,
      "reward": 1.7250000685453415,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7250000294297934,
      "rewards/format_reward_func": 1.0,
      "step": 10334
    },
    {
      "completion_length": 253.477689743042,
      "epoch": 1.7329728823504758,
      "grad_norm": 0.13879610996945366,
      "kl": 0.160552978515625,
      "learning_rate": 4.971624027469526e-07,
      "loss": 0.0002,
      "reward": 1.7696429193019867,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7741071730852127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10336
    },
    {
      "completion_length": 257.62055015563965,
      "epoch": 1.7333081855903432,
      "grad_norm": 0.21084404479090876,
      "kl": 0.20538330078125,
      "learning_rate": 4.97160600224399e-07,
      "loss": 0.0002,
      "reward": 1.7928571999073029,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 10338
    },
    {
      "completion_length": 250.45090675354004,
      "epoch": 1.7336434888302108,
      "grad_norm": 0.26688576816393683,
      "kl": 0.146881103515625,
      "learning_rate": 4.971587971327901e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.782142873853445,
      "rewards/format_reward_func": 1.0,
      "step": 10340
    },
    {
      "completion_length": 256.0759038925171,
      "epoch": 1.7339787920700784,
      "grad_norm": 0.113620714880813,
      "kl": 0.16156005859375,
      "learning_rate": 4.9715699347213e-07,
      "loss": 0.0002,
      "reward": 1.7857143431901932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 10342
    },
    {
      "completion_length": 257.41072273254395,
      "epoch": 1.734314095309946,
      "grad_norm": 0.2530098783662599,
      "kl": 0.1668701171875,
      "learning_rate": 4.971551892424228e-07,
      "loss": 0.0002,
      "reward": 1.7357143387198448,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.735714316368103,
      "rewards/format_reward_func": 1.0,
      "step": 10344
    },
    {
      "completion_length": 258.1651916503906,
      "epoch": 1.7346493985498135,
      "grad_norm": 0.13072198172795174,
      "kl": 0.095489501953125,
      "learning_rate": 4.971533844436728e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 10346
    },
    {
      "completion_length": 265.5535840988159,
      "epoch": 1.734984701789681,
      "grad_norm": 0.13941893460031515,
      "kl": 0.114532470703125,
      "learning_rate": 4.97151579075884e-07,
      "loss": 0.0001,
      "reward": 1.7267858013510704,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7312500383704901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10348
    },
    {
      "completion_length": 268.7009029388428,
      "epoch": 1.7353200050295485,
      "grad_norm": 0.28658059038787587,
      "kl": 0.103271484375,
      "learning_rate": 4.971497731390607e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 10350
    },
    {
      "completion_length": 259.558048248291,
      "epoch": 1.7356553082694162,
      "grad_norm": 0.1752486314212871,
      "kl": 0.12921142578125,
      "learning_rate": 4.971479666332069e-07,
      "loss": 0.0001,
      "reward": 1.7732143476605415,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.777678593993187,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10352
    },
    {
      "completion_length": 275.13393783569336,
      "epoch": 1.7359906115092838,
      "grad_norm": 0.1740133493170833,
      "kl": 0.183502197265625,
      "learning_rate": 4.971461595583269e-07,
      "loss": 0.0002,
      "reward": 1.7375000566244125,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7419643104076385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10354
    },
    {
      "completion_length": 276.25447845458984,
      "epoch": 1.7363259147491512,
      "grad_norm": 0.4997503302559109,
      "kl": 0.151702880859375,
      "learning_rate": 4.971443519144248e-07,
      "loss": 0.0002,
      "reward": 1.758928619325161,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.763392886146903,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10356
    },
    {
      "completion_length": 268.64287090301514,
      "epoch": 1.7366612179890188,
      "grad_norm": 0.20603458803696528,
      "kl": 0.166961669921875,
      "learning_rate": 4.971425437015048e-07,
      "loss": 0.0002,
      "reward": 1.8267857804894447,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8312500268220901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10358
    },
    {
      "completion_length": 269.7678680419922,
      "epoch": 1.7369965212288863,
      "grad_norm": 0.1892884256617599,
      "kl": 0.1112518310546875,
      "learning_rate": 4.97140734919571e-07,
      "loss": 0.0001,
      "reward": 1.7821429371833801,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7910714484751225,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10360
    },
    {
      "completion_length": 267.5357265472412,
      "epoch": 1.737331824468754,
      "grad_norm": 0.21339591862795737,
      "kl": 0.096923828125,
      "learning_rate": 4.971389255686275e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.08586296625435352,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 10362
    },
    {
      "completion_length": 266.5714406967163,
      "epoch": 1.7376671277086215,
      "grad_norm": 0.15824648300184477,
      "kl": 0.1436004638671875,
      "learning_rate": 4.971371156486786e-07,
      "loss": 0.0001,
      "reward": 1.8250000402331352,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8250000365078449,
      "rewards/format_reward_func": 1.0,
      "step": 10364
    },
    {
      "completion_length": 266.89733600616455,
      "epoch": 1.7380024309484892,
      "grad_norm": 0.3613629986828252,
      "kl": 0.1967926025390625,
      "learning_rate": 4.971353051597285e-07,
      "loss": 0.0002,
      "reward": 1.791071504354477,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7955357320606709,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10366
    },
    {
      "completion_length": 260.7232265472412,
      "epoch": 1.7383377341883566,
      "grad_norm": 0.002761877959319243,
      "kl": 0.1133575439453125,
      "learning_rate": 4.971334941017813e-07,
      "loss": 0.0001,
      "reward": 1.7589285969734192,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7633929047733545,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10368
    },
    {
      "completion_length": 267.3794765472412,
      "epoch": 1.738673037428224,
      "grad_norm": 0.12438441265646877,
      "kl": 0.15130615234375,
      "learning_rate": 4.971316824748412e-07,
      "loss": 0.0002,
      "reward": 1.7464286535978317,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 10370
    },
    {
      "completion_length": 271.1116189956665,
      "epoch": 1.7390083406680916,
      "grad_norm": 0.25002979116110813,
      "kl": 0.125457763671875,
      "learning_rate": 4.971298702789123e-07,
      "loss": 0.0001,
      "reward": 1.7839286401867867,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7883928790688515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10372
    },
    {
      "completion_length": 275.9107246398926,
      "epoch": 1.7393436439079593,
      "grad_norm": 0.20122847648749992,
      "kl": 0.1316986083984375,
      "learning_rate": 4.971280575139988e-07,
      "loss": 0.0001,
      "reward": 1.7553572058677673,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.759821455925703,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10374
    },
    {
      "completion_length": 271.5223340988159,
      "epoch": 1.739678947147827,
      "grad_norm": 0.14811595712044243,
      "kl": 0.1340484619140625,
      "learning_rate": 4.971262441801048e-07,
      "loss": 0.0001,
      "reward": 1.7482143491506577,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.752678606659174,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10376
    },
    {
      "completion_length": 284.4330425262451,
      "epoch": 1.7400142503876945,
      "grad_norm": 0.22086490716191035,
      "kl": 0.118499755859375,
      "learning_rate": 4.971244302772346e-07,
      "loss": 0.0001,
      "reward": 1.787500061094761,
      "reward_std": 0.07828682195395231,
      "rewards/equation_reward_func": 0.8008928783237934,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10378
    },
    {
      "completion_length": 267.0089416503906,
      "epoch": 1.740349553627562,
      "grad_norm": 0.057972474967720546,
      "kl": 0.1278839111328125,
      "learning_rate": 4.971226158053923e-07,
      "loss": 0.0001,
      "reward": 1.7946429252624512,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7991071753203869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10380
    },
    {
      "completion_length": 282.9732275009155,
      "epoch": 1.7406848568674294,
      "grad_norm": 0.295008429979625,
      "kl": 0.219482421875,
      "learning_rate": 4.971208007645823e-07,
      "loss": 0.0002,
      "reward": 1.728571504354477,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.7375000305473804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10382
    },
    {
      "completion_length": 270.80804920196533,
      "epoch": 1.741020160107297,
      "grad_norm": 0.22125537829038144,
      "kl": 0.129486083984375,
      "learning_rate": 4.971189851548084e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7732143215835094,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10384
    },
    {
      "completion_length": 285.18304538726807,
      "epoch": 1.7413554633471646,
      "grad_norm": 0.5638466687686062,
      "kl": 0.160888671875,
      "learning_rate": 4.971171689760751e-07,
      "loss": 0.0002,
      "reward": 1.7625000402331352,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643200933933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10386
    },
    {
      "completion_length": 275.16965103149414,
      "epoch": 1.7416907665870323,
      "grad_norm": 0.2000542526716108,
      "kl": 0.141265869140625,
      "learning_rate": 4.971153522283864e-07,
      "loss": 0.0001,
      "reward": 1.7714286521077156,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7803571708500385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10388
    },
    {
      "completion_length": 273.29465675354004,
      "epoch": 1.7420260698268997,
      "grad_norm": 0.19184259176887247,
      "kl": 0.140655517578125,
      "learning_rate": 4.971135349117465e-07,
      "loss": 0.0001,
      "reward": 1.7428572326898575,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7517857402563095,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10390
    },
    {
      "completion_length": 267.9821538925171,
      "epoch": 1.7423613730667673,
      "grad_norm": 0.11815547024679297,
      "kl": 0.15802001953125,
      "learning_rate": 4.971117170261596e-07,
      "loss": 0.0002,
      "reward": 1.7714286372065544,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 10392
    },
    {
      "completion_length": 272.5803699493408,
      "epoch": 1.7426966763066347,
      "grad_norm": 0.24361023479353522,
      "kl": 0.267608642578125,
      "learning_rate": 4.9710989857163e-07,
      "loss": 0.0003,
      "reward": 1.7142858058214188,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7142857424914837,
      "rewards/format_reward_func": 1.0,
      "step": 10394
    },
    {
      "completion_length": 258.5982275009155,
      "epoch": 1.7430319795465024,
      "grad_norm": 0.24585877829799133,
      "kl": 0.13165283203125,
      "learning_rate": 4.971080795481618e-07,
      "loss": 0.0001,
      "reward": 1.8285714909434319,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8285714462399483,
      "rewards/format_reward_func": 1.0,
      "step": 10396
    },
    {
      "completion_length": 266.8125123977661,
      "epoch": 1.74336728278637,
      "grad_norm": 0.2220689272461111,
      "kl": 0.150238037109375,
      "learning_rate": 4.971062599557591e-07,
      "loss": 0.0002,
      "reward": 1.76071435213089,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 10398
    },
    {
      "completion_length": 260.52679920196533,
      "epoch": 1.7437025860262376,
      "grad_norm": 0.19939037132635878,
      "kl": 0.109222412109375,
      "learning_rate": 4.971044397944261e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143260538578,
      "rewards/format_reward_func": 1.0,
      "step": 10400
    },
    {
      "completion_length": 272.54911613464355,
      "epoch": 1.744037889266105,
      "grad_norm": 0.19888458583738894,
      "kl": 0.2293853759765625,
      "learning_rate": 4.971026190641672e-07,
      "loss": 0.0002,
      "reward": 1.7321429252624512,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428991854191,
      "rewards/format_reward_func": 1.0,
      "step": 10402
    },
    {
      "completion_length": 273.9732275009155,
      "epoch": 1.7443731925059724,
      "grad_norm": 0.27151938888768473,
      "kl": 0.178863525390625,
      "learning_rate": 4.971007977649864e-07,
      "loss": 0.0002,
      "reward": 1.785714365541935,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 10404
    },
    {
      "completion_length": 266.08930110931396,
      "epoch": 1.74470849574584,
      "grad_norm": 0.31683103034099785,
      "kl": 0.1129150390625,
      "learning_rate": 4.97098975896888e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 10406
    },
    {
      "completion_length": 267.5089406967163,
      "epoch": 1.7450437989857077,
      "grad_norm": 0.13888860982697956,
      "kl": 0.1087799072265625,
      "learning_rate": 4.970971534598761e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7767857424914837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10408
    },
    {
      "completion_length": 263.92411708831787,
      "epoch": 1.7453791022255754,
      "grad_norm": 0.27825671143278985,
      "kl": 0.143402099609375,
      "learning_rate": 4.970953304539549e-07,
      "loss": 0.0001,
      "reward": 1.7571429461240768,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428753435612,
      "rewards/format_reward_func": 1.0,
      "step": 10410
    },
    {
      "completion_length": 273.5178699493408,
      "epoch": 1.7457144054654428,
      "grad_norm": 0.08538777497485277,
      "kl": 0.151153564453125,
      "learning_rate": 4.970935068791286e-07,
      "loss": 0.0002,
      "reward": 1.7410714775323868,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7455357443541288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10412
    },
    {
      "completion_length": 270.0982275009155,
      "epoch": 1.7460497087053104,
      "grad_norm": 0.5390369677057072,
      "kl": 0.193695068359375,
      "learning_rate": 4.970916827354016e-07,
      "loss": 0.0002,
      "reward": 1.7732143476605415,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7866071835160255,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10414
    },
    {
      "completion_length": 261.7366199493408,
      "epoch": 1.7463850119451778,
      "grad_norm": 0.18971974736151406,
      "kl": 0.1143035888671875,
      "learning_rate": 4.970898580227778e-07,
      "loss": 0.0001,
      "reward": 1.8285714611411095,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8285714648663998,
      "rewards/format_reward_func": 1.0,
      "step": 10416
    },
    {
      "completion_length": 285.3214406967163,
      "epoch": 1.7467203151850454,
      "grad_norm": 0.3431656777052588,
      "kl": 0.220428466796875,
      "learning_rate": 4.970880327412616e-07,
      "loss": 0.0002,
      "reward": 1.7875000685453415,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7919643148779869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10418
    },
    {
      "completion_length": 276.5134029388428,
      "epoch": 1.747055618424913,
      "grad_norm": 0.14949047704968912,
      "kl": 0.2712860107421875,
      "learning_rate": 4.97086206890857e-07,
      "loss": 0.0003,
      "reward": 1.7553572058677673,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214484751225,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10420
    },
    {
      "completion_length": 265.5669755935669,
      "epoch": 1.7473909216647807,
      "grad_norm": 0.20053087897408742,
      "kl": 0.1323699951171875,
      "learning_rate": 4.970843804715684e-07,
      "loss": 0.0001,
      "reward": 1.7910714969038963,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.795535746961832,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10422
    },
    {
      "completion_length": 281.4196557998657,
      "epoch": 1.7477262249046481,
      "grad_norm": 0.34421321644086805,
      "kl": 0.195068359375,
      "learning_rate": 4.970825534834e-07,
      "loss": 0.0002,
      "reward": 1.7285715118050575,
      "reward_std": 0.12121830135583878,
      "rewards/equation_reward_func": 0.728571455925703,
      "rewards/format_reward_func": 1.0,
      "step": 10424
    },
    {
      "completion_length": 272.6919765472412,
      "epoch": 1.7480615281445155,
      "grad_norm": 0.3231745079674326,
      "kl": 0.154296875,
      "learning_rate": 4.970807259263559e-07,
      "loss": 0.0002,
      "reward": 1.7678572088479996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 10426
    },
    {
      "completion_length": 279.2589464187622,
      "epoch": 1.7483968313843832,
      "grad_norm": 0.17148280698992707,
      "kl": 0.165374755859375,
      "learning_rate": 4.970788978004404e-07,
      "loss": 0.0002,
      "reward": 1.7607143595814705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143260538578,
      "rewards/format_reward_func": 1.0,
      "step": 10428
    },
    {
      "completion_length": 275.589298248291,
      "epoch": 1.7487321346242508,
      "grad_norm": 0.25112206633806494,
      "kl": 0.146392822265625,
      "learning_rate": 4.970770691056577e-07,
      "loss": 0.0001,
      "reward": 1.7053572237491608,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7098214644938707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10430
    },
    {
      "completion_length": 265.3437614440918,
      "epoch": 1.7490674378641184,
      "grad_norm": 0.17369490573789476,
      "kl": 0.12432861328125,
      "learning_rate": 4.97075239842012e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 10432
    },
    {
      "completion_length": 273.96429443359375,
      "epoch": 1.749402741103986,
      "grad_norm": 0.24760782391253822,
      "kl": 0.1802978515625,
      "learning_rate": 4.970734100095073e-07,
      "loss": 0.0002,
      "reward": 1.7446429207921028,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7580357417464256,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10434
    },
    {
      "completion_length": 272.47322845458984,
      "epoch": 1.7497380443438535,
      "grad_norm": 0.20643982792458254,
      "kl": 0.1326141357421875,
      "learning_rate": 4.970715796081482e-07,
      "loss": 0.0001,
      "reward": 1.817857176065445,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.817857164889574,
      "rewards/format_reward_func": 1.0,
      "step": 10436
    },
    {
      "completion_length": 276.3794746398926,
      "epoch": 1.750073347583721,
      "grad_norm": 0.10986143055110638,
      "kl": 0.2333984375,
      "learning_rate": 4.970697486379386e-07,
      "loss": 0.0002,
      "reward": 1.7446429207921028,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10438
    },
    {
      "completion_length": 257.4509029388428,
      "epoch": 1.7504086508235885,
      "grad_norm": 0.19597224889941217,
      "kl": 0.110137939453125,
      "learning_rate": 4.970679170988829e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285850524902,
      "rewards/format_reward_func": 1.0,
      "step": 10440
    },
    {
      "completion_length": 261.3794746398926,
      "epoch": 1.7507439540634562,
      "grad_norm": 0.32525695998592424,
      "kl": 0.164581298828125,
      "learning_rate": 4.970660849909852e-07,
      "loss": 0.0002,
      "reward": 1.7642857879400253,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 10442
    },
    {
      "completion_length": 264.62500858306885,
      "epoch": 1.7510792573033238,
      "grad_norm": 0.2660366832229654,
      "kl": 0.09619140625,
      "learning_rate": 4.970642523142498e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 10444
    },
    {
      "completion_length": 271.4866180419922,
      "epoch": 1.7514145605431912,
      "grad_norm": 0.2229115237298831,
      "kl": 0.159698486328125,
      "learning_rate": 4.970624190686808e-07,
      "loss": 0.0002,
      "reward": 1.773214340209961,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7776786126196384,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10446
    },
    {
      "completion_length": 255.22322750091553,
      "epoch": 1.7517498637830589,
      "grad_norm": 0.34719989862558487,
      "kl": 0.1978912353515625,
      "learning_rate": 4.970605852542826e-07,
      "loss": 0.0002,
      "reward": 1.8125000447034836,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8169643171131611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10448
    },
    {
      "completion_length": 264.40626335144043,
      "epoch": 1.7520851670229263,
      "grad_norm": 0.19374510192625774,
      "kl": 0.1017913818359375,
      "learning_rate": 4.970587508710593e-07,
      "loss": 0.0001,
      "reward": 1.74642863124609,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464285995811224,
      "rewards/format_reward_func": 1.0,
      "step": 10450
    },
    {
      "completion_length": 267.5491199493408,
      "epoch": 1.752420470262794,
      "grad_norm": 0.18141824950577376,
      "kl": 0.22296142578125,
      "learning_rate": 4.970569159190152e-07,
      "loss": 0.0002,
      "reward": 1.742857202887535,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571842610836,
      "rewards/format_reward_func": 1.0,
      "step": 10452
    },
    {
      "completion_length": 257.7500123977661,
      "epoch": 1.7527557735026615,
      "grad_norm": 0.00973250312477692,
      "kl": 0.120849609375,
      "learning_rate": 4.970550803981544e-07,
      "loss": 0.0001,
      "reward": 1.7250000685453415,
      "reward_std": 0.005050762556493282,
      "rewards/equation_reward_func": 0.7250000536441803,
      "rewards/format_reward_func": 1.0,
      "step": 10454
    },
    {
      "completion_length": 264.75447845458984,
      "epoch": 1.7530910767425292,
      "grad_norm": 0.16646004320135122,
      "kl": 0.0950927734375,
      "learning_rate": 4.970532443084812e-07,
      "loss": 0.0001,
      "reward": 1.7553571984171867,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7589286118745804,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 10456
    },
    {
      "completion_length": 272.44197368621826,
      "epoch": 1.7534263799823966,
      "grad_norm": 0.2769682286353471,
      "kl": 0.12298583984375,
      "learning_rate": 4.970514076499999e-07,
      "loss": 0.0001,
      "reward": 1.7678571939468384,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.776785746216774,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10458
    },
    {
      "completion_length": 259.90626430511475,
      "epoch": 1.753761683222264,
      "grad_norm": 0.20317105516113643,
      "kl": 0.0985107421875,
      "learning_rate": 4.970495704227146e-07,
      "loss": 0.0001,
      "reward": 1.7928572222590446,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 10460
    },
    {
      "completion_length": 267.4732275009155,
      "epoch": 1.7540969864621316,
      "grad_norm": 0.28731729589505295,
      "kl": 0.119537353515625,
      "learning_rate": 4.970477326266297e-07,
      "loss": 0.0001,
      "reward": 1.7178572490811348,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7178571820259094,
      "rewards/format_reward_func": 1.0,
      "step": 10462
    },
    {
      "completion_length": 258.66072940826416,
      "epoch": 1.7544322897019993,
      "grad_norm": 0.37070153761318986,
      "kl": 0.100830078125,
      "learning_rate": 4.970458942617493e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 10464
    },
    {
      "completion_length": 271.5000123977661,
      "epoch": 1.754767592941867,
      "grad_norm": 0.20976854165769124,
      "kl": 0.156036376953125,
      "learning_rate": 4.970440553280776e-07,
      "loss": 0.0002,
      "reward": 1.7732143551111221,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7776785865426064,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10466
    },
    {
      "completion_length": 261.25893783569336,
      "epoch": 1.7551028961817343,
      "grad_norm": 0.3999970736225111,
      "kl": 0.1058502197265625,
      "learning_rate": 4.970422158256188e-07,
      "loss": 0.0001,
      "reward": 1.762500062584877,
      "reward_std": 0.09343910776078701,
      "rewards/equation_reward_func": 0.7669643014669418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10468
    },
    {
      "completion_length": 270.8616189956665,
      "epoch": 1.755438199421602,
      "grad_norm": 0.2932392086982285,
      "kl": 0.102630615234375,
      "learning_rate": 4.970403757543773e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714626312256,
      "rewards/format_reward_func": 1.0,
      "step": 10470
    },
    {
      "completion_length": 258.6919755935669,
      "epoch": 1.7557735026614694,
      "grad_norm": 0.7825140351594014,
      "kl": 0.227874755859375,
      "learning_rate": 4.970385351143573e-07,
      "loss": 0.0002,
      "reward": 1.8178571835160255,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8178571574389935,
      "rewards/format_reward_func": 1.0,
      "step": 10472
    },
    {
      "completion_length": 265.7009048461914,
      "epoch": 1.756108805901337,
      "grad_norm": 0.23065535913563237,
      "kl": 0.1201171875,
      "learning_rate": 4.97036693905563e-07,
      "loss": 0.0001,
      "reward": 1.800000049173832,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 10474
    },
    {
      "completion_length": 255.95536708831787,
      "epoch": 1.7564441091412046,
      "grad_norm": 0.35248789588895424,
      "kl": 0.3164215087890625,
      "learning_rate": 4.970348521279986e-07,
      "loss": 0.0003,
      "reward": 1.8142857626080513,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.823214303702116,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10476
    },
    {
      "completion_length": 261.8125123977661,
      "epoch": 1.7567794123810723,
      "grad_norm": 0.19963963290351514,
      "kl": 0.1441192626953125,
      "learning_rate": 4.970330097816683e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 10478
    },
    {
      "completion_length": 270.3884057998657,
      "epoch": 1.7571147156209397,
      "grad_norm": 0.23056821333387442,
      "kl": 0.110260009765625,
      "learning_rate": 4.970311668665766e-07,
      "loss": 0.0001,
      "reward": 1.764285758137703,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 10480
    },
    {
      "completion_length": 266.4776916503906,
      "epoch": 1.757450018860807,
      "grad_norm": 0.1639649340695387,
      "kl": 0.137908935546875,
      "learning_rate": 4.970293233827274e-07,
      "loss": 0.0001,
      "reward": 1.6678572073578835,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.676785746589303,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10482
    },
    {
      "completion_length": 257.2053699493408,
      "epoch": 1.7577853221006747,
      "grad_norm": 0.20303344762139539,
      "kl": 0.107940673828125,
      "learning_rate": 4.970274793301252e-07,
      "loss": 0.0001,
      "reward": 1.7678572237491608,
      "reward_std": 0.04545686487108469,
      "rewards/equation_reward_func": 0.7767857424914837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10484
    },
    {
      "completion_length": 274.4018020629883,
      "epoch": 1.7581206253405424,
      "grad_norm": 0.261607313743246,
      "kl": 0.113311767578125,
      "learning_rate": 4.970256347087741e-07,
      "loss": 0.0001,
      "reward": 1.789285771548748,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 10486
    },
    {
      "completion_length": 258.8750114440918,
      "epoch": 1.75845592858041,
      "grad_norm": 0.48522185192261696,
      "kl": 0.1284637451171875,
      "learning_rate": 4.970237895186784e-07,
      "loss": 0.0001,
      "reward": 1.8339286223053932,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8383928686380386,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10488
    },
    {
      "completion_length": 273.1071529388428,
      "epoch": 1.7587912318202774,
      "grad_norm": 0.18348579583019425,
      "kl": 0.12353515625,
      "learning_rate": 4.970219437598423e-07,
      "loss": 0.0001,
      "reward": 1.7428572177886963,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.751785721629858,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10490
    },
    {
      "completion_length": 270.8884057998657,
      "epoch": 1.759126535060145,
      "grad_norm": 0.3575486729727749,
      "kl": 0.1207275390625,
      "learning_rate": 4.970200974322702e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7767857499420643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10492
    },
    {
      "completion_length": 269.8750114440918,
      "epoch": 1.7594618383000125,
      "grad_norm": 0.2593350581842222,
      "kl": 0.160858154296875,
      "learning_rate": 4.970182505359662e-07,
      "loss": 0.0002,
      "reward": 1.7571429088711739,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 10494
    },
    {
      "completion_length": 266.58929920196533,
      "epoch": 1.75979714153988,
      "grad_norm": 0.2131207192247197,
      "kl": 0.10052490234375,
      "learning_rate": 4.970164030709346e-07,
      "loss": 0.0001,
      "reward": 1.7625000774860382,
      "reward_std": 0.06313453428447247,
      "rewards/equation_reward_func": 0.7758928798139095,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10496
    },
    {
      "completion_length": 262.8928680419922,
      "epoch": 1.7601324447797477,
      "grad_norm": 0.23990261996700582,
      "kl": 0.1071319580078125,
      "learning_rate": 4.970145550371797e-07,
      "loss": 0.0001,
      "reward": 1.7303572222590446,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7348214536905289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10498
    },
    {
      "completion_length": 265.95090770721436,
      "epoch": 1.7604677480196154,
      "grad_norm": 0.28258834064510424,
      "kl": 0.177490234375,
      "learning_rate": 4.970127064347056e-07,
      "loss": 0.0002,
      "reward": 1.7017857879400253,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7062500268220901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10500
    },
    {
      "completion_length": 255.3259038925171,
      "epoch": 1.7608030512594828,
      "grad_norm": 0.25810737442998816,
      "kl": 0.099578857421875,
      "learning_rate": 4.970108572635168e-07,
      "loss": 0.0001,
      "reward": 1.825000062584877,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8250000141561031,
      "rewards/format_reward_func": 1.0,
      "step": 10502
    },
    {
      "completion_length": 255.51787185668945,
      "epoch": 1.7611383544993502,
      "grad_norm": 0.309821970825553,
      "kl": 0.150054931640625,
      "learning_rate": 4.970090075236173e-07,
      "loss": 0.0002,
      "reward": 1.7303572073578835,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7437500394880772,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10504
    },
    {
      "completion_length": 260.08929920196533,
      "epoch": 1.7614736577392178,
      "grad_norm": 0.24711570040427921,
      "kl": 0.1146240234375,
      "learning_rate": 4.970071572150116e-07,
      "loss": 0.0001,
      "reward": 1.7982143238186836,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.8026785999536514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10506
    },
    {
      "completion_length": 260.5982265472412,
      "epoch": 1.7618089609790855,
      "grad_norm": 0.18349956303515857,
      "kl": 0.11907958984375,
      "learning_rate": 4.970053063377037e-07,
      "loss": 0.0001,
      "reward": 1.7285714969038963,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714633762836,
      "rewards/format_reward_func": 1.0,
      "step": 10508
    },
    {
      "completion_length": 245.54911994934082,
      "epoch": 1.762144264218953,
      "grad_norm": 0.07936488872801638,
      "kl": 0.09381103515625,
      "learning_rate": 4.97003454891698e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 10510
    },
    {
      "completion_length": 255.3884048461914,
      "epoch": 1.7624795674588207,
      "grad_norm": 0.09023603463700747,
      "kl": 0.0997314453125,
      "learning_rate": 4.970016028769989e-07,
      "loss": 0.0001,
      "reward": 1.7625000476837158,
      "reward_std": 0.022728432901203632,
      "rewards/equation_reward_func": 0.7669643312692642,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10512
    },
    {
      "completion_length": 251.54465675354004,
      "epoch": 1.7628148706986881,
      "grad_norm": 0.14167783274190454,
      "kl": 0.120635986328125,
      "learning_rate": 4.969997502936105e-07,
      "loss": 0.0001,
      "reward": 1.7375000789761543,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7419643178582191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10514
    },
    {
      "completion_length": 238.62054634094238,
      "epoch": 1.7631501739385556,
      "grad_norm": 0.21653483508274973,
      "kl": 0.100006103515625,
      "learning_rate": 4.96997897141537e-07,
      "loss": 0.0001,
      "reward": 1.8071429282426834,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428947150707,
      "rewards/format_reward_func": 1.0,
      "step": 10516
    },
    {
      "completion_length": 244.4732265472412,
      "epoch": 1.7634854771784232,
      "grad_norm": 0.21315796766163686,
      "kl": 0.09234619140625,
      "learning_rate": 4.969960434207828e-07,
      "loss": 0.0001,
      "reward": 1.7383929342031479,
      "reward_std": 0.056821079924702644,
      "rewards/equation_reward_func": 0.7446428798139095,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 10518
    },
    {
      "completion_length": 247.1517972946167,
      "epoch": 1.7638207804182908,
      "grad_norm": 0.18540650627396144,
      "kl": 0.097076416015625,
      "learning_rate": 4.969941891313522e-07,
      "loss": 0.0001,
      "reward": 1.848214328289032,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8526785857975483,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10520
    },
    {
      "completion_length": 245.6741180419922,
      "epoch": 1.7641560836581585,
      "grad_norm": 0.16238503100743001,
      "kl": 0.101806640625,
      "learning_rate": 4.969923342732493e-07,
      "loss": 0.0001,
      "reward": 1.7982143387198448,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.8026786036789417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10522
    },
    {
      "completion_length": 251.41072750091553,
      "epoch": 1.7644913868980259,
      "grad_norm": 0.435907843732035,
      "kl": 0.09576416015625,
      "learning_rate": 4.969904788464786e-07,
      "loss": 0.0001,
      "reward": 1.7392858266830444,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 1.0,
      "step": 10524
    },
    {
      "completion_length": 253.42411708831787,
      "epoch": 1.7648266901378935,
      "grad_norm": 0.18620437703841827,
      "kl": 0.0916748046875,
      "learning_rate": 4.969886228510442e-07,
      "loss": 0.0001,
      "reward": 1.7285714894533157,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7375000268220901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10526
    },
    {
      "completion_length": 251.94197845458984,
      "epoch": 1.765161993377761,
      "grad_norm": 0.2205986553207691,
      "kl": 0.0987396240234375,
      "learning_rate": 4.969867662869503e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 10528
    },
    {
      "completion_length": 246.54019165039062,
      "epoch": 1.7654972966176286,
      "grad_norm": 0.1830755227045685,
      "kl": 0.120941162109375,
      "learning_rate": 4.969849091542014e-07,
      "loss": 0.0001,
      "reward": 1.8000000640749931,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000193715096,
      "rewards/format_reward_func": 1.0,
      "step": 10530
    },
    {
      "completion_length": 240.3125123977661,
      "epoch": 1.7658325998574962,
      "grad_norm": 0.15024188183011833,
      "kl": 0.1271514892578125,
      "learning_rate": 4.969830514528016e-07,
      "loss": 0.0001,
      "reward": 1.7607143372297287,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.7696428839117289,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10532
    },
    {
      "completion_length": 238.25447368621826,
      "epoch": 1.7661679030973638,
      "grad_norm": 0.149959763615365,
      "kl": 0.091094970703125,
      "learning_rate": 4.969811931827552e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 10534
    },
    {
      "completion_length": 246.3928689956665,
      "epoch": 1.7665032063372312,
      "grad_norm": 0.3019943513178325,
      "kl": 0.098297119140625,
      "learning_rate": 4.969793343440666e-07,
      "loss": 0.0001,
      "reward": 1.8035714998841286,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 10536
    },
    {
      "completion_length": 238.12500858306885,
      "epoch": 1.7668385095770986,
      "grad_norm": 0.3349463351982298,
      "kl": 0.09002685546875,
      "learning_rate": 4.969774749367401e-07,
      "loss": 0.0001,
      "reward": 1.7446429282426834,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10538
    },
    {
      "completion_length": 232.12054538726807,
      "epoch": 1.7671738128169663,
      "grad_norm": 0.36886859954996376,
      "kl": 0.0915985107421875,
      "learning_rate": 4.969756149607796e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 10540
    },
    {
      "completion_length": 242.53572750091553,
      "epoch": 1.767509116056834,
      "grad_norm": 0.2734354801264316,
      "kl": 0.1054229736328125,
      "learning_rate": 4.969737544161899e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857357859612,
      "rewards/format_reward_func": 1.0,
      "step": 10542
    },
    {
      "completion_length": 246.2500114440918,
      "epoch": 1.7678444192967016,
      "grad_norm": 0.004447381062977461,
      "kl": 0.0968017578125,
      "learning_rate": 4.96971893302975e-07,
      "loss": 0.0001,
      "reward": 1.6928572058677673,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.6928571686148643,
      "rewards/format_reward_func": 1.0,
      "step": 10544
    },
    {
      "completion_length": 237.5000123977661,
      "epoch": 1.768179722536569,
      "grad_norm": 0.1948318054506246,
      "kl": 0.115692138671875,
      "learning_rate": 4.969700316211392e-07,
      "loss": 0.0001,
      "reward": 1.814285784959793,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857402563095,
      "rewards/format_reward_func": 1.0,
      "step": 10546
    },
    {
      "completion_length": 239.33929634094238,
      "epoch": 1.7685150257764366,
      "grad_norm": 0.3184589691220786,
      "kl": 0.099212646484375,
      "learning_rate": 4.969681693706868e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 10548
    },
    {
      "completion_length": 235.21429538726807,
      "epoch": 1.768850329016304,
      "grad_norm": 0.20532648968982828,
      "kl": 0.129791259765625,
      "learning_rate": 4.969663065516222e-07,
      "loss": 0.0001,
      "reward": 1.7803572341799736,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7839286029338837,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 10550
    },
    {
      "completion_length": 231.64286708831787,
      "epoch": 1.7691856322561716,
      "grad_norm": 0.24991697110021083,
      "kl": 0.096435546875,
      "learning_rate": 4.969644431639495e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 10552
    },
    {
      "completion_length": 234.38393878936768,
      "epoch": 1.7695209354960393,
      "grad_norm": 0.21260438942552592,
      "kl": 0.11773681640625,
      "learning_rate": 4.969625792076731e-07,
      "loss": 0.0001,
      "reward": 1.692857213318348,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.6928571686148643,
      "rewards/format_reward_func": 1.0,
      "step": 10554
    },
    {
      "completion_length": 238.30358123779297,
      "epoch": 1.769856238735907,
      "grad_norm": 0.1314157473470005,
      "kl": 0.099853515625,
      "learning_rate": 4.969607146827972e-07,
      "loss": 0.0001,
      "reward": 1.789285771548748,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 10556
    },
    {
      "completion_length": 234.6071548461914,
      "epoch": 1.7701915419757743,
      "grad_norm": 0.12255947847610604,
      "kl": 0.11346435546875,
      "learning_rate": 4.969588495893263e-07,
      "loss": 0.0001,
      "reward": 1.8017857670783997,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8053571842610836,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 10558
    },
    {
      "completion_length": 232.83483028411865,
      "epoch": 1.7705268452156417,
      "grad_norm": 0.3855074347280174,
      "kl": 0.141265869140625,
      "learning_rate": 4.969569839272645e-07,
      "loss": 0.0001,
      "reward": 1.7089286670088768,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7125000320374966,
      "rewards/format_reward_func": 0.9964285790920258,
      "step": 10560
    },
    {
      "completion_length": 233.3303680419922,
      "epoch": 1.7708621484555094,
      "grad_norm": 0.18710436318418036,
      "kl": 0.1106109619140625,
      "learning_rate": 4.969551176966162e-07,
      "loss": 0.0001,
      "reward": 1.7928572297096252,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 10562
    },
    {
      "completion_length": 237.5669765472412,
      "epoch": 1.771197451695377,
      "grad_norm": 0.2694915854598506,
      "kl": 0.1116485595703125,
      "learning_rate": 4.969532508973856e-07,
      "loss": 0.0001,
      "reward": 1.8071429058909416,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428909897804,
      "rewards/format_reward_func": 1.0,
      "step": 10564
    },
    {
      "completion_length": 249.62054634094238,
      "epoch": 1.7715327549352446,
      "grad_norm": 0.18554608403376768,
      "kl": 0.2052001953125,
      "learning_rate": 4.969513835295771e-07,
      "loss": 0.0002,
      "reward": 1.767857201397419,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571790456772,
      "rewards/format_reward_func": 1.0,
      "step": 10566
    },
    {
      "completion_length": 248.68751049041748,
      "epoch": 1.7718680581751123,
      "grad_norm": 0.14565025528136757,
      "kl": 0.138702392578125,
      "learning_rate": 4.96949515593195e-07,
      "loss": 0.0001,
      "reward": 1.732142947614193,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7321428805589676,
      "rewards/format_reward_func": 1.0,
      "step": 10568
    },
    {
      "completion_length": 260.7232275009155,
      "epoch": 1.7722033614149797,
      "grad_norm": 0.2558570624218233,
      "kl": 0.375213623046875,
      "learning_rate": 4.969476470882435e-07,
      "loss": 0.0004,
      "reward": 1.7821428999304771,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428868919611,
      "rewards/format_reward_func": 1.0,
      "step": 10570
    },
    {
      "completion_length": 255.16072463989258,
      "epoch": 1.772538664654847,
      "grad_norm": 0.2710318747661508,
      "kl": 0.5247802734375,
      "learning_rate": 4.969457780147268e-07,
      "loss": 0.0005,
      "reward": 1.717857226729393,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7178571671247482,
      "rewards/format_reward_func": 1.0,
      "step": 10572
    },
    {
      "completion_length": 239.89733219146729,
      "epoch": 1.7728739678947147,
      "grad_norm": 0.1223392200473689,
      "kl": 0.10552978515625,
      "learning_rate": 4.969439083726496e-07,
      "loss": 0.0001,
      "reward": 1.8357143327593803,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8357143178582191,
      "rewards/format_reward_func": 1.0,
      "step": 10574
    },
    {
      "completion_length": 257.7053689956665,
      "epoch": 1.7732092711345824,
      "grad_norm": 0.3783049950951984,
      "kl": 0.526641845703125,
      "learning_rate": 4.969420381620158e-07,
      "loss": 0.0005,
      "reward": 1.7339286506175995,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7473214641213417,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10576
    },
    {
      "completion_length": 256.7142972946167,
      "epoch": 1.77354457437445,
      "grad_norm": 0.23260212557444576,
      "kl": 0.114898681640625,
      "learning_rate": 4.9694016738283e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 10578
    },
    {
      "completion_length": 254.59376335144043,
      "epoch": 1.7738798776143174,
      "grad_norm": 0.31166389144220735,
      "kl": 0.380523681640625,
      "learning_rate": 4.969382960350962e-07,
      "loss": 0.0004,
      "reward": 1.7821429073810577,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 10580
    },
    {
      "completion_length": 270.8705472946167,
      "epoch": 1.774215180854185,
      "grad_norm": 0.16185259134789332,
      "kl": 0.2837677001953125,
      "learning_rate": 4.969364241188191e-07,
      "loss": 0.0003,
      "reward": 1.6928572282195091,
      "reward_std": 0.08081220462918282,
      "rewards/equation_reward_func": 0.7196428887546062,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 10582
    },
    {
      "completion_length": 262.33036708831787,
      "epoch": 1.7745504840940525,
      "grad_norm": 0.20630636229383023,
      "kl": 0.138214111328125,
      "learning_rate": 4.969345516340026e-07,
      "loss": 0.0001,
      "reward": 1.7732143551111221,
      "reward_std": 0.10354063473641872,
      "rewards/equation_reward_func": 0.8044643104076385,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 10584
    },
    {
      "completion_length": 266.24108505249023,
      "epoch": 1.77488578733392,
      "grad_norm": 0.22700971231892747,
      "kl": 0.325653076171875,
      "learning_rate": 4.969326785806513e-07,
      "loss": 0.0003,
      "reward": 1.7053571864962578,
      "reward_std": 0.123743686825037,
      "rewards/equation_reward_func": 0.736607177183032,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 10586
    },
    {
      "completion_length": 255.2634048461914,
      "epoch": 1.7752210905737877,
      "grad_norm": 0.31500744912161194,
      "kl": 0.1110076904296875,
      "learning_rate": 4.969308049587694e-07,
      "loss": 0.0001,
      "reward": 1.7071429044008255,
      "reward_std": 0.08081220369786024,
      "rewards/equation_reward_func": 0.7250000312924385,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 10588
    },
    {
      "completion_length": 269.6607275009155,
      "epoch": 1.7755563938136554,
      "grad_norm": 0.29098020918540685,
      "kl": 0.1125640869140625,
      "learning_rate": 4.969289307683612e-07,
      "loss": 0.0001,
      "reward": 1.698214367032051,
      "reward_std": 0.10354063659906387,
      "rewards/equation_reward_func": 0.7205357439815998,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 10590
    },
    {
      "completion_length": 266.28125953674316,
      "epoch": 1.7758916970535228,
      "grad_norm": 0.3319292014716968,
      "kl": 0.608734130859375,
      "learning_rate": 4.969270560094311e-07,
      "loss": 0.0006,
      "reward": 1.6875000596046448,
      "reward_std": 0.1085914010182023,
      "rewards/equation_reward_func": 0.7187500447034836,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 10592
    },
    {
      "completion_length": 261.5401945114136,
      "epoch": 1.7762270002933902,
      "grad_norm": 0.26305753900705553,
      "kl": 0.69952392578125,
      "learning_rate": 4.969251806819834e-07,
      "loss": 0.0007,
      "reward": 1.7392857745289803,
      "reward_std": 0.10606601648032665,
      "rewards/equation_reward_func": 0.7660714574158192,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 10594
    },
    {
      "completion_length": 256.352689743042,
      "epoch": 1.7765623035332578,
      "grad_norm": 0.11624276670860509,
      "kl": 0.237396240234375,
      "learning_rate": 4.969233047860223e-07,
      "loss": 0.0002,
      "reward": 1.73214291036129,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7491071708500385,
      "rewards/format_reward_func": 0.9830357208848,
      "step": 10596
    },
    {
      "completion_length": 266.1116180419922,
      "epoch": 1.7768976067731255,
      "grad_norm": 0.24888262657285953,
      "kl": 0.37750244140625,
      "learning_rate": 4.969214283215523e-07,
      "loss": 0.0004,
      "reward": 1.700000062584877,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7089285850524902,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10598
    },
    {
      "completion_length": 255.25894165039062,
      "epoch": 1.777232910012993,
      "grad_norm": 0.3873078676136931,
      "kl": 0.159027099609375,
      "learning_rate": 4.969195512885775e-07,
      "loss": 0.0002,
      "reward": 1.757142923772335,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428753435612,
      "rewards/format_reward_func": 1.0,
      "step": 10600
    },
    {
      "completion_length": 258.6294775009155,
      "epoch": 1.7775682132528605,
      "grad_norm": 0.16038748989821341,
      "kl": 0.148468017578125,
      "learning_rate": 4.969176736871024e-07,
      "loss": 0.0001,
      "reward": 1.7142858132719994,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857499420643,
      "rewards/format_reward_func": 1.0,
      "step": 10602
    },
    {
      "completion_length": 252.9553680419922,
      "epoch": 1.7779035164927282,
      "grad_norm": 0.2520951190407464,
      "kl": 0.157440185546875,
      "learning_rate": 4.969157955171313e-07,
      "loss": 0.0002,
      "reward": 1.7000000476837158,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.708928607404232,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10604
    },
    {
      "completion_length": 247.06697463989258,
      "epoch": 1.7782388197325956,
      "grad_norm": 0.26364201619762373,
      "kl": 0.3067626953125,
      "learning_rate": 4.969139167786684e-07,
      "loss": 0.0003,
      "reward": 1.7928572222590446,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 10606
    },
    {
      "completion_length": 245.12947463989258,
      "epoch": 1.7785741229724632,
      "grad_norm": 0.1645764154368642,
      "kl": 0.153106689453125,
      "learning_rate": 4.969120374717182e-07,
      "loss": 0.0002,
      "reward": 1.7678572162985802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571790456772,
      "rewards/format_reward_func": 1.0,
      "step": 10608
    },
    {
      "completion_length": 237.33929538726807,
      "epoch": 1.7789094262123308,
      "grad_norm": 0.16426994973474854,
      "kl": 0.1124267578125,
      "learning_rate": 4.969101575962849e-07,
      "loss": 0.0001,
      "reward": 1.8178572058677673,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8178571611642838,
      "rewards/format_reward_func": 1.0,
      "step": 10610
    },
    {
      "completion_length": 247.6964406967163,
      "epoch": 1.7792447294521985,
      "grad_norm": 0.182551150654352,
      "kl": 0.1142730712890625,
      "learning_rate": 4.969082771523728e-07,
      "loss": 0.0001,
      "reward": 1.7982143461704254,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8026785962283611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10612
    },
    {
      "completion_length": 247.01786518096924,
      "epoch": 1.7795800326920659,
      "grad_norm": 0.23539622917480219,
      "kl": 0.1351470947265625,
      "learning_rate": 4.969063961399865e-07,
      "loss": 0.0001,
      "reward": 1.7866071984171867,
      "reward_std": 0.049244935624301434,
      "rewards/equation_reward_func": 0.7883928902447224,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 10614
    },
    {
      "completion_length": 255.25893878936768,
      "epoch": 1.7799153359319333,
      "grad_norm": 0.20134669385634055,
      "kl": 0.119110107421875,
      "learning_rate": 4.9690451455913e-07,
      "loss": 0.0001,
      "reward": 1.7660714983940125,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7705357372760773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10616
    },
    {
      "completion_length": 240.79911708831787,
      "epoch": 1.780250639171801,
      "grad_norm": 0.07603412774145471,
      "kl": 0.10906982421875,
      "learning_rate": 4.969026324098076e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 10618
    },
    {
      "completion_length": 249.82590293884277,
      "epoch": 1.7805859424116686,
      "grad_norm": 0.26972691648450575,
      "kl": 0.1138153076171875,
      "learning_rate": 4.96900749692024e-07,
      "loss": 0.0001,
      "reward": 1.7776786237955093,
      "reward_std": 0.07197336852550507,
      "rewards/equation_reward_func": 0.7839285917580128,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 10620
    },
    {
      "completion_length": 239.44197750091553,
      "epoch": 1.7809212456515362,
      "grad_norm": 0.2305440981797223,
      "kl": 0.130126953125,
      "learning_rate": 4.968988664057834e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 10622
    },
    {
      "completion_length": 243.85268878936768,
      "epoch": 1.7812565488914036,
      "grad_norm": 0.26768835156722426,
      "kl": 0.092803955078125,
      "learning_rate": 4.968969825510899e-07,
      "loss": 0.0001,
      "reward": 1.7964286357164383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 10624
    },
    {
      "completion_length": 247.4196548461914,
      "epoch": 1.7815918521312712,
      "grad_norm": 0.23568997635818786,
      "kl": 0.155609130859375,
      "learning_rate": 4.968950981279481e-07,
      "loss": 0.0002,
      "reward": 1.7553572058677673,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7598214708268642,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10626
    },
    {
      "completion_length": 248.5937614440918,
      "epoch": 1.7819271553711387,
      "grad_norm": 0.2304402395430419,
      "kl": 0.101806640625,
      "learning_rate": 4.968932131363621e-07,
      "loss": 0.0001,
      "reward": 1.775000050663948,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7839285954833031,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10628
    },
    {
      "completion_length": 252.67858505249023,
      "epoch": 1.7822624586110063,
      "grad_norm": 0.14773177459543393,
      "kl": 0.088470458984375,
      "learning_rate": 4.968913275763365e-07,
      "loss": 0.0001,
      "reward": 1.753571480512619,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.762500025331974,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10630
    },
    {
      "completion_length": 241.102689743042,
      "epoch": 1.782597761850874,
      "grad_norm": 0.23951759989016289,
      "kl": 0.0987548828125,
      "learning_rate": 4.968894414478756e-07,
      "loss": 0.0001,
      "reward": 1.817857213318348,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8178571686148643,
      "rewards/format_reward_func": 1.0,
      "step": 10632
    },
    {
      "completion_length": 259.4732246398926,
      "epoch": 1.7829330650907416,
      "grad_norm": 0.26787557518060745,
      "kl": 0.1324462890625,
      "learning_rate": 4.968875547509836e-07,
      "loss": 0.0001,
      "reward": 1.7696429267525673,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7741071619093418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10634
    },
    {
      "completion_length": 251.00447750091553,
      "epoch": 1.783268368330609,
      "grad_norm": 0.5559621875801689,
      "kl": 0.1373291015625,
      "learning_rate": 4.968856674856648e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 10636
    },
    {
      "completion_length": 244.40626049041748,
      "epoch": 1.7836036715704764,
      "grad_norm": 0.2907232267990311,
      "kl": 0.132568359375,
      "learning_rate": 4.968837796519238e-07,
      "loss": 0.0001,
      "reward": 1.7214286401867867,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7214285992085934,
      "rewards/format_reward_func": 1.0,
      "step": 10638
    },
    {
      "completion_length": 245.6741180419922,
      "epoch": 1.783938974810344,
      "grad_norm": 0.2367535675360497,
      "kl": 0.134552001953125,
      "learning_rate": 4.968818912497647e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7875000275671482,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10640
    },
    {
      "completion_length": 249.16072463989258,
      "epoch": 1.7842742780502117,
      "grad_norm": 0.279259478311908,
      "kl": 0.3076171875,
      "learning_rate": 4.968800022791921e-07,
      "loss": 0.0003,
      "reward": 1.751785770058632,
      "reward_std": 0.0328299580141902,
      "rewards/equation_reward_func": 0.7651786021888256,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10642
    },
    {
      "completion_length": 244.25447463989258,
      "epoch": 1.7846095812900793,
      "grad_norm": 0.30117573873766484,
      "kl": 0.3115234375,
      "learning_rate": 4.9687811274021e-07,
      "loss": 0.0003,
      "reward": 1.7196429371833801,
      "reward_std": 0.1035406356677413,
      "rewards/equation_reward_func": 0.7330357506871223,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10644
    },
    {
      "completion_length": 247.9375123977661,
      "epoch": 1.784944884529947,
      "grad_norm": 0.31154505385018966,
      "kl": 0.31414794921875,
      "learning_rate": 4.968762226328231e-07,
      "loss": 0.0003,
      "reward": 1.7392857819795609,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7482143118977547,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10646
    },
    {
      "completion_length": 252.08929634094238,
      "epoch": 1.7852801877698143,
      "grad_norm": 0.16945113473111542,
      "kl": 0.0981597900390625,
      "learning_rate": 4.968743319570354e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714399069548,
      "rewards/format_reward_func": 1.0,
      "step": 10648
    },
    {
      "completion_length": 240.6160831451416,
      "epoch": 1.7856154910096818,
      "grad_norm": 0.22287418212263907,
      "kl": 0.100494384765625,
      "learning_rate": 4.968724407128516e-07,
      "loss": 0.0001,
      "reward": 1.8160714879631996,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8196428790688515,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 10650
    },
    {
      "completion_length": 235.75893878936768,
      "epoch": 1.7859507942495494,
      "grad_norm": 0.1826597093835376,
      "kl": 0.0981597900390625,
      "learning_rate": 4.968705489002759e-07,
      "loss": 0.0001,
      "reward": 1.751785770058632,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7562500387430191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10652
    },
    {
      "completion_length": 238.5044755935669,
      "epoch": 1.786286097489417,
      "grad_norm": 0.1955615982514768,
      "kl": 0.101715087890625,
      "learning_rate": 4.968686565193127e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 10654
    },
    {
      "completion_length": 247.62054443359375,
      "epoch": 1.7866214007292847,
      "grad_norm": 0.20413985650724134,
      "kl": 0.120880126953125,
      "learning_rate": 4.968667635699662e-07,
      "loss": 0.0001,
      "reward": 1.8000000640749931,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.808928593993187,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10656
    },
    {
      "completion_length": 247.2232255935669,
      "epoch": 1.786956703969152,
      "grad_norm": 0.2791594220333788,
      "kl": 0.272247314453125,
      "learning_rate": 4.968648700522411e-07,
      "loss": 0.0003,
      "reward": 1.6928571984171867,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.6928571853786707,
      "rewards/format_reward_func": 1.0,
      "step": 10658
    },
    {
      "completion_length": 234.5089406967163,
      "epoch": 1.7872920072090197,
      "grad_norm": 0.3520679227334401,
      "kl": 0.1893463134765625,
      "learning_rate": 4.968629759661414e-07,
      "loss": 0.0002,
      "reward": 1.7607143595814705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 10660
    },
    {
      "completion_length": 242.85715293884277,
      "epoch": 1.7876273104488871,
      "grad_norm": 0.22464576239652737,
      "kl": 0.2414398193359375,
      "learning_rate": 4.968610813116716e-07,
      "loss": 0.0002,
      "reward": 1.7678572162985802,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7767857387661934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10662
    },
    {
      "completion_length": 238.91072463989258,
      "epoch": 1.7879626136887548,
      "grad_norm": 0.2595509564574555,
      "kl": 0.10626220703125,
      "learning_rate": 4.96859186088836e-07,
      "loss": 0.0001,
      "reward": 1.735714353621006,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143238186836,
      "rewards/format_reward_func": 1.0,
      "step": 10664
    },
    {
      "completion_length": 231.3705472946167,
      "epoch": 1.7882979169286224,
      "grad_norm": 0.5999381979692516,
      "kl": 0.176300048828125,
      "learning_rate": 4.968572902976392e-07,
      "loss": 0.0002,
      "reward": 1.8178571835160255,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8178571611642838,
      "rewards/format_reward_func": 1.0,
      "step": 10666
    },
    {
      "completion_length": 243.08482933044434,
      "epoch": 1.78863322016849,
      "grad_norm": 0.45039498742021106,
      "kl": 0.152374267578125,
      "learning_rate": 4.968553939380852e-07,
      "loss": 0.0002,
      "reward": 1.7696429416537285,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7741071581840515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10668
    },
    {
      "completion_length": 238.62055015563965,
      "epoch": 1.7889685234083574,
      "grad_norm": 0.31604306983363795,
      "kl": 0.114898681640625,
      "learning_rate": 4.968534970101786e-07,
      "loss": 0.0001,
      "reward": 1.8392857685685158,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8392857275903225,
      "rewards/format_reward_func": 1.0,
      "step": 10670
    },
    {
      "completion_length": 239.0000123977661,
      "epoch": 1.7893038266482248,
      "grad_norm": 0.1465661738385793,
      "kl": 0.151458740234375,
      "learning_rate": 4.968515995139237e-07,
      "loss": 0.0002,
      "reward": 1.803571492433548,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8035714440047741,
      "rewards/format_reward_func": 1.0,
      "step": 10672
    },
    {
      "completion_length": 244.06697750091553,
      "epoch": 1.7896391298880925,
      "grad_norm": 0.29623229382166644,
      "kl": 0.254180908203125,
      "learning_rate": 4.968497014493251e-07,
      "loss": 0.0003,
      "reward": 1.7482143267989159,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7616071812808514,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10674
    },
    {
      "completion_length": 245.59376335144043,
      "epoch": 1.7899744331279601,
      "grad_norm": 0.26622090051005864,
      "kl": 0.1448974609375,
      "learning_rate": 4.968478028163867e-07,
      "loss": 0.0001,
      "reward": 1.7625000774860382,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7669643089175224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10676
    },
    {
      "completion_length": 246.6607255935669,
      "epoch": 1.7903097363678278,
      "grad_norm": 0.16048621493513235,
      "kl": 0.18072509765625,
      "learning_rate": 4.968459036151132e-07,
      "loss": 0.0002,
      "reward": 1.76071435213089,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143241912127,
      "rewards/format_reward_func": 1.0,
      "step": 10678
    },
    {
      "completion_length": 240.69197368621826,
      "epoch": 1.7906450396076952,
      "grad_norm": 0.1796921365565492,
      "kl": 0.12127685546875,
      "learning_rate": 4.96844003845509e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000469386578,
      "rewards/format_reward_func": 1.0,
      "step": 10680
    },
    {
      "completion_length": 227.19643688201904,
      "epoch": 1.7909803428475628,
      "grad_norm": 0.14189694182681398,
      "kl": 0.176177978515625,
      "learning_rate": 4.968421035075784e-07,
      "loss": 0.0002,
      "reward": 1.7946429252624512,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7991071790456772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10682
    },
    {
      "completion_length": 240.73661518096924,
      "epoch": 1.7913156460874302,
      "grad_norm": 0.2688246315021042,
      "kl": 0.352508544921875,
      "learning_rate": 4.968402026013256e-07,
      "loss": 0.0004,
      "reward": 1.7517858073115349,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500275671482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10684
    },
    {
      "completion_length": 234.1294755935669,
      "epoch": 1.7916509493272978,
      "grad_norm": 0.3868588654613052,
      "kl": 0.184417724609375,
      "learning_rate": 4.968383011267553e-07,
      "loss": 0.0002,
      "reward": 1.7785715088248253,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 10686
    },
    {
      "completion_length": 235.66518878936768,
      "epoch": 1.7919862525671655,
      "grad_norm": 0.23331563541205555,
      "kl": 0.129638671875,
      "learning_rate": 4.968363990838716e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 10688
    },
    {
      "completion_length": 239.1384048461914,
      "epoch": 1.7923215558070331,
      "grad_norm": 0.14848342113524915,
      "kl": 0.100128173828125,
      "learning_rate": 4.96834496472679e-07,
      "loss": 0.0001,
      "reward": 1.8107143640518188,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8107143044471741,
      "rewards/format_reward_func": 1.0,
      "step": 10690
    },
    {
      "completion_length": 235.27679347991943,
      "epoch": 1.7926568590469005,
      "grad_norm": 0.12301443116228022,
      "kl": 0.1092529296875,
      "learning_rate": 4.968325932931819e-07,
      "loss": 0.0001,
      "reward": 1.7821428924798965,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428831666708,
      "rewards/format_reward_func": 1.0,
      "step": 10692
    },
    {
      "completion_length": 245.8437623977661,
      "epoch": 1.792992162286768,
      "grad_norm": 0.24794920535729859,
      "kl": 0.235076904296875,
      "learning_rate": 4.968306895453846e-07,
      "loss": 0.0002,
      "reward": 1.7142857983708382,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7142857424914837,
      "rewards/format_reward_func": 1.0,
      "step": 10694
    },
    {
      "completion_length": 239.08036613464355,
      "epoch": 1.7933274655266356,
      "grad_norm": 0.7665548328413366,
      "kl": 0.161376953125,
      "learning_rate": 4.968287852292916e-07,
      "loss": 0.0002,
      "reward": 1.7857143506407738,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 10696
    },
    {
      "completion_length": 240.67858600616455,
      "epoch": 1.7936627687665032,
      "grad_norm": 0.26873292722443687,
      "kl": 0.1585845947265625,
      "learning_rate": 4.968268803449072e-07,
      "loss": 0.0002,
      "reward": 1.7642857879400253,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857357859612,
      "rewards/format_reward_func": 1.0,
      "step": 10698
    },
    {
      "completion_length": 239.7053689956665,
      "epoch": 1.7939980720063708,
      "grad_norm": 0.36024769526209427,
      "kl": 0.099151611328125,
      "learning_rate": 4.968249748922358e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 10700
    },
    {
      "completion_length": 235.33037090301514,
      "epoch": 1.7943333752462385,
      "grad_norm": 0.1891792212945995,
      "kl": 0.097320556640625,
      "learning_rate": 4.968230688712818e-07,
      "loss": 0.0001,
      "reward": 1.721428669989109,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.721428606659174,
      "rewards/format_reward_func": 1.0,
      "step": 10702
    },
    {
      "completion_length": 223.64733123779297,
      "epoch": 1.794668678486106,
      "grad_norm": 0.18664824171925037,
      "kl": 0.09637451171875,
      "learning_rate": 4.968211622820495e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 10704
    },
    {
      "completion_length": 230.7187614440918,
      "epoch": 1.7950039817259733,
      "grad_norm": 0.19383546116090142,
      "kl": 0.1127166748046875,
      "learning_rate": 4.968192551245435e-07,
      "loss": 0.0001,
      "reward": 1.7892857640981674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857529222965,
      "rewards/format_reward_func": 1.0,
      "step": 10706
    },
    {
      "completion_length": 231.93750953674316,
      "epoch": 1.795339284965841,
      "grad_norm": 0.2077794241490124,
      "kl": 0.093048095703125,
      "learning_rate": 4.968173473987681e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 10708
    },
    {
      "completion_length": 234.94643783569336,
      "epoch": 1.7956745882057086,
      "grad_norm": 0.21449086730659028,
      "kl": 0.096038818359375,
      "learning_rate": 4.968154391047274e-07,
      "loss": 0.0001,
      "reward": 1.7714286595582962,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714285925030708,
      "rewards/format_reward_func": 1.0,
      "step": 10710
    },
    {
      "completion_length": 238.0491180419922,
      "epoch": 1.7960098914455762,
      "grad_norm": 0.29946112981399303,
      "kl": 0.09326171875,
      "learning_rate": 4.968135302424262e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 10712
    },
    {
      "completion_length": 236.37054634094238,
      "epoch": 1.7963451946854436,
      "grad_norm": 0.11589060979527094,
      "kl": 0.091156005859375,
      "learning_rate": 4.968116208118688e-07,
      "loss": 0.0001,
      "reward": 1.8080357685685158,
      "reward_std": 0.059346460737288,
      "rewards/equation_reward_func": 0.8098214641213417,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 10714
    },
    {
      "completion_length": 245.41965198516846,
      "epoch": 1.796680497925311,
      "grad_norm": 0.5277318509362663,
      "kl": 0.11151123046875,
      "learning_rate": 4.968097108130595e-07,
      "loss": 0.0001,
      "reward": 1.7250000983476639,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000424683094,
      "rewards/format_reward_func": 1.0,
      "step": 10716
    },
    {
      "completion_length": 237.00447368621826,
      "epoch": 1.7970158011651787,
      "grad_norm": 0.4631048577308088,
      "kl": 0.155487060546875,
      "learning_rate": 4.968078002460027e-07,
      "loss": 0.0002,
      "reward": 1.8000000566244125,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000193715096,
      "rewards/format_reward_func": 1.0,
      "step": 10718
    },
    {
      "completion_length": 232.86161613464355,
      "epoch": 1.7973511044050463,
      "grad_norm": 0.31052716637182964,
      "kl": 0.121002197265625,
      "learning_rate": 4.968058891107029e-07,
      "loss": 0.0001,
      "reward": 1.7794643267989159,
      "reward_std": 0.049244935624301434,
      "rewards/equation_reward_func": 0.7812500279396772,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 10720
    },
    {
      "completion_length": 240.9241180419922,
      "epoch": 1.797686407644914,
      "grad_norm": 0.49598330005641567,
      "kl": 0.12677001953125,
      "learning_rate": 4.968039774071644e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 10722
    },
    {
      "completion_length": 236.0134048461914,
      "epoch": 1.7980217108847816,
      "grad_norm": 0.40670055538330085,
      "kl": 0.1128997802734375,
      "learning_rate": 4.968020651353916e-07,
      "loss": 0.0001,
      "reward": 1.739285796880722,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7392857372760773,
      "rewards/format_reward_func": 1.0,
      "step": 10724
    },
    {
      "completion_length": 230.78572463989258,
      "epoch": 1.798357014124649,
      "grad_norm": 0.2347570156213548,
      "kl": 0.13226318359375,
      "learning_rate": 4.968001522953889e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000342726707,
      "rewards/format_reward_func": 1.0,
      "step": 10726
    },
    {
      "completion_length": 238.46429634094238,
      "epoch": 1.7986923173645164,
      "grad_norm": 0.15634411925731584,
      "kl": 0.101654052734375,
      "learning_rate": 4.967982388871608e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 10728
    },
    {
      "completion_length": 240.1785831451416,
      "epoch": 1.799027620604384,
      "grad_norm": 0.1919579346758296,
      "kl": 0.098297119140625,
      "learning_rate": 4.967963249107117e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286129921675,
      "rewards/format_reward_func": 1.0,
      "step": 10730
    },
    {
      "completion_length": 236.01340293884277,
      "epoch": 1.7993629238442517,
      "grad_norm": 0.19750935368932737,
      "kl": 0.104766845703125,
      "learning_rate": 4.967944103660458e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 10732
    },
    {
      "completion_length": 242.94197463989258,
      "epoch": 1.7996982270841193,
      "grad_norm": 0.18550836957480152,
      "kl": 0.106964111328125,
      "learning_rate": 4.967924952531678e-07,
      "loss": 0.0001,
      "reward": 1.753571480512619,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714730620384,
      "rewards/format_reward_func": 1.0,
      "step": 10734
    },
    {
      "completion_length": 239.72769165039062,
      "epoch": 1.8000335303239867,
      "grad_norm": 0.14716396585075947,
      "kl": 0.11041259765625,
      "learning_rate": 4.96790579572082e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7535714562982321,
      "rewards/format_reward_func": 1.0,
      "step": 10736
    },
    {
      "completion_length": 234.71429634094238,
      "epoch": 1.8003688335638544,
      "grad_norm": 0.1392211187955793,
      "kl": 0.102142333984375,
      "learning_rate": 4.967886633227927e-07,
      "loss": 0.0001,
      "reward": 1.7250000685453415,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000275671482,
      "rewards/format_reward_func": 1.0,
      "step": 10738
    },
    {
      "completion_length": 236.9375114440918,
      "epoch": 1.8007041368037218,
      "grad_norm": 0.2994866979627325,
      "kl": 0.105682373046875,
      "learning_rate": 4.967867465053044e-07,
      "loss": 0.0001,
      "reward": 1.7839286401867867,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7883928865194321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10740
    },
    {
      "completion_length": 234.30804634094238,
      "epoch": 1.8010394400435894,
      "grad_norm": 0.18439435986178013,
      "kl": 0.102508544921875,
      "learning_rate": 4.967848291196216e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 10742
    },
    {
      "completion_length": 244.1160831451416,
      "epoch": 1.801374743283457,
      "grad_norm": 0.19176630269728612,
      "kl": 0.114044189453125,
      "learning_rate": 4.967829111657485e-07,
      "loss": 0.0001,
      "reward": 1.7071429267525673,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7071428839117289,
      "rewards/format_reward_func": 1.0,
      "step": 10744
    },
    {
      "completion_length": 249.7500123977661,
      "epoch": 1.8017100465233247,
      "grad_norm": 0.3886127234059232,
      "kl": 0.1446533203125,
      "learning_rate": 4.967809926436897e-07,
      "loss": 0.0001,
      "reward": 1.7732143327593803,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7776786126196384,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10746
    },
    {
      "completion_length": 240.1875114440918,
      "epoch": 1.802045349763192,
      "grad_norm": 0.25582454254449444,
      "kl": 0.12982177734375,
      "learning_rate": 4.967790735534495e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 10748
    },
    {
      "completion_length": 243.196439743042,
      "epoch": 1.8023806530030595,
      "grad_norm": 0.1093102806927311,
      "kl": 0.12841796875,
      "learning_rate": 4.967771538950325e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143022119999,
      "rewards/format_reward_func": 1.0,
      "step": 10750
    },
    {
      "completion_length": 233.93304538726807,
      "epoch": 1.8027159562429271,
      "grad_norm": 0.2041405389472402,
      "kl": 0.097808837890625,
      "learning_rate": 4.967752336684428e-07,
      "loss": 0.0001,
      "reward": 1.8178572058677673,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8178571686148643,
      "rewards/format_reward_func": 1.0,
      "step": 10752
    },
    {
      "completion_length": 243.42857933044434,
      "epoch": 1.8030512594827948,
      "grad_norm": 0.31358890129485795,
      "kl": 0.1039886474609375,
      "learning_rate": 4.967733128736852e-07,
      "loss": 0.0001,
      "reward": 1.7696429044008255,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7830357551574707,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10754
    },
    {
      "completion_length": 252.57144165039062,
      "epoch": 1.8033865627226624,
      "grad_norm": 0.5249165360425068,
      "kl": 0.2020263671875,
      "learning_rate": 4.967713915107639e-07,
      "loss": 0.0002,
      "reward": 1.7062500938773155,
      "reward_std": 0.06187184248119593,
      "rewards/equation_reward_func": 0.7125000283122063,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 10756
    },
    {
      "completion_length": 249.23662090301514,
      "epoch": 1.8037218659625298,
      "grad_norm": 0.0033794405458902822,
      "kl": 0.105987548828125,
      "learning_rate": 4.967694695796833e-07,
      "loss": 0.0001,
      "reward": 1.7357143387198448,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7357143145054579,
      "rewards/format_reward_func": 1.0,
      "step": 10758
    },
    {
      "completion_length": 249.14733219146729,
      "epoch": 1.8040571692023974,
      "grad_norm": 0.25247943380170706,
      "kl": 0.135498046875,
      "learning_rate": 4.967675470804479e-07,
      "loss": 0.0001,
      "reward": 1.7482143491506577,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.752678606659174,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10760
    },
    {
      "completion_length": 258.09822940826416,
      "epoch": 1.8043924724422649,
      "grad_norm": 0.32874928742714987,
      "kl": 0.146820068359375,
      "learning_rate": 4.967656240130621e-07,
      "loss": 0.0001,
      "reward": 1.7107143849134445,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7107143104076385,
      "rewards/format_reward_func": 1.0,
      "step": 10762
    },
    {
      "completion_length": 245.71876335144043,
      "epoch": 1.8047277756821325,
      "grad_norm": 0.2009743767668466,
      "kl": 0.3307342529296875,
      "learning_rate": 4.967637003775303e-07,
      "loss": 0.0003,
      "reward": 1.7857143506407738,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143245637417,
      "rewards/format_reward_func": 1.0,
      "step": 10764
    },
    {
      "completion_length": 254.3928689956665,
      "epoch": 1.8050630789220001,
      "grad_norm": 0.24327630118252214,
      "kl": 0.181610107421875,
      "learning_rate": 4.967617761738571e-07,
      "loss": 0.0002,
      "reward": 1.7000000849366188,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7000000346451998,
      "rewards/format_reward_func": 1.0,
      "step": 10766
    },
    {
      "completion_length": 247.15626049041748,
      "epoch": 1.8053983821618678,
      "grad_norm": 0.4902753874075953,
      "kl": 0.193634033203125,
      "learning_rate": 4.967598514020467e-07,
      "loss": 0.0002,
      "reward": 1.7750000581145287,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 10768
    },
    {
      "completion_length": 251.9866180419922,
      "epoch": 1.8057336854017352,
      "grad_norm": 0.22101468467873372,
      "kl": 0.237457275390625,
      "learning_rate": 4.967579260621036e-07,
      "loss": 0.0002,
      "reward": 1.7589286416769028,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928954601288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10770
    },
    {
      "completion_length": 262.8035840988159,
      "epoch": 1.8060689886416026,
      "grad_norm": 0.38687489836976613,
      "kl": 0.288330078125,
      "learning_rate": 4.967560001540324e-07,
      "loss": 0.0003,
      "reward": 1.7178572118282318,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.71785718947649,
      "rewards/format_reward_func": 1.0,
      "step": 10772
    },
    {
      "completion_length": 258.84375858306885,
      "epoch": 1.8064042918814702,
      "grad_norm": 0.1725018185461603,
      "kl": 0.2459564208984375,
      "learning_rate": 4.967540736778373e-07,
      "loss": 0.0002,
      "reward": 1.7500000968575478,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 10774
    },
    {
      "completion_length": 245.78126525878906,
      "epoch": 1.8067395951213379,
      "grad_norm": 0.12436038886983099,
      "kl": 0.120635986328125,
      "learning_rate": 4.967521466335228e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7660714536905289,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10776
    },
    {
      "completion_length": 259.9955472946167,
      "epoch": 1.8070748983612055,
      "grad_norm": 0.3770273477501551,
      "kl": 0.53118896484375,
      "learning_rate": 4.967502190210934e-07,
      "loss": 0.0005,
      "reward": 1.7482143715023994,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.752678606659174,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10778
    },
    {
      "completion_length": 242.696439743042,
      "epoch": 1.8074102016010731,
      "grad_norm": 1.098628616791553,
      "kl": 0.73077392578125,
      "learning_rate": 4.967482908405536e-07,
      "loss": 0.0007,
      "reward": 1.7785715013742447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 10780
    },
    {
      "completion_length": 244.02233123779297,
      "epoch": 1.8077455048409405,
      "grad_norm": 0.3056318524063688,
      "kl": 0.1494140625,
      "learning_rate": 4.967463620919077e-07,
      "loss": 0.0001,
      "reward": 1.7035715132951736,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7035714611411095,
      "rewards/format_reward_func": 1.0,
      "step": 10782
    },
    {
      "completion_length": 251.45536708831787,
      "epoch": 1.808080808080808,
      "grad_norm": 0.2017605345416473,
      "kl": 0.2183837890625,
      "learning_rate": 4.967444327751601e-07,
      "loss": 0.0002,
      "reward": 1.74642863124609,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7553571723401546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10784
    },
    {
      "completion_length": 246.1384038925171,
      "epoch": 1.8084161113206756,
      "grad_norm": 0.17238070059866234,
      "kl": 0.24993896484375,
      "learning_rate": 4.967425028903153e-07,
      "loss": 0.0002,
      "reward": 1.7892857789993286,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857436090708,
      "rewards/format_reward_func": 1.0,
      "step": 10786
    },
    {
      "completion_length": 247.71429538726807,
      "epoch": 1.8087514145605432,
      "grad_norm": 0.21629355159149774,
      "kl": 0.41064453125,
      "learning_rate": 4.96740572437378e-07,
      "loss": 0.0004,
      "reward": 1.7392857894301414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 1.0,
      "step": 10788
    },
    {
      "completion_length": 242.66965579986572,
      "epoch": 1.8090867178004109,
      "grad_norm": 0.3175852033098951,
      "kl": 0.1038665771484375,
      "learning_rate": 4.967386414163522e-07,
      "loss": 0.0001,
      "reward": 1.8196429461240768,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8241071701049805,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10790
    },
    {
      "completion_length": 255.8884038925171,
      "epoch": 1.8094220210402783,
      "grad_norm": 0.32156099273712735,
      "kl": 0.1937713623046875,
      "learning_rate": 4.967367098272427e-07,
      "loss": 0.0002,
      "reward": 1.7410715073347092,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.754464328289032,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10792
    },
    {
      "completion_length": 255.883939743042,
      "epoch": 1.809757324280146,
      "grad_norm": 0.0992087667618495,
      "kl": 0.1259918212890625,
      "learning_rate": 4.967347776700538e-07,
      "loss": 0.0001,
      "reward": 1.6910714656114578,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7133928891271353,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 10794
    },
    {
      "completion_length": 247.7053680419922,
      "epoch": 1.8100926275200133,
      "grad_norm": 0.18729247886434083,
      "kl": 0.1168670654296875,
      "learning_rate": 4.967328449447898e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.787500012665987,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10796
    },
    {
      "completion_length": 260.98215675354004,
      "epoch": 1.810427930759881,
      "grad_norm": 0.21310321098668664,
      "kl": 0.1923980712890625,
      "learning_rate": 4.967309116514555e-07,
      "loss": 0.0002,
      "reward": 1.7964286133646965,
      "reward_std": 0.035355339758098125,
      "rewards/equation_reward_func": 0.8053571805357933,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10798
    },
    {
      "completion_length": 256.29911708831787,
      "epoch": 1.8107632339997486,
      "grad_norm": 0.3038643309134824,
      "kl": 0.1059722900390625,
      "learning_rate": 4.967289777900551e-07,
      "loss": 0.0001,
      "reward": 1.6625000908970833,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.6758928969502449,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10800
    },
    {
      "completion_length": 240.25001049041748,
      "epoch": 1.8110985372396162,
      "grad_norm": 0.283155134206445,
      "kl": 0.132720947265625,
      "learning_rate": 4.96727043360593e-07,
      "loss": 0.0001,
      "reward": 1.7196429520845413,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7241071723401546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10802
    },
    {
      "completion_length": 237.68750953674316,
      "epoch": 1.8114338404794836,
      "grad_norm": 0.08161599208536394,
      "kl": 0.1114654541015625,
      "learning_rate": 4.967251083630739e-07,
      "loss": 0.0001,
      "reward": 1.785714328289032,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143245637417,
      "rewards/format_reward_func": 1.0,
      "step": 10804
    },
    {
      "completion_length": 246.34822273254395,
      "epoch": 1.811769143719351,
      "grad_norm": 0.23240967627608058,
      "kl": 0.122589111328125,
      "learning_rate": 4.967231727975021e-07,
      "loss": 0.0001,
      "reward": 1.762500062584877,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.766964316368103,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10806
    },
    {
      "completion_length": 248.69197463989258,
      "epoch": 1.8121044469592187,
      "grad_norm": 0.1302638213649644,
      "kl": 0.102691650390625,
      "learning_rate": 4.967212366638821e-07,
      "loss": 0.0001,
      "reward": 1.7464286163449287,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7553571723401546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10808
    },
    {
      "completion_length": 249.09822463989258,
      "epoch": 1.8124397501990863,
      "grad_norm": 0.309884045882955,
      "kl": 0.0946807861328125,
      "learning_rate": 4.967192999622183e-07,
      "loss": 0.0001,
      "reward": 1.7375000789761543,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7419643215835094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10810
    },
    {
      "completion_length": 233.42858028411865,
      "epoch": 1.812775053438954,
      "grad_norm": 0.23595094134065409,
      "kl": 0.10211181640625,
      "learning_rate": 4.967173626925152e-07,
      "loss": 0.0001,
      "reward": 1.7589286118745804,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7633928954601288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10812
    },
    {
      "completion_length": 245.852689743042,
      "epoch": 1.8131103566788214,
      "grad_norm": 0.32348892322344996,
      "kl": 0.143402099609375,
      "learning_rate": 4.967154248547773e-07,
      "loss": 0.0001,
      "reward": 1.751785770058632,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500312924385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10814
    },
    {
      "completion_length": 227.3035831451416,
      "epoch": 1.813445659918689,
      "grad_norm": 0.07325437364817967,
      "kl": 0.0951690673828125,
      "learning_rate": 4.967134864490089e-07,
      "loss": 0.0001,
      "reward": 1.782142885029316,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7821428943425417,
      "rewards/format_reward_func": 1.0,
      "step": 10816
    },
    {
      "completion_length": 239.6428689956665,
      "epoch": 1.8137809631585564,
      "grad_norm": 0.25369394576949483,
      "kl": 0.090789794921875,
      "learning_rate": 4.967115474752146e-07,
      "loss": 0.0001,
      "reward": 1.821428619325161,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8214285932481289,
      "rewards/format_reward_func": 1.0,
      "step": 10818
    },
    {
      "completion_length": 241.89733219146729,
      "epoch": 1.814116266398424,
      "grad_norm": 0.18867217775157238,
      "kl": 0.130615234375,
      "learning_rate": 4.967096079333989e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 10820
    },
    {
      "completion_length": 235.8750123977661,
      "epoch": 1.8144515696382917,
      "grad_norm": 0.21150017673269703,
      "kl": 0.095611572265625,
      "learning_rate": 4.967076678235662e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 10822
    },
    {
      "completion_length": 251.977689743042,
      "epoch": 1.8147868728781593,
      "grad_norm": 0.13915586483642634,
      "kl": 0.124359130859375,
      "learning_rate": 4.96705727145721e-07,
      "loss": 0.0001,
      "reward": 1.783928632736206,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7875000275671482,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 10824
    },
    {
      "completion_length": 236.93304634094238,
      "epoch": 1.8151221761180267,
      "grad_norm": 0.32561266503298675,
      "kl": 0.105865478515625,
      "learning_rate": 4.967037858998677e-07,
      "loss": 0.0001,
      "reward": 1.8178571835160255,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.817857164889574,
      "rewards/format_reward_func": 1.0,
      "step": 10826
    },
    {
      "completion_length": 252.33929634094238,
      "epoch": 1.8154574793578941,
      "grad_norm": 0.28224839859070017,
      "kl": 0.1246337890625,
      "learning_rate": 4.967018440860109e-07,
      "loss": 0.0001,
      "reward": 1.8142857775092125,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.8232143074274063,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10828
    },
    {
      "completion_length": 253.1205472946167,
      "epoch": 1.8157927825977618,
      "grad_norm": 0.15807833293193896,
      "kl": 0.157958984375,
      "learning_rate": 4.966999017041549e-07,
      "loss": 0.0002,
      "reward": 1.7892857789993286,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857473343611,
      "rewards/format_reward_func": 1.0,
      "step": 10830
    },
    {
      "completion_length": 251.11608505249023,
      "epoch": 1.8161280858376294,
      "grad_norm": 0.20844670553261815,
      "kl": 0.195587158203125,
      "learning_rate": 4.966979587543043e-07,
      "loss": 0.0002,
      "reward": 1.735714353621006,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143256813288,
      "rewards/format_reward_func": 1.0,
      "step": 10832
    },
    {
      "completion_length": 250.04465198516846,
      "epoch": 1.816463389077497,
      "grad_norm": 0.22690609755391183,
      "kl": 0.124908447265625,
      "learning_rate": 4.966960152364635e-07,
      "loss": 0.0001,
      "reward": 1.7410714700818062,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7455357387661934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10834
    },
    {
      "completion_length": 255.36608409881592,
      "epoch": 1.8167986923173647,
      "grad_norm": 0.1446434148430334,
      "kl": 0.151519775390625,
      "learning_rate": 4.96694071150637e-07,
      "loss": 0.0002,
      "reward": 1.7446429207921028,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7580357454717159,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10836
    },
    {
      "completion_length": 250.4866189956665,
      "epoch": 1.817133995557232,
      "grad_norm": 0.47846980714272636,
      "kl": 0.1998748779296875,
      "learning_rate": 4.966921264968293e-07,
      "loss": 0.0002,
      "reward": 1.6732143759727478,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.6776786055415869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10838
    },
    {
      "completion_length": 243.34375953674316,
      "epoch": 1.8174692987970995,
      "grad_norm": 0.13234759313529285,
      "kl": 0.109649658203125,
      "learning_rate": 4.966901812750448e-07,
      "loss": 0.0001,
      "reward": 1.8428571745753288,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8428571745753288,
      "rewards/format_reward_func": 1.0,
      "step": 10840
    },
    {
      "completion_length": 242.13394165039062,
      "epoch": 1.8178046020369671,
      "grad_norm": 0.13118584337901576,
      "kl": 0.1727294921875,
      "learning_rate": 4.966882354852882e-07,
      "loss": 0.0002,
      "reward": 1.7589286267757416,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7633928954601288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10842
    },
    {
      "completion_length": 254.74554634094238,
      "epoch": 1.8181399052768348,
      "grad_norm": 0.19629673610097528,
      "kl": 0.15948486328125,
      "learning_rate": 4.966862891275637e-07,
      "loss": 0.0002,
      "reward": 1.7750000357627869,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7839286141097546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10844
    },
    {
      "completion_length": 255.2276906967163,
      "epoch": 1.8184752085167024,
      "grad_norm": 0.21825418465981786,
      "kl": 0.15118408203125,
      "learning_rate": 4.966843422018758e-07,
      "loss": 0.0002,
      "reward": 1.771428644657135,
      "reward_std": 0.10101525392383337,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 10846
    },
    {
      "completion_length": 246.93304920196533,
      "epoch": 1.8188105117565698,
      "grad_norm": 0.14504716918157162,
      "kl": 0.1495361328125,
      "learning_rate": 4.966823947082292e-07,
      "loss": 0.0001,
      "reward": 1.7732143625617027,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7776785977184772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10848
    },
    {
      "completion_length": 253.6205472946167,
      "epoch": 1.8191458149964372,
      "grad_norm": 0.20655397846657847,
      "kl": 0.129791259765625,
      "learning_rate": 4.966804466466282e-07,
      "loss": 0.0001,
      "reward": 1.735714353621006,
      "reward_std": 0.10101525392383337,
      "rewards/equation_reward_func": 0.7535714693367481,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 10850
    },
    {
      "completion_length": 251.24554347991943,
      "epoch": 1.8194811182363049,
      "grad_norm": 0.1738275679980491,
      "kl": 0.1940460205078125,
      "learning_rate": 4.966784980170774e-07,
      "loss": 0.0002,
      "reward": 1.7714286223053932,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7892857305705547,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 10852
    },
    {
      "completion_length": 247.00447750091553,
      "epoch": 1.8198164214761725,
      "grad_norm": 0.20985469649492897,
      "kl": 1.001861572265625,
      "learning_rate": 4.966765488195812e-07,
      "loss": 0.001,
      "reward": 1.753571480512619,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.7803571671247482,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 10854
    },
    {
      "completion_length": 236.7812623977661,
      "epoch": 1.8201517247160401,
      "grad_norm": 0.11620140241283157,
      "kl": 0.25567626953125,
      "learning_rate": 4.966745990541442e-07,
      "loss": 0.0003,
      "reward": 1.7553572058677673,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7687500305473804,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10856
    },
    {
      "completion_length": 253.74555015563965,
      "epoch": 1.8204870279559078,
      "grad_norm": 0.1833107759586828,
      "kl": 0.108062744140625,
      "learning_rate": 4.966726487207708e-07,
      "loss": 0.0001,
      "reward": 1.7071429267525673,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7250000350177288,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 10858
    },
    {
      "completion_length": 239.290189743042,
      "epoch": 1.8208223311957752,
      "grad_norm": 0.23300359377419902,
      "kl": 0.562469482421875,
      "learning_rate": 4.966706978194655e-07,
      "loss": 0.0006,
      "reward": 1.7089286297559738,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7133928816765547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10860
    },
    {
      "completion_length": 236.2634038925171,
      "epoch": 1.8211576344356426,
      "grad_norm": 0.18866839307399952,
      "kl": 0.4730224609375,
      "learning_rate": 4.966687463502327e-07,
      "loss": 0.0005,
      "reward": 1.7660714760422707,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7705357298254967,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10862
    },
    {
      "completion_length": 229.13393783569336,
      "epoch": 1.8214929376755102,
      "grad_norm": 0.10210684142020995,
      "kl": 0.330596923828125,
      "learning_rate": 4.966667943130771e-07,
      "loss": 0.0003,
      "reward": 1.7607143595814705,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 10864
    },
    {
      "completion_length": 234.0446538925171,
      "epoch": 1.8218282409153779,
      "grad_norm": 0.23129147813982034,
      "kl": 0.1144256591796875,
      "learning_rate": 4.966648417080031e-07,
      "loss": 0.0001,
      "reward": 1.7500000894069672,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 10866
    },
    {
      "completion_length": 235.9107255935669,
      "epoch": 1.8221635441552455,
      "grad_norm": 0.1554049848557028,
      "kl": 0.1997222900390625,
      "learning_rate": 4.96662888535015e-07,
      "loss": 0.0002,
      "reward": 1.7857143506407738,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 10868
    },
    {
      "completion_length": 235.03572463989258,
      "epoch": 1.822498847395113,
      "grad_norm": 0.40407649067600887,
      "kl": 0.53131103515625,
      "learning_rate": 4.966609347941176e-07,
      "loss": 0.0005,
      "reward": 1.8035714700818062,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 10870
    },
    {
      "completion_length": 233.65179824829102,
      "epoch": 1.8228341506349806,
      "grad_norm": 0.13115162629134425,
      "kl": 0.3810577392578125,
      "learning_rate": 4.966589804853153e-07,
      "loss": 0.0004,
      "reward": 1.7642857804894447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857581377029,
      "rewards/format_reward_func": 1.0,
      "step": 10872
    },
    {
      "completion_length": 224.61161708831787,
      "epoch": 1.823169453874848,
      "grad_norm": 0.22174956953423863,
      "kl": 0.0928497314453125,
      "learning_rate": 4.966570256086126e-07,
      "loss": 0.0001,
      "reward": 1.7928572222590446,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 10874
    },
    {
      "completion_length": 231.61608123779297,
      "epoch": 1.8235047571147156,
      "grad_norm": 0.2175888324277477,
      "kl": 0.1910400390625,
      "learning_rate": 4.966550701640139e-07,
      "loss": 0.0002,
      "reward": 1.7464286610484123,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 10876
    },
    {
      "completion_length": 224.2098331451416,
      "epoch": 1.8238400603545832,
      "grad_norm": 0.13761918610296675,
      "kl": 0.294647216796875,
      "learning_rate": 4.966531141515237e-07,
      "loss": 0.0003,
      "reward": 1.7464286163449287,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7464286144822836,
      "rewards/format_reward_func": 1.0,
      "step": 10878
    },
    {
      "completion_length": 219.59375858306885,
      "epoch": 1.8241753635944509,
      "grad_norm": 0.12025355892629688,
      "kl": 0.126495361328125,
      "learning_rate": 4.966511575711467e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143223285675,
      "rewards/format_reward_func": 1.0,
      "step": 10880
    },
    {
      "completion_length": 228.77233219146729,
      "epoch": 1.8245106668343183,
      "grad_norm": 0.3027116876913923,
      "kl": 0.84716796875,
      "learning_rate": 4.966492004228872e-07,
      "loss": 0.0008,
      "reward": 1.7678572088479996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 10882
    },
    {
      "completion_length": 220.33929538726807,
      "epoch": 1.8248459700741857,
      "grad_norm": 0.15113626205446923,
      "kl": 0.207672119140625,
      "learning_rate": 4.966472427067499e-07,
      "loss": 0.0002,
      "reward": 1.8571428954601288,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8571428768336773,
      "rewards/format_reward_func": 1.0,
      "step": 10884
    },
    {
      "completion_length": 223.70536613464355,
      "epoch": 1.8251812733140533,
      "grad_norm": 0.30749767893684915,
      "kl": 0.09979248046875,
      "learning_rate": 4.966452844227391e-07,
      "loss": 0.0001,
      "reward": 1.796428643167019,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7964286059141159,
      "rewards/format_reward_func": 1.0,
      "step": 10886
    },
    {
      "completion_length": 225.04018878936768,
      "epoch": 1.825516576553921,
      "grad_norm": 0.2152092830261441,
      "kl": 0.13427734375,
      "learning_rate": 4.966433255708594e-07,
      "loss": 0.0001,
      "reward": 1.728571504354477,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7285714522004128,
      "rewards/format_reward_func": 1.0,
      "step": 10888
    },
    {
      "completion_length": 237.9241189956665,
      "epoch": 1.8258518797937886,
      "grad_norm": 0.11807239100575599,
      "kl": 0.154876708984375,
      "learning_rate": 4.966413661511154e-07,
      "loss": 0.0002,
      "reward": 1.7250000461935997,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.725000036880374,
      "rewards/format_reward_func": 1.0,
      "step": 10890
    },
    {
      "completion_length": 228.87501049041748,
      "epoch": 1.826187183033656,
      "grad_norm": 0.273143116307382,
      "kl": 0.3214111328125,
      "learning_rate": 4.966394061635115e-07,
      "loss": 0.0003,
      "reward": 1.7428572103381157,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 10892
    },
    {
      "completion_length": 223.47322368621826,
      "epoch": 1.8265224862735236,
      "grad_norm": 0.1521072199861038,
      "kl": 0.11309814453125,
      "learning_rate": 4.966374456080522e-07,
      "loss": 0.0001,
      "reward": 1.735714353621006,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.735714316368103,
      "rewards/format_reward_func": 1.0,
      "step": 10894
    },
    {
      "completion_length": 237.30358123779297,
      "epoch": 1.826857789513391,
      "grad_norm": 0.6139851648931479,
      "kl": 0.197723388671875,
      "learning_rate": 4.966354844847421e-07,
      "loss": 0.0002,
      "reward": 1.778571479022503,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 10896
    },
    {
      "completion_length": 238.2946548461914,
      "epoch": 1.8271930927532587,
      "grad_norm": 0.29583958632372603,
      "kl": 0.115325927734375,
      "learning_rate": 4.966335227935856e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571604192257,
      "rewards/format_reward_func": 1.0,
      "step": 10898
    },
    {
      "completion_length": 225.09822368621826,
      "epoch": 1.8275283959931263,
      "grad_norm": 0.24015298347241795,
      "kl": 0.225799560546875,
      "learning_rate": 4.966315605345873e-07,
      "loss": 0.0002,
      "reward": 1.850000038743019,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8500000089406967,
      "rewards/format_reward_func": 1.0,
      "step": 10900
    },
    {
      "completion_length": 235.571439743042,
      "epoch": 1.827863699232994,
      "grad_norm": 0.2932754055945961,
      "kl": 0.11669921875,
      "learning_rate": 4.966295977077518e-07,
      "loss": 0.0001,
      "reward": 1.7410715073347092,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.745535746216774,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10902
    },
    {
      "completion_length": 238.17411613464355,
      "epoch": 1.8281990024728614,
      "grad_norm": 0.26385235317977035,
      "kl": 0.1941680908203125,
      "learning_rate": 4.966276343130835e-07,
      "loss": 0.0002,
      "reward": 1.7142858132719994,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857536673546,
      "rewards/format_reward_func": 1.0,
      "step": 10904
    },
    {
      "completion_length": 234.19643783569336,
      "epoch": 1.8285343057127288,
      "grad_norm": 0.3444873572004797,
      "kl": 0.1352691650390625,
      "learning_rate": 4.966256703505869e-07,
      "loss": 0.0001,
      "reward": 1.8178571835160255,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.817857176065445,
      "rewards/format_reward_func": 1.0,
      "step": 10906
    },
    {
      "completion_length": 247.98215198516846,
      "epoch": 1.8288696089525964,
      "grad_norm": 0.2911970609785864,
      "kl": 0.11505126953125,
      "learning_rate": 4.966237058202665e-07,
      "loss": 0.0001,
      "reward": 1.7178572192788124,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7267857529222965,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10908
    },
    {
      "completion_length": 239.40179538726807,
      "epoch": 1.829204912192464,
      "grad_norm": 0.25664083609430993,
      "kl": 0.21173095703125,
      "learning_rate": 4.96621740722127e-07,
      "loss": 0.0002,
      "reward": 1.7535714730620384,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714562982321,
      "rewards/format_reward_func": 1.0,
      "step": 10910
    },
    {
      "completion_length": 233.74554634094238,
      "epoch": 1.8295402154323317,
      "grad_norm": 0.2616752508689885,
      "kl": 0.0946807861328125,
      "learning_rate": 4.966197750561728e-07,
      "loss": 0.0001,
      "reward": 1.7839286178350449,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10912
    },
    {
      "completion_length": 234.1294755935669,
      "epoch": 1.8298755186721993,
      "grad_norm": 0.2995033876494516,
      "kl": 0.1689453125,
      "learning_rate": 4.966178088224084e-07,
      "loss": 0.0002,
      "reward": 1.7892857789993286,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 10914
    },
    {
      "completion_length": 242.03126525878906,
      "epoch": 1.8302108219120667,
      "grad_norm": 0.31543069402561186,
      "kl": 0.10711669921875,
      "learning_rate": 4.966158420208383e-07,
      "loss": 0.0001,
      "reward": 1.7232143506407738,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7276785969734192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10916
    },
    {
      "completion_length": 219.94197273254395,
      "epoch": 1.8305461251519342,
      "grad_norm": 0.21660990217758797,
      "kl": 0.1119232177734375,
      "learning_rate": 4.966138746514672e-07,
      "loss": 0.0001,
      "reward": 1.8107143342494965,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 10918
    },
    {
      "completion_length": 227.89733123779297,
      "epoch": 1.8308814283918018,
      "grad_norm": 0.16069491284800438,
      "kl": 0.1295623779296875,
      "learning_rate": 4.966119067142995e-07,
      "loss": 0.0001,
      "reward": 1.7714286148548126,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7803571783006191,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10920
    },
    {
      "completion_length": 238.59376049041748,
      "epoch": 1.8312167316316694,
      "grad_norm": 0.18957200729810142,
      "kl": 0.117218017578125,
      "learning_rate": 4.966099382093397e-07,
      "loss": 0.0001,
      "reward": 1.7482143640518188,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526785992085934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10922
    },
    {
      "completion_length": 232.71875858306885,
      "epoch": 1.831552034871537,
      "grad_norm": 0.12322576113924998,
      "kl": 0.1805267333984375,
      "learning_rate": 4.966079691365925e-07,
      "loss": 0.0002,
      "reward": 1.7803571969270706,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10924
    },
    {
      "completion_length": 233.70983028411865,
      "epoch": 1.8318873381114045,
      "grad_norm": 0.2518008327942554,
      "kl": 0.1314697265625,
      "learning_rate": 4.966059994960622e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 10926
    },
    {
      "completion_length": 247.36608123779297,
      "epoch": 1.832222641351272,
      "grad_norm": 0.3195704223914685,
      "kl": 0.25115966796875,
      "learning_rate": 4.966040292877534e-07,
      "loss": 0.0003,
      "reward": 1.7607143446803093,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7696428894996643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10928
    },
    {
      "completion_length": 232.74554538726807,
      "epoch": 1.8325579445911395,
      "grad_norm": 0.10482360688190698,
      "kl": 0.15802001953125,
      "learning_rate": 4.966020585116709e-07,
      "loss": 0.0002,
      "reward": 1.767857201397419,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571734577417,
      "rewards/format_reward_func": 1.0,
      "step": 10930
    },
    {
      "completion_length": 230.0937623977661,
      "epoch": 1.8328932478310072,
      "grad_norm": 0.3141847682191237,
      "kl": 0.16552734375,
      "learning_rate": 4.966000871678189e-07,
      "loss": 0.0002,
      "reward": 1.757142923772335,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 10932
    },
    {
      "completion_length": 232.38393878936768,
      "epoch": 1.8332285510708748,
      "grad_norm": 0.16685600890208938,
      "kl": 0.35284423828125,
      "learning_rate": 4.965981152562021e-07,
      "loss": 0.0004,
      "reward": 1.7446429058909416,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7491071783006191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10934
    },
    {
      "completion_length": 240.52679634094238,
      "epoch": 1.8335638543107424,
      "grad_norm": 0.5557902127047117,
      "kl": 0.145477294921875,
      "learning_rate": 4.96596142776825e-07,
      "loss": 0.0001,
      "reward": 1.758928619325161,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.763392873108387,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10936
    },
    {
      "completion_length": 227.29911708831787,
      "epoch": 1.8338991575506098,
      "grad_norm": 0.25553049632062863,
      "kl": 0.1214599609375,
      "learning_rate": 4.965941697296922e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.757142897695303,
      "rewards/format_reward_func": 1.0,
      "step": 10938
    },
    {
      "completion_length": 244.99554538726807,
      "epoch": 1.8342344607904773,
      "grad_norm": 0.36957619283406395,
      "kl": 0.170867919921875,
      "learning_rate": 4.965921961148081e-07,
      "loss": 0.0002,
      "reward": 1.7339286506175995,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7464285884052515,
      "rewards/format_reward_func": 0.9875000044703484,
      "step": 10940
    },
    {
      "completion_length": 235.54465675354004,
      "epoch": 1.8345697640303449,
      "grad_norm": 0.5029314215918635,
      "kl": 0.261199951171875,
      "learning_rate": 4.965902219321773e-07,
      "loss": 0.0003,
      "reward": 1.8000000715255737,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000193715096,
      "rewards/format_reward_func": 1.0,
      "step": 10942
    },
    {
      "completion_length": 234.7009038925171,
      "epoch": 1.8349050672702125,
      "grad_norm": 0.19529034614225257,
      "kl": 0.1197967529296875,
      "learning_rate": 4.965882471818045e-07,
      "loss": 0.0001,
      "reward": 1.7428571954369545,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7517857477068901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10944
    },
    {
      "completion_length": 234.1696538925171,
      "epoch": 1.8352403705100802,
      "grad_norm": 0.21069830505407974,
      "kl": 0.131500244140625,
      "learning_rate": 4.965862718636941e-07,
      "loss": 0.0001,
      "reward": 1.7125000581145287,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7258928883820772,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 10946
    },
    {
      "completion_length": 230.1964406967163,
      "epoch": 1.8355756737499476,
      "grad_norm": 0.33352966239595755,
      "kl": 0.177215576171875,
      "learning_rate": 4.965842959778505e-07,
      "loss": 0.0002,
      "reward": 1.7482143342494965,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7526786029338837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10948
    },
    {
      "completion_length": 233.35715198516846,
      "epoch": 1.8359109769898152,
      "grad_norm": 0.26718184673680234,
      "kl": 0.15020751953125,
      "learning_rate": 4.965823195242786e-07,
      "loss": 0.0002,
      "reward": 1.7732143625617027,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.777678593993187,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10950
    },
    {
      "completion_length": 228.6741189956665,
      "epoch": 1.8362462802296826,
      "grad_norm": 0.6468786012472365,
      "kl": 0.130218505859375,
      "learning_rate": 4.965803425029828e-07,
      "loss": 0.0001,
      "reward": 1.751785784959793,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7562500331550837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10952
    },
    {
      "completion_length": 217.52679538726807,
      "epoch": 1.8365815834695502,
      "grad_norm": 0.20986412344697003,
      "kl": 0.132232666015625,
      "learning_rate": 4.965783649139675e-07,
      "loss": 0.0001,
      "reward": 1.787500038743019,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7919643186032772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10954
    },
    {
      "completion_length": 227.6696538925171,
      "epoch": 1.8369168867094179,
      "grad_norm": 0.13116619544808694,
      "kl": 0.1138916015625,
      "learning_rate": 4.965763867572375e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 10956
    },
    {
      "completion_length": 218.23661708831787,
      "epoch": 1.8372521899492855,
      "grad_norm": 0.20983926729483543,
      "kl": 0.1107177734375,
      "learning_rate": 4.965744080327972e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571626543999,
      "rewards/format_reward_func": 1.0,
      "step": 10958
    },
    {
      "completion_length": 223.20090293884277,
      "epoch": 1.837587493189153,
      "grad_norm": 0.3545627872806163,
      "kl": 0.10614013671875,
      "learning_rate": 4.965724287406512e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000394880772,
      "rewards/format_reward_func": 1.0,
      "step": 10960
    },
    {
      "completion_length": 224.86608123779297,
      "epoch": 1.8379227964290203,
      "grad_norm": 0.2105883357527289,
      "kl": 0.1058807373046875,
      "learning_rate": 4.96570448880804e-07,
      "loss": 0.0001,
      "reward": 1.7500000894069672,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7589286044239998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10962
    },
    {
      "completion_length": 229.571439743042,
      "epoch": 1.838258099668888,
      "grad_norm": 0.22364455871043729,
      "kl": 0.12042236328125,
      "learning_rate": 4.965684684532603e-07,
      "loss": 0.0001,
      "reward": 1.7468750923871994,
      "reward_std": 0.037249374436214566,
      "rewards/equation_reward_func": 0.7517857421189547,
      "rewards/format_reward_func": 0.9950892888009548,
      "step": 10964
    },
    {
      "completion_length": 224.60715293884277,
      "epoch": 1.8385934029087556,
      "grad_norm": 0.26191034704334365,
      "kl": 0.14447021484375,
      "learning_rate": 4.965664874580244e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.755357164889574,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10966
    },
    {
      "completion_length": 225.0803680419922,
      "epoch": 1.8389287061486232,
      "grad_norm": 0.3470244049399262,
      "kl": 0.118988037109375,
      "learning_rate": 4.965645058951011e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 10968
    },
    {
      "completion_length": 223.38840293884277,
      "epoch": 1.8392640093884907,
      "grad_norm": 0.3880943554691539,
      "kl": 0.1046142578125,
      "learning_rate": 4.965625237644949e-07,
      "loss": 0.0001,
      "reward": 1.7428572326898575,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 10970
    },
    {
      "completion_length": 221.9375114440918,
      "epoch": 1.8395993126283583,
      "grad_norm": 0.5691958115794947,
      "kl": 0.1212158203125,
      "learning_rate": 4.965605410662104e-07,
      "loss": 0.0001,
      "reward": 1.766071505844593,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7705357372760773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10972
    },
    {
      "completion_length": 222.00893878936768,
      "epoch": 1.8399346158682257,
      "grad_norm": 0.17615148433837938,
      "kl": 0.101470947265625,
      "learning_rate": 4.965585578002521e-07,
      "loss": 0.0001,
      "reward": 1.8000000342726707,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 10974
    },
    {
      "completion_length": 242.60268878936768,
      "epoch": 1.8402699191080933,
      "grad_norm": 0.18370374417663582,
      "kl": 0.130126953125,
      "learning_rate": 4.965565739666245e-07,
      "loss": 0.0001,
      "reward": 1.755357213318348,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7598214484751225,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10976
    },
    {
      "completion_length": 233.42411613464355,
      "epoch": 1.840605222347961,
      "grad_norm": 0.3542768180228526,
      "kl": 0.112030029296875,
      "learning_rate": 4.965545895653324e-07,
      "loss": 0.0001,
      "reward": 1.7607143744826317,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7696428783237934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10978
    },
    {
      "completion_length": 234.0491180419922,
      "epoch": 1.8409405255878286,
      "grad_norm": 0.21181866329362126,
      "kl": 0.11077880859375,
      "learning_rate": 4.965526045963801e-07,
      "loss": 0.0001,
      "reward": 1.7535715326666832,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7625000197440386,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 10980
    },
    {
      "completion_length": 235.08483505249023,
      "epoch": 1.841275828827696,
      "grad_norm": 0.14119896069516588,
      "kl": 0.11920166015625,
      "learning_rate": 4.965506190597723e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 10982
    },
    {
      "completion_length": 228.2991180419922,
      "epoch": 1.8416111320675634,
      "grad_norm": 0.394193532340823,
      "kl": 0.11993408203125,
      "learning_rate": 4.965486329555136e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000447034836,
      "rewards/format_reward_func": 1.0,
      "step": 10984
    },
    {
      "completion_length": 229.94643878936768,
      "epoch": 1.841946435307431,
      "grad_norm": 0.22609019385205742,
      "kl": 0.100433349609375,
      "learning_rate": 4.965466462836085e-07,
      "loss": 0.0001,
      "reward": 1.8000000715255737,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000156462193,
      "rewards/format_reward_func": 1.0,
      "step": 10986
    },
    {
      "completion_length": 237.7276906967163,
      "epoch": 1.8422817385472987,
      "grad_norm": 0.3759695986030544,
      "kl": 0.1258544921875,
      "learning_rate": 4.965446590440616e-07,
      "loss": 0.0001,
      "reward": 1.8035715073347092,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.8035714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 10988
    },
    {
      "completion_length": 217.1428680419922,
      "epoch": 1.8426170417871663,
      "grad_norm": 0.2734768653982204,
      "kl": 0.109344482421875,
      "learning_rate": 4.965426712368776e-07,
      "loss": 0.0001,
      "reward": 1.7500000521540642,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7500000447034836,
      "rewards/format_reward_func": 1.0,
      "step": 10990
    },
    {
      "completion_length": 228.47322368621826,
      "epoch": 1.842952345027034,
      "grad_norm": 0.30183089643210215,
      "kl": 0.117431640625,
      "learning_rate": 4.965406828620607e-07,
      "loss": 0.0001,
      "reward": 1.730357214808464,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7348214518278837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 10992
    },
    {
      "completion_length": 227.852689743042,
      "epoch": 1.8432876482669014,
      "grad_norm": 0.16928201748186822,
      "kl": 0.099700927734375,
      "learning_rate": 4.96538693919616e-07,
      "loss": 0.0001,
      "reward": 1.7714286148548126,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286111295223,
      "rewards/format_reward_func": 1.0,
      "step": 10994
    },
    {
      "completion_length": 224.49108123779297,
      "epoch": 1.8436229515067688,
      "grad_norm": 0.2128443559967921,
      "kl": 0.099517822265625,
      "learning_rate": 4.965367044095477e-07,
      "loss": 0.0001,
      "reward": 1.7678572237491608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 10996
    },
    {
      "completion_length": 232.45090293884277,
      "epoch": 1.8439582547466364,
      "grad_norm": 0.35638057676443574,
      "kl": 0.159698486328125,
      "learning_rate": 4.965347143318605e-07,
      "loss": 0.0002,
      "reward": 1.7821429297327995,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 10998
    },
    {
      "completion_length": 220.62054634094238,
      "epoch": 1.844293557986504,
      "grad_norm": 0.2383076241088169,
      "kl": 0.09564208984375,
      "learning_rate": 4.96532723686559e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464285958558321,
      "rewards/format_reward_func": 1.0,
      "step": 11000
    },
    {
      "completion_length": 238.58036994934082,
      "epoch": 1.8446288612263717,
      "grad_norm": 0.3791546490491425,
      "kl": 0.20074462890625,
      "learning_rate": 4.965307324736477e-07,
      "loss": 0.0002,
      "reward": 1.8285714909434319,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.8285714481025934,
      "rewards/format_reward_func": 1.0,
      "step": 11002
    },
    {
      "completion_length": 223.78126049041748,
      "epoch": 1.8449641644662391,
      "grad_norm": 0.16352363771720657,
      "kl": 0.114044189453125,
      "learning_rate": 4.965287406931313e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7964286059141159,
      "rewards/format_reward_func": 1.0,
      "step": 11004
    },
    {
      "completion_length": 227.56251049041748,
      "epoch": 1.8452994677061068,
      "grad_norm": 0.3131439397412478,
      "kl": 0.1202392578125,
      "learning_rate": 4.965267483450144e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.08081220090389252,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 11006
    },
    {
      "completion_length": 234.16518878936768,
      "epoch": 1.8456347709459742,
      "grad_norm": 0.20428976348291153,
      "kl": 0.10565185546875,
      "learning_rate": 4.965247554293014e-07,
      "loss": 0.0001,
      "reward": 1.8607143238186836,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8607143051922321,
      "rewards/format_reward_func": 1.0,
      "step": 11008
    },
    {
      "completion_length": 222.62054538726807,
      "epoch": 1.8459700741858418,
      "grad_norm": 0.27048745939962315,
      "kl": 0.1072845458984375,
      "learning_rate": 4.965227619459971e-07,
      "loss": 0.0001,
      "reward": 1.7821429371833801,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 11010
    },
    {
      "completion_length": 221.56251049041748,
      "epoch": 1.8463053774257094,
      "grad_norm": 0.255154030721768,
      "kl": 0.090850830078125,
      "learning_rate": 4.96520767895106e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7964285835623741,
      "rewards/format_reward_func": 1.0,
      "step": 11012
    },
    {
      "completion_length": 215.44197368621826,
      "epoch": 1.846640680665577,
      "grad_norm": 0.2318606194787304,
      "kl": 0.101898193359375,
      "learning_rate": 4.965187732766326e-07,
      "loss": 0.0001,
      "reward": 1.8035714849829674,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8035714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 11014
    },
    {
      "completion_length": 221.33929538726807,
      "epoch": 1.8469759839054445,
      "grad_norm": 0.14313873384667694,
      "kl": 0.1305389404296875,
      "learning_rate": 4.965167780905817e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714656114578,
      "rewards/format_reward_func": 1.0,
      "step": 11016
    },
    {
      "completion_length": 231.19197845458984,
      "epoch": 1.847311287145312,
      "grad_norm": 0.22620464844274196,
      "kl": 0.10772705078125,
      "learning_rate": 4.965147823369576e-07,
      "loss": 0.0001,
      "reward": 1.7321429401636124,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428842842579,
      "rewards/format_reward_func": 1.0,
      "step": 11018
    },
    {
      "completion_length": 225.977689743042,
      "epoch": 1.8476465903851795,
      "grad_norm": 0.19280302891875264,
      "kl": 0.0964508056640625,
      "learning_rate": 4.965127860157652e-07,
      "loss": 0.0001,
      "reward": 1.76517865806818,
      "reward_std": 0.059346460737288,
      "rewards/equation_reward_func": 0.766964316368103,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 11020
    },
    {
      "completion_length": 232.98215198516846,
      "epoch": 1.8479818936250472,
      "grad_norm": 0.2722219674141182,
      "kl": 0.118072509765625,
      "learning_rate": 4.96510789127009e-07,
      "loss": 0.0001,
      "reward": 1.6964286640286446,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.6964286100119352,
      "rewards/format_reward_func": 1.0,
      "step": 11022
    },
    {
      "completion_length": 225.9107255935669,
      "epoch": 1.8483171968649148,
      "grad_norm": 0.25389911395651005,
      "kl": 0.100128173828125,
      "learning_rate": 4.965087916706934e-07,
      "loss": 0.0001,
      "reward": 1.7464286610484123,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 11024
    },
    {
      "completion_length": 221.84822368621826,
      "epoch": 1.8486525001047822,
      "grad_norm": 0.21637961265921582,
      "kl": 0.102935791015625,
      "learning_rate": 4.965067936468234e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 11026
    },
    {
      "completion_length": 230.5803680419922,
      "epoch": 1.8489878033446498,
      "grad_norm": 0.18228930297643892,
      "kl": 0.1324462890625,
      "learning_rate": 4.965047950554032e-07,
      "loss": 0.0001,
      "reward": 1.725000075995922,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7250000312924385,
      "rewards/format_reward_func": 1.0,
      "step": 11028
    },
    {
      "completion_length": 223.92858123779297,
      "epoch": 1.8493231065845173,
      "grad_norm": 0.23163988946281985,
      "kl": 0.0888214111328125,
      "learning_rate": 4.965027958964376e-07,
      "loss": 0.0001,
      "reward": 1.703571505844593,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7035714611411095,
      "rewards/format_reward_func": 1.0,
      "step": 11030
    },
    {
      "completion_length": 234.64733219146729,
      "epoch": 1.849658409824385,
      "grad_norm": 0.15882551249432497,
      "kl": 0.11004638671875,
      "learning_rate": 4.965007961699312e-07,
      "loss": 0.0001,
      "reward": 1.787500038743019,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7919643111526966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11032
    },
    {
      "completion_length": 234.52679634094238,
      "epoch": 1.8499937130642525,
      "grad_norm": 0.16316429098513455,
      "kl": 0.10003662109375,
      "learning_rate": 4.964987958758885e-07,
      "loss": 0.0001,
      "reward": 1.760714367032051,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 11034
    },
    {
      "completion_length": 235.0267972946167,
      "epoch": 1.8503290163041202,
      "grad_norm": 0.1258603120735231,
      "kl": 0.24822998046875,
      "learning_rate": 4.964967950143143e-07,
      "loss": 0.0002,
      "reward": 1.733928620815277,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7383928894996643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11036
    },
    {
      "completion_length": 227.22768688201904,
      "epoch": 1.8506643195439876,
      "grad_norm": 0.28442494129277424,
      "kl": 0.114990234375,
      "learning_rate": 4.964947935852129e-07,
      "loss": 0.0001,
      "reward": 1.7500000596046448,
      "reward_std": 0.07071067858487368,
      "rewards/equation_reward_func": 0.7589286081492901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11038
    },
    {
      "completion_length": 229.85268878936768,
      "epoch": 1.850999622783855,
      "grad_norm": 0.3542580625263232,
      "kl": 0.116607666015625,
      "learning_rate": 4.964927915885893e-07,
      "loss": 0.0001,
      "reward": 1.703571505844593,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7035714723169804,
      "rewards/format_reward_func": 1.0,
      "step": 11040
    },
    {
      "completion_length": 231.94643878936768,
      "epoch": 1.8513349260237226,
      "grad_norm": 0.2923367114733701,
      "kl": 0.1073455810546875,
      "learning_rate": 4.964907890244478e-07,
      "loss": 0.0001,
      "reward": 1.7946429252624512,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7991071753203869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11042
    },
    {
      "completion_length": 225.80804252624512,
      "epoch": 1.8516702292635903,
      "grad_norm": 0.2327532429020024,
      "kl": 0.1014404296875,
      "learning_rate": 4.964887858927931e-07,
      "loss": 0.0001,
      "reward": 1.775000050663948,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7839286103844643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11044
    },
    {
      "completion_length": 228.22768783569336,
      "epoch": 1.852005532503458,
      "grad_norm": 0.277353104211083,
      "kl": 0.1070098876953125,
      "learning_rate": 4.964867821936298e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 11046
    },
    {
      "completion_length": 226.15179634094238,
      "epoch": 1.8523408357433255,
      "grad_norm": 0.23078546163027888,
      "kl": 0.116180419921875,
      "learning_rate": 4.964847779269625e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 11048
    },
    {
      "completion_length": 227.08036708831787,
      "epoch": 1.852676138983193,
      "grad_norm": 0.2231158849058702,
      "kl": 0.099700927734375,
      "learning_rate": 4.96482773092796e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857581377029,
      "rewards/format_reward_func": 1.0,
      "step": 11050
    },
    {
      "completion_length": 227.75893783569336,
      "epoch": 1.8530114422230604,
      "grad_norm": 0.140503338610388,
      "kl": 0.134063720703125,
      "learning_rate": 4.964807676911347e-07,
      "loss": 0.0001,
      "reward": 1.7000000849366188,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7000000327825546,
      "rewards/format_reward_func": 1.0,
      "step": 11052
    },
    {
      "completion_length": 226.5759038925171,
      "epoch": 1.853346745462928,
      "grad_norm": 0.2436791254328514,
      "kl": 0.125732421875,
      "learning_rate": 4.964787617219832e-07,
      "loss": 0.0001,
      "reward": 1.8214286044239998,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8214285932481289,
      "rewards/format_reward_func": 1.0,
      "step": 11054
    },
    {
      "completion_length": 224.81697368621826,
      "epoch": 1.8536820487027956,
      "grad_norm": 0.2266395282483944,
      "kl": 0.10595703125,
      "learning_rate": 4.964767551853463e-07,
      "loss": 0.0001,
      "reward": 1.751785784959793,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500238418579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11056
    },
    {
      "completion_length": 218.30804634094238,
      "epoch": 1.8540173519426633,
      "grad_norm": 0.12301408490279202,
      "kl": 0.106903076171875,
      "learning_rate": 4.964747480812285e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 11058
    },
    {
      "completion_length": 225.54911613464355,
      "epoch": 1.8543526551825307,
      "grad_norm": 0.2668418387078583,
      "kl": 0.118621826171875,
      "learning_rate": 4.964727404096344e-07,
      "loss": 0.0001,
      "reward": 1.6875000819563866,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.6919643320143223,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11060
    },
    {
      "completion_length": 220.75893592834473,
      "epoch": 1.8546879584223983,
      "grad_norm": 0.0035520221745497823,
      "kl": 0.1208953857421875,
      "learning_rate": 4.964707321705687e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 11062
    },
    {
      "completion_length": 237.15179443359375,
      "epoch": 1.8550232616622657,
      "grad_norm": 0.36594927961505463,
      "kl": 0.11773681640625,
      "learning_rate": 4.96468723364036e-07,
      "loss": 0.0001,
      "reward": 1.796428643167019,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 11064
    },
    {
      "completion_length": 222.36608028411865,
      "epoch": 1.8553585649021334,
      "grad_norm": 0.30599537635174073,
      "kl": 0.106658935546875,
      "learning_rate": 4.964667139900409e-07,
      "loss": 0.0001,
      "reward": 1.785714365541935,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 11066
    },
    {
      "completion_length": 223.11161708831787,
      "epoch": 1.855693868142001,
      "grad_norm": 0.47049750051047357,
      "kl": 0.12823486328125,
      "learning_rate": 4.96464704048588e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 11068
    },
    {
      "completion_length": 215.1741180419922,
      "epoch": 1.8560291713818686,
      "grad_norm": 0.2115054002297252,
      "kl": 0.0959320068359375,
      "learning_rate": 4.964626935396821e-07,
      "loss": 0.0001,
      "reward": 1.7892857491970062,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857473343611,
      "rewards/format_reward_func": 1.0,
      "step": 11070
    },
    {
      "completion_length": 217.03572273254395,
      "epoch": 1.856364474621736,
      "grad_norm": 0.18145256750998845,
      "kl": 0.108856201171875,
      "learning_rate": 4.964606824633276e-07,
      "loss": 0.0001,
      "reward": 1.760714367032051,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 11072
    },
    {
      "completion_length": 221.37500953674316,
      "epoch": 1.8566997778616035,
      "grad_norm": 0.20762713551014986,
      "kl": 0.108001708984375,
      "learning_rate": 4.964586708195292e-07,
      "loss": 0.0001,
      "reward": 1.8285714760422707,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8285714536905289,
      "rewards/format_reward_func": 1.0,
      "step": 11074
    },
    {
      "completion_length": 228.83483219146729,
      "epoch": 1.857035081101471,
      "grad_norm": 0.18659784904839863,
      "kl": 0.14788818359375,
      "learning_rate": 4.964566586082916e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 11076
    },
    {
      "completion_length": 222.52233028411865,
      "epoch": 1.8573703843413387,
      "grad_norm": 0.19205268624150945,
      "kl": 0.10675048828125,
      "learning_rate": 4.964546458296194e-07,
      "loss": 0.0001,
      "reward": 1.8214286416769028,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8214285969734192,
      "rewards/format_reward_func": 1.0,
      "step": 11078
    },
    {
      "completion_length": 221.46429634094238,
      "epoch": 1.8577056875812064,
      "grad_norm": 0.2383659915228919,
      "kl": 0.12945556640625,
      "learning_rate": 4.964526324835172e-07,
      "loss": 0.0001,
      "reward": 1.732142947614193,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7321428935974836,
      "rewards/format_reward_func": 1.0,
      "step": 11080
    },
    {
      "completion_length": 211.19643878936768,
      "epoch": 1.8580409908210738,
      "grad_norm": 0.11302266977949367,
      "kl": 0.1028900146484375,
      "learning_rate": 4.964506185699897e-07,
      "loss": 0.0001,
      "reward": 1.8178571835160255,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.817857176065445,
      "rewards/format_reward_func": 1.0,
      "step": 11082
    },
    {
      "completion_length": 223.49554538726807,
      "epoch": 1.8583762940609414,
      "grad_norm": 0.23475124330163089,
      "kl": 0.14801025390625,
      "learning_rate": 4.964486040890415e-07,
      "loss": 0.0001,
      "reward": 1.710714377462864,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7107143215835094,
      "rewards/format_reward_func": 1.0,
      "step": 11084
    },
    {
      "completion_length": 219.42411708831787,
      "epoch": 1.8587115973008088,
      "grad_norm": 0.32645925916745405,
      "kl": 0.200714111328125,
      "learning_rate": 4.964465890406773e-07,
      "loss": 0.0002,
      "reward": 1.8071429133415222,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428909897804,
      "rewards/format_reward_func": 1.0,
      "step": 11086
    },
    {
      "completion_length": 223.35268878936768,
      "epoch": 1.8590469005406765,
      "grad_norm": 0.2379439074253615,
      "kl": 0.1185760498046875,
      "learning_rate": 4.964445734249015e-07,
      "loss": 0.0001,
      "reward": 1.78035718947649,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7848214656114578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11088
    },
    {
      "completion_length": 218.3526906967163,
      "epoch": 1.859382203780544,
      "grad_norm": 0.490915657263521,
      "kl": 0.137847900390625,
      "learning_rate": 4.96442557241719e-07,
      "loss": 0.0001,
      "reward": 1.721428669989109,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.721428606659174,
      "rewards/format_reward_func": 1.0,
      "step": 11090
    },
    {
      "completion_length": 219.01786518096924,
      "epoch": 1.8597175070204117,
      "grad_norm": 0.10672365885385886,
      "kl": 0.186767578125,
      "learning_rate": 4.964405404911344e-07,
      "loss": 0.0002,
      "reward": 1.7196429297327995,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7241071797907352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11092
    },
    {
      "completion_length": 216.41072368621826,
      "epoch": 1.8600528102602791,
      "grad_norm": 0.3824340210957931,
      "kl": 0.21832275390625,
      "learning_rate": 4.964385231731523e-07,
      "loss": 0.0002,
      "reward": 1.7464286461472511,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 11094
    },
    {
      "completion_length": 223.16072463989258,
      "epoch": 1.8603881135001465,
      "grad_norm": 0.23764974494911736,
      "kl": 0.187652587890625,
      "learning_rate": 4.964365052877773e-07,
      "loss": 0.0002,
      "reward": 1.7571429312229156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 11096
    },
    {
      "completion_length": 236.05358409881592,
      "epoch": 1.8607234167400142,
      "grad_norm": 0.2853067876926874,
      "kl": 0.215728759765625,
      "learning_rate": 4.96434486835014e-07,
      "loss": 0.0002,
      "reward": 1.7089286744594574,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.713392898440361,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11098
    },
    {
      "completion_length": 223.25001049041748,
      "epoch": 1.8610587199798818,
      "grad_norm": 0.3091993117408902,
      "kl": 0.2955322265625,
      "learning_rate": 4.964324678148674e-07,
      "loss": 0.0003,
      "reward": 1.725000075995922,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000275671482,
      "rewards/format_reward_func": 1.0,
      "step": 11100
    },
    {
      "completion_length": 215.76340293884277,
      "epoch": 1.8613940232197494,
      "grad_norm": 0.2468966823450781,
      "kl": 0.138336181640625,
      "learning_rate": 4.964304482273417e-07,
      "loss": 0.0001,
      "reward": 1.8142857626080513,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8142857290804386,
      "rewards/format_reward_func": 1.0,
      "step": 11102
    },
    {
      "completion_length": 222.14286613464355,
      "epoch": 1.8617293264596169,
      "grad_norm": 0.14444154096411288,
      "kl": 0.2125244140625,
      "learning_rate": 4.964284280724418e-07,
      "loss": 0.0002,
      "reward": 1.7392857745289803,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857372760773,
      "rewards/format_reward_func": 1.0,
      "step": 11104
    },
    {
      "completion_length": 212.61608028411865,
      "epoch": 1.8620646296994845,
      "grad_norm": 0.2764355620324343,
      "kl": 0.18096923828125,
      "learning_rate": 4.964264073501723e-07,
      "loss": 0.0002,
      "reward": 1.73214291036129,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7321428880095482,
      "rewards/format_reward_func": 1.0,
      "step": 11106
    },
    {
      "completion_length": 226.53572368621826,
      "epoch": 1.862399932939352,
      "grad_norm": 0.13167498736808136,
      "kl": 0.2672882080078125,
      "learning_rate": 4.964243860605378e-07,
      "loss": 0.0003,
      "reward": 1.8000000715255737,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 11108
    },
    {
      "completion_length": 218.80804538726807,
      "epoch": 1.8627352361792195,
      "grad_norm": 0.45637270918773215,
      "kl": 0.367919921875,
      "learning_rate": 4.96422364203543e-07,
      "loss": 0.0004,
      "reward": 1.8285714760422707,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8285714499652386,
      "rewards/format_reward_func": 1.0,
      "step": 11110
    },
    {
      "completion_length": 221.94643783569336,
      "epoch": 1.8630705394190872,
      "grad_norm": 0.24640437904082058,
      "kl": 0.381561279296875,
      "learning_rate": 4.964203417791926e-07,
      "loss": 0.0004,
      "reward": 1.7607143595814705,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7687500230967999,
      "rewards/format_reward_func": 0.9919642955064774,
      "step": 11112
    },
    {
      "completion_length": 222.49554538726807,
      "epoch": 1.8634058426589548,
      "grad_norm": 0.08996637829383648,
      "kl": 0.1153564453125,
      "learning_rate": 4.964183187874912e-07,
      "loss": 0.0001,
      "reward": 1.7321429401636124,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7321428954601288,
      "rewards/format_reward_func": 1.0,
      "step": 11114
    },
    {
      "completion_length": 210.2009038925171,
      "epoch": 1.8637411458988222,
      "grad_norm": 0.30682680867562007,
      "kl": 0.525390625,
      "learning_rate": 4.964162952284435e-07,
      "loss": 0.0005,
      "reward": 1.7892857789993286,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857529222965,
      "rewards/format_reward_func": 1.0,
      "step": 11116
    },
    {
      "completion_length": 219.20983123779297,
      "epoch": 1.8640764491386896,
      "grad_norm": 0.21939131000569564,
      "kl": 0.161834716796875,
      "learning_rate": 4.964142711020539e-07,
      "loss": 0.0002,
      "reward": 1.7696429342031479,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11118
    },
    {
      "completion_length": 220.9196548461914,
      "epoch": 1.8644117523785573,
      "grad_norm": 0.24876917172333088,
      "kl": 0.163299560546875,
      "learning_rate": 4.964122464083275e-07,
      "loss": 0.0002,
      "reward": 1.7357143610715866,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 11120
    },
    {
      "completion_length": 216.77233123779297,
      "epoch": 1.864747055618425,
      "grad_norm": 0.23915510555329103,
      "kl": 0.4630126953125,
      "learning_rate": 4.964102211472687e-07,
      "loss": 0.0005,
      "reward": 1.725000075995922,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7250000387430191,
      "rewards/format_reward_func": 1.0,
      "step": 11122
    },
    {
      "completion_length": 221.33929538726807,
      "epoch": 1.8650823588582925,
      "grad_norm": 0.24977953166575798,
      "kl": 0.19873046875,
      "learning_rate": 4.964081953188822e-07,
      "loss": 0.0002,
      "reward": 1.8071429058909416,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428760886192,
      "rewards/format_reward_func": 1.0,
      "step": 11124
    },
    {
      "completion_length": 226.56250762939453,
      "epoch": 1.8654176620981602,
      "grad_norm": 0.4095663196737921,
      "kl": 0.487213134765625,
      "learning_rate": 4.964061689231727e-07,
      "loss": 0.0005,
      "reward": 1.7750000581145287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000171363354,
      "rewards/format_reward_func": 1.0,
      "step": 11126
    },
    {
      "completion_length": 228.93750858306885,
      "epoch": 1.8657529653380276,
      "grad_norm": 0.30583252427411733,
      "kl": 0.337738037109375,
      "learning_rate": 4.964041419601448e-07,
      "loss": 0.0003,
      "reward": 1.7250000685453415,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000238418579,
      "rewards/format_reward_func": 1.0,
      "step": 11128
    },
    {
      "completion_length": 224.008939743042,
      "epoch": 1.866088268577895,
      "grad_norm": 0.13443705049530696,
      "kl": 0.214447021484375,
      "learning_rate": 4.964021144298032e-07,
      "loss": 0.0002,
      "reward": 1.7375000640749931,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7419643141329288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11130
    },
    {
      "completion_length": 229.95536708831787,
      "epoch": 1.8664235718177626,
      "grad_norm": 0.6325187220459428,
      "kl": 1.204498291015625,
      "learning_rate": 4.964000863321526e-07,
      "loss": 0.0012,
      "reward": 1.7160715088248253,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7205357551574707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11132
    },
    {
      "completion_length": 231.96429634094238,
      "epoch": 1.8667588750576303,
      "grad_norm": 0.11808471962014658,
      "kl": 0.1728515625,
      "learning_rate": 4.963980576671977e-07,
      "loss": 0.0002,
      "reward": 1.7857143431901932,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143059372902,
      "rewards/format_reward_func": 1.0,
      "step": 11134
    },
    {
      "completion_length": 233.20090293884277,
      "epoch": 1.867094178297498,
      "grad_norm": 0.17947428004964994,
      "kl": 0.18927001953125,
      "learning_rate": 4.96396028434943e-07,
      "loss": 0.0002,
      "reward": 1.7571429014205933,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 11136
    },
    {
      "completion_length": 234.05804920196533,
      "epoch": 1.8674294815373653,
      "grad_norm": 0.23106300919668743,
      "kl": 0.202972412109375,
      "learning_rate": 4.963939986353934e-07,
      "loss": 0.0002,
      "reward": 1.765178643167019,
      "reward_std": 0.07954951096326113,
      "rewards/equation_reward_func": 0.7669643089175224,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 11138
    },
    {
      "completion_length": 228.23661708831787,
      "epoch": 1.867764784777233,
      "grad_norm": 0.2550263269785493,
      "kl": 0.206146240234375,
      "learning_rate": 4.963919682685533e-07,
      "loss": 0.0002,
      "reward": 1.8214286416769028,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.821428582072258,
      "rewards/format_reward_func": 1.0,
      "step": 11140
    },
    {
      "completion_length": 228.85715198516846,
      "epoch": 1.8681000880171004,
      "grad_norm": 0.17487852501349144,
      "kl": 0.12628173828125,
      "learning_rate": 4.963899373344276e-07,
      "loss": 0.0001,
      "reward": 1.8607143312692642,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8607143051922321,
      "rewards/format_reward_func": 1.0,
      "step": 11142
    },
    {
      "completion_length": 234.2991180419922,
      "epoch": 1.868435391256968,
      "grad_norm": 0.1105321170059028,
      "kl": 0.12158203125,
      "learning_rate": 4.963879058330209e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571660071611,
      "rewards/format_reward_func": 1.0,
      "step": 11144
    },
    {
      "completion_length": 234.05804538726807,
      "epoch": 1.8687706944968356,
      "grad_norm": 0.14043168729058803,
      "kl": 0.199127197265625,
      "learning_rate": 4.963858737643379e-07,
      "loss": 0.0002,
      "reward": 1.7964286357164383,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7964286059141159,
      "rewards/format_reward_func": 1.0,
      "step": 11146
    },
    {
      "completion_length": 227.93750762939453,
      "epoch": 1.8691059977367033,
      "grad_norm": 0.23978752547022278,
      "kl": 0.2164306640625,
      "learning_rate": 4.963838411283834e-07,
      "loss": 0.0002,
      "reward": 1.7035715207457542,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7035714685916901,
      "rewards/format_reward_func": 1.0,
      "step": 11148
    },
    {
      "completion_length": 239.415189743042,
      "epoch": 1.8694413009765707,
      "grad_norm": 0.17226054792818743,
      "kl": 0.137786865234375,
      "learning_rate": 4.963818079251618e-07,
      "loss": 0.0001,
      "reward": 1.746428668498993,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.746428593993187,
      "rewards/format_reward_func": 1.0,
      "step": 11150
    },
    {
      "completion_length": 234.7366189956665,
      "epoch": 1.869776604216438,
      "grad_norm": 0.563115423768281,
      "kl": 0.161102294921875,
      "learning_rate": 4.963797741546779e-07,
      "loss": 0.0002,
      "reward": 1.7535715028643608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.753571443259716,
      "rewards/format_reward_func": 1.0,
      "step": 11152
    },
    {
      "completion_length": 231.68750762939453,
      "epoch": 1.8701119074563057,
      "grad_norm": 0.13526243051831918,
      "kl": 0.117584228515625,
      "learning_rate": 4.963777398169365e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571902215481,
      "rewards/format_reward_func": 1.0,
      "step": 11154
    },
    {
      "completion_length": 230.14286708831787,
      "epoch": 1.8704472106961734,
      "grad_norm": 0.39458122341108914,
      "kl": 0.1586151123046875,
      "learning_rate": 4.963757049119421e-07,
      "loss": 0.0002,
      "reward": 1.7982143387198448,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.802678607404232,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11156
    },
    {
      "completion_length": 232.852689743042,
      "epoch": 1.870782513936041,
      "grad_norm": 0.2544117423378524,
      "kl": 0.13641357421875,
      "learning_rate": 4.963736694396996e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 11158
    },
    {
      "completion_length": 230.69643878936768,
      "epoch": 1.8711178171759084,
      "grad_norm": 0.2988204347526544,
      "kl": 0.113555908203125,
      "learning_rate": 4.963716334002135e-07,
      "loss": 0.0001,
      "reward": 1.7589286416769028,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928954601288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11160
    },
    {
      "completion_length": 244.852689743042,
      "epoch": 1.871453120415776,
      "grad_norm": 0.2732977459117658,
      "kl": 0.110443115234375,
      "learning_rate": 4.963695967934886e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714525729418,
      "rewards/format_reward_func": 1.0,
      "step": 11162
    },
    {
      "completion_length": 231.44643783569336,
      "epoch": 1.8717884236556435,
      "grad_norm": 0.20912566072607522,
      "kl": 0.106414794921875,
      "learning_rate": 4.963675596195295e-07,
      "loss": 0.0001,
      "reward": 1.7089286521077156,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7133928909897804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11164
    },
    {
      "completion_length": 234.3437614440918,
      "epoch": 1.872123726895511,
      "grad_norm": 0.4023109968185387,
      "kl": 0.12664794921875,
      "learning_rate": 4.963655218783409e-07,
      "loss": 0.0001,
      "reward": 1.7339286282658577,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.738392885774374,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11166
    },
    {
      "completion_length": 235.37054824829102,
      "epoch": 1.8724590301353787,
      "grad_norm": 0.2784812271855098,
      "kl": 0.12969970703125,
      "learning_rate": 4.963634835699275e-07,
      "loss": 0.0001,
      "reward": 1.7482143566012383,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526785992085934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11168
    },
    {
      "completion_length": 235.15179538726807,
      "epoch": 1.8727943333752464,
      "grad_norm": 0.17483775044451044,
      "kl": 0.113037109375,
      "learning_rate": 4.963614446942941e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 11170
    },
    {
      "completion_length": 233.7410831451416,
      "epoch": 1.8731296366151138,
      "grad_norm": 0.12216847250099579,
      "kl": 0.0994415283203125,
      "learning_rate": 4.963594052514453e-07,
      "loss": 0.0001,
      "reward": 1.8071429282426834,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8071428835391998,
      "rewards/format_reward_func": 1.0,
      "step": 11172
    },
    {
      "completion_length": 230.18304634094238,
      "epoch": 1.8734649398549812,
      "grad_norm": 0.4168596205302882,
      "kl": 0.11090087890625,
      "learning_rate": 4.963573652413858e-07,
      "loss": 0.0001,
      "reward": 1.766071505844593,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7705357521772385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11174
    },
    {
      "completion_length": 229.29911613464355,
      "epoch": 1.8738002430948488,
      "grad_norm": 0.2791899452311946,
      "kl": 0.118316650390625,
      "learning_rate": 4.963553246641203e-07,
      "loss": 0.0001,
      "reward": 1.8357143625617027,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.8357143104076385,
      "rewards/format_reward_func": 1.0,
      "step": 11176
    },
    {
      "completion_length": 231.80358028411865,
      "epoch": 1.8741355463347165,
      "grad_norm": 0.16582448671376948,
      "kl": 0.12030029296875,
      "learning_rate": 4.963532835196534e-07,
      "loss": 0.0001,
      "reward": 1.807142898440361,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.807142898440361,
      "rewards/format_reward_func": 1.0,
      "step": 11178
    },
    {
      "completion_length": 234.4509048461914,
      "epoch": 1.874470849574584,
      "grad_norm": 0.2595491378811056,
      "kl": 0.122222900390625,
      "learning_rate": 4.9635124180799e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857231199741,
      "rewards/format_reward_func": 1.0,
      "step": 11180
    },
    {
      "completion_length": 240.40179634094238,
      "epoch": 1.8748061528144517,
      "grad_norm": 0.14194016819276536,
      "kl": 0.11871337890625,
      "learning_rate": 4.963491995291347e-07,
      "loss": 0.0001,
      "reward": 1.7875000461935997,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.791964303702116,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11182
    },
    {
      "completion_length": 232.55804538726807,
      "epoch": 1.8751414560543191,
      "grad_norm": 0.12892934407277282,
      "kl": 0.11798095703125,
      "learning_rate": 4.963471566830922e-07,
      "loss": 0.0001,
      "reward": 1.7553572207689285,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7598214745521545,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11184
    },
    {
      "completion_length": 235.81251049041748,
      "epoch": 1.8754767592941866,
      "grad_norm": 0.25378944592714786,
      "kl": 0.11944580078125,
      "learning_rate": 4.963451132698672e-07,
      "loss": 0.0001,
      "reward": 1.8357143253087997,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8357142992317677,
      "rewards/format_reward_func": 1.0,
      "step": 11186
    },
    {
      "completion_length": 250.5669755935669,
      "epoch": 1.8758120625340542,
      "grad_norm": 0.615027955265174,
      "kl": 0.1882781982421875,
      "learning_rate": 4.963430692894644e-07,
      "loss": 0.0002,
      "reward": 1.7375000938773155,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7419643066823483,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11188
    },
    {
      "completion_length": 244.79465293884277,
      "epoch": 1.8761473657739218,
      "grad_norm": 0.2411759958789673,
      "kl": 0.1248626708984375,
      "learning_rate": 4.963410247418886e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000417232513,
      "rewards/format_reward_func": 1.0,
      "step": 11190
    },
    {
      "completion_length": 244.06250953674316,
      "epoch": 1.8764826690137895,
      "grad_norm": 0.3607120042957956,
      "kl": 0.274566650390625,
      "learning_rate": 4.963389796271443e-07,
      "loss": 0.0003,
      "reward": 1.6946429535746574,
      "reward_std": 0.09848987031728029,
      "rewards/equation_reward_func": 0.6991071961820126,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11192
    },
    {
      "completion_length": 241.02680015563965,
      "epoch": 1.8768179722536569,
      "grad_norm": 0.179436973418576,
      "kl": 0.150360107421875,
      "learning_rate": 4.963369339452363e-07,
      "loss": 0.0002,
      "reward": 1.773214340209961,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7776785902678967,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11194
    },
    {
      "completion_length": 249.25447463989258,
      "epoch": 1.8771532754935245,
      "grad_norm": 0.22366962597111079,
      "kl": 0.215087890625,
      "learning_rate": 4.963348876961695e-07,
      "loss": 0.0002,
      "reward": 1.7660714909434319,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7705357447266579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11196
    },
    {
      "completion_length": 239.75893783569336,
      "epoch": 1.877488578733392,
      "grad_norm": 0.5282950834770903,
      "kl": 0.2510986328125,
      "learning_rate": 4.963328408799484e-07,
      "loss": 0.0003,
      "reward": 1.757142923772335,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 11198
    },
    {
      "completion_length": 244.62500858306885,
      "epoch": 1.8778238819732596,
      "grad_norm": 0.17760628742076728,
      "kl": 0.205841064453125,
      "learning_rate": 4.963307934965777e-07,
      "loss": 0.0002,
      "reward": 1.789285771548748,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7982143275439739,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11200
    },
    {
      "completion_length": 248.48215579986572,
      "epoch": 1.8781591852131272,
      "grad_norm": 0.19333171248821832,
      "kl": 0.146392822265625,
      "learning_rate": 4.963287455460622e-07,
      "loss": 0.0001,
      "reward": 1.7821428999304771,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 11202
    },
    {
      "completion_length": 246.0000123977661,
      "epoch": 1.8784944884529948,
      "grad_norm": 0.26767903257350883,
      "kl": 0.162445068359375,
      "learning_rate": 4.963266970284067e-07,
      "loss": 0.0002,
      "reward": 1.7267857789993286,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.731250025331974,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11204
    },
    {
      "completion_length": 249.42858219146729,
      "epoch": 1.8788297916928622,
      "grad_norm": 0.15949540283927346,
      "kl": 0.176422119140625,
      "learning_rate": 4.963246479436157e-07,
      "loss": 0.0002,
      "reward": 1.8000000566244125,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000342726707,
      "rewards/format_reward_func": 1.0,
      "step": 11206
    },
    {
      "completion_length": 235.9687614440918,
      "epoch": 1.8791650949327297,
      "grad_norm": 0.22833940214050732,
      "kl": 0.18597412109375,
      "learning_rate": 4.96322598291694e-07,
      "loss": 0.0002,
      "reward": 1.7892857640981674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857305705547,
      "rewards/format_reward_func": 1.0,
      "step": 11208
    },
    {
      "completion_length": 252.52233695983887,
      "epoch": 1.8795003981725973,
      "grad_norm": 0.16628686527578662,
      "kl": 0.2346954345703125,
      "learning_rate": 4.963205480726465e-07,
      "loss": 0.0002,
      "reward": 1.7964285984635353,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 11210
    },
    {
      "completion_length": 240.1741180419922,
      "epoch": 1.879835701412465,
      "grad_norm": 0.3176503045231156,
      "kl": 0.28228759765625,
      "learning_rate": 4.963184972864776e-07,
      "loss": 0.0003,
      "reward": 1.7607143819332123,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 11212
    },
    {
      "completion_length": 243.00001525878906,
      "epoch": 1.8801710046523326,
      "grad_norm": 0.17669946946339002,
      "kl": 0.12799072265625,
      "learning_rate": 4.963164459331924e-07,
      "loss": 0.0001,
      "reward": 1.7875000312924385,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7919643241912127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11214
    },
    {
      "completion_length": 241.74108219146729,
      "epoch": 1.8805063078922,
      "grad_norm": 0.11810350754256321,
      "kl": 0.107452392578125,
      "learning_rate": 4.963143940127953e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857491970062,
      "rewards/format_reward_func": 1.0,
      "step": 11216
    },
    {
      "completion_length": 239.05804443359375,
      "epoch": 1.8808416111320676,
      "grad_norm": 0.24622105000391542,
      "kl": 0.14434814453125,
      "learning_rate": 4.963123415252911e-07,
      "loss": 0.0001,
      "reward": 1.730357214808464,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7348214611411095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11218
    },
    {
      "completion_length": 247.7544765472412,
      "epoch": 1.881176914371935,
      "grad_norm": 0.20972882425954445,
      "kl": 0.20135498046875,
      "learning_rate": 4.963102884706845e-07,
      "loss": 0.0002,
      "reward": 1.7428572103381157,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571693599224,
      "rewards/format_reward_func": 1.0,
      "step": 11220
    },
    {
      "completion_length": 232.7142972946167,
      "epoch": 1.8815122176118027,
      "grad_norm": 0.2750278904808387,
      "kl": 0.192230224609375,
      "learning_rate": 4.963082348489804e-07,
      "loss": 0.0002,
      "reward": 1.7642857655882835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 11222
    },
    {
      "completion_length": 242.1250123977661,
      "epoch": 1.8818475208516703,
      "grad_norm": 0.3315015096714337,
      "kl": 0.336578369140625,
      "learning_rate": 4.963061806601835e-07,
      "loss": 0.0003,
      "reward": 1.7642857730388641,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 11224
    },
    {
      "completion_length": 234.46429443359375,
      "epoch": 1.882182824091538,
      "grad_norm": 0.09096484363912223,
      "kl": 0.1370697021484375,
      "learning_rate": 4.963041259042984e-07,
      "loss": 0.0001,
      "reward": 1.7589286491274834,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7633928805589676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11226
    },
    {
      "completion_length": 230.73215198516846,
      "epoch": 1.8825181273314053,
      "grad_norm": 0.16681827582170852,
      "kl": 0.13641357421875,
      "learning_rate": 4.963020705813297e-07,
      "loss": 0.0001,
      "reward": 1.835714340209961,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8357142992317677,
      "rewards/format_reward_func": 1.0,
      "step": 11228
    },
    {
      "completion_length": 234.8303680419922,
      "epoch": 1.8828534305712727,
      "grad_norm": 0.37642315561234024,
      "kl": 0.120574951171875,
      "learning_rate": 4.963000146912825e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714286111295223,
      "rewards/format_reward_func": 1.0,
      "step": 11230
    },
    {
      "completion_length": 240.70983219146729,
      "epoch": 1.8831887338111404,
      "grad_norm": 0.23413257547900723,
      "kl": 0.123199462890625,
      "learning_rate": 4.962979582341613e-07,
      "loss": 0.0001,
      "reward": 1.8071429282426834,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428760886192,
      "rewards/format_reward_func": 1.0,
      "step": 11232
    },
    {
      "completion_length": 236.7098331451416,
      "epoch": 1.883524037051008,
      "grad_norm": 0.2838174612146134,
      "kl": 0.137664794921875,
      "learning_rate": 4.962959012099709e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 11234
    },
    {
      "completion_length": 237.9821548461914,
      "epoch": 1.8838593402908757,
      "grad_norm": 0.3923008916019813,
      "kl": 0.157470703125,
      "learning_rate": 4.96293843618716e-07,
      "loss": 0.0002,
      "reward": 1.7750000581145287,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 11236
    },
    {
      "completion_length": 238.98215293884277,
      "epoch": 1.884194643530743,
      "grad_norm": 0.2385679361089355,
      "kl": 0.1591796875,
      "learning_rate": 4.962917854604013e-07,
      "loss": 0.0002,
      "reward": 1.7642857655882835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857581377029,
      "rewards/format_reward_func": 1.0,
      "step": 11238
    },
    {
      "completion_length": 235.0759048461914,
      "epoch": 1.8845299467706107,
      "grad_norm": 0.14434117277451863,
      "kl": 0.12469482421875,
      "learning_rate": 4.962897267350316e-07,
      "loss": 0.0001,
      "reward": 1.7357143685221672,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7357143256813288,
      "rewards/format_reward_func": 1.0,
      "step": 11240
    },
    {
      "completion_length": 223.95090293884277,
      "epoch": 1.884865250010478,
      "grad_norm": 0.03234523914231118,
      "kl": 0.12738037109375,
      "learning_rate": 4.962876674426116e-07,
      "loss": 0.0001,
      "reward": 1.8107143342494965,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143267989159,
      "rewards/format_reward_func": 1.0,
      "step": 11242
    },
    {
      "completion_length": 244.4196538925171,
      "epoch": 1.8852005532503457,
      "grad_norm": 0.3593552515735288,
      "kl": 0.238555908203125,
      "learning_rate": 4.962856075831462e-07,
      "loss": 0.0002,
      "reward": 1.6785715222358704,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.6785714607685804,
      "rewards/format_reward_func": 1.0,
      "step": 11244
    },
    {
      "completion_length": 246.20537090301514,
      "epoch": 1.8855358564902134,
      "grad_norm": 0.37697301198383576,
      "kl": 0.193267822265625,
      "learning_rate": 4.962835471566399e-07,
      "loss": 0.0002,
      "reward": 1.7839286252856255,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7883928716182709,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11246
    },
    {
      "completion_length": 248.55804824829102,
      "epoch": 1.885871159730081,
      "grad_norm": 0.23996368764434198,
      "kl": 0.1488037109375,
      "learning_rate": 4.962814861630977e-07,
      "loss": 0.0001,
      "reward": 1.7375000938773155,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7419643066823483,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11248
    },
    {
      "completion_length": 253.0000114440918,
      "epoch": 1.8862064629699484,
      "grad_norm": 0.18681771723186258,
      "kl": 0.12933349609375,
      "learning_rate": 4.96279424602524e-07,
      "loss": 0.0001,
      "reward": 1.7517857924103737,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500238418579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11250
    },
    {
      "completion_length": 238.5446548461914,
      "epoch": 1.8865417662098158,
      "grad_norm": 0.10783026267363756,
      "kl": 0.18572998046875,
      "learning_rate": 4.962773624749239e-07,
      "loss": 0.0002,
      "reward": 1.796428620815277,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 11252
    },
    {
      "completion_length": 229.48215293884277,
      "epoch": 1.8868770694496835,
      "grad_norm": 0.19649646385163824,
      "kl": 0.105621337890625,
      "learning_rate": 4.96275299780302e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 11254
    },
    {
      "completion_length": 248.9866189956665,
      "epoch": 1.887212372689551,
      "grad_norm": 0.28251356223769303,
      "kl": 0.43621826171875,
      "learning_rate": 4.96273236518663e-07,
      "loss": 0.0004,
      "reward": 1.7767857685685158,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.781250037252903,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11256
    },
    {
      "completion_length": 229.65179634094238,
      "epoch": 1.8875476759294187,
      "grad_norm": 0.09993092682238044,
      "kl": 0.1275787353515625,
      "learning_rate": 4.962711726900117e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 11258
    },
    {
      "completion_length": 235.91518878936768,
      "epoch": 1.8878829791692864,
      "grad_norm": 0.2383186917676842,
      "kl": 0.11273193359375,
      "learning_rate": 4.962691082943528e-07,
      "loss": 0.0001,
      "reward": 1.7482143491506577,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7526785843074322,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11260
    },
    {
      "completion_length": 240.62947368621826,
      "epoch": 1.8882182824091538,
      "grad_norm": 0.12872696701242822,
      "kl": 0.2894744873046875,
      "learning_rate": 4.962670433316912e-07,
      "loss": 0.0003,
      "reward": 1.7660714834928513,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11262
    },
    {
      "completion_length": 237.12054634094238,
      "epoch": 1.8885535856490212,
      "grad_norm": 0.2637527234686633,
      "kl": 0.201141357421875,
      "learning_rate": 4.962649778020316e-07,
      "loss": 0.0002,
      "reward": 1.7535715103149414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714637488127,
      "rewards/format_reward_func": 1.0,
      "step": 11264
    },
    {
      "completion_length": 241.26340198516846,
      "epoch": 1.8888888888888888,
      "grad_norm": 0.19499182893264433,
      "kl": 0.1104736328125,
      "learning_rate": 4.962629117053786e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 11266
    },
    {
      "completion_length": 247.0312623977661,
      "epoch": 1.8892241921287565,
      "grad_norm": 0.34806357799507837,
      "kl": 0.128662109375,
      "learning_rate": 4.962608450417371e-07,
      "loss": 0.0001,
      "reward": 1.6964286416769028,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.6964286100119352,
      "rewards/format_reward_func": 1.0,
      "step": 11268
    },
    {
      "completion_length": 243.08037090301514,
      "epoch": 1.889559495368624,
      "grad_norm": 0.13210932993568505,
      "kl": 0.141876220703125,
      "learning_rate": 4.962587778111119e-07,
      "loss": 0.0001,
      "reward": 1.832142896950245,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8321428745985031,
      "rewards/format_reward_func": 1.0,
      "step": 11270
    },
    {
      "completion_length": 247.37947845458984,
      "epoch": 1.8898947986084915,
      "grad_norm": 0.37825347713712715,
      "kl": 0.171356201171875,
      "learning_rate": 4.962567100135075e-07,
      "loss": 0.0002,
      "reward": 1.7642857804894447,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7642857544124126,
      "rewards/format_reward_func": 1.0,
      "step": 11272
    },
    {
      "completion_length": 250.8884038925171,
      "epoch": 1.8902301018483592,
      "grad_norm": 0.25807449253969955,
      "kl": 0.112640380859375,
      "learning_rate": 4.962546416489289e-07,
      "loss": 0.0001,
      "reward": 1.7589286491274834,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7633928954601288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11274
    },
    {
      "completion_length": 252.50894165039062,
      "epoch": 1.8905654050882266,
      "grad_norm": 0.3161104987993201,
      "kl": 0.10699462890625,
      "learning_rate": 4.962525727173809e-07,
      "loss": 0.0001,
      "reward": 1.746428668498993,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 11276
    },
    {
      "completion_length": 250.54912090301514,
      "epoch": 1.8909007083280942,
      "grad_norm": 0.26533155739866404,
      "kl": 0.1735382080078125,
      "learning_rate": 4.962505032188682e-07,
      "loss": 0.0002,
      "reward": 1.725000075995922,
      "reward_std": 0.05555838719010353,
      "rewards/equation_reward_func": 0.7250000331550837,
      "rewards/format_reward_func": 1.0,
      "step": 11278
    },
    {
      "completion_length": 252.9866180419922,
      "epoch": 1.8912360115679618,
      "grad_norm": 0.269657416281771,
      "kl": 0.264495849609375,
      "learning_rate": 4.962484331533955e-07,
      "loss": 0.0003,
      "reward": 1.7000000849366188,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7000000365078449,
      "rewards/format_reward_func": 1.0,
      "step": 11280
    },
    {
      "completion_length": 255.80358600616455,
      "epoch": 1.8915713148078295,
      "grad_norm": 0.11395970195478738,
      "kl": 0.197479248046875,
      "learning_rate": 4.962463625209676e-07,
      "loss": 0.0002,
      "reward": 1.773214340209961,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7776786051690578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11282
    },
    {
      "completion_length": 252.80804634094238,
      "epoch": 1.8919066180476969,
      "grad_norm": 0.20015288301172346,
      "kl": 0.2366180419921875,
      "learning_rate": 4.962442913215892e-07,
      "loss": 0.0002,
      "reward": 1.7607143446803093,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.7696428839117289,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11284
    },
    {
      "completion_length": 265.46429920196533,
      "epoch": 1.8922419212875643,
      "grad_norm": 0.3071617414978232,
      "kl": 0.1317138671875,
      "learning_rate": 4.962422195552652e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7767857313156128,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11286
    },
    {
      "completion_length": 264.34822845458984,
      "epoch": 1.892577224527432,
      "grad_norm": 0.5844285027634494,
      "kl": 0.6663818359375,
      "learning_rate": 4.962401472220004e-07,
      "loss": 0.0007,
      "reward": 1.7375000640749931,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643178582191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11288
    },
    {
      "completion_length": 257.7276906967163,
      "epoch": 1.8929125277672996,
      "grad_norm": 0.18094547642963446,
      "kl": 0.4002685546875,
      "learning_rate": 4.962380743217994e-07,
      "loss": 0.0004,
      "reward": 1.7250000983476639,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7250000275671482,
      "rewards/format_reward_func": 1.0,
      "step": 11290
    },
    {
      "completion_length": 256.72322845458984,
      "epoch": 1.8932478310071672,
      "grad_norm": 0.31669459769976144,
      "kl": 0.608367919921875,
      "learning_rate": 4.96236000854667e-07,
      "loss": 0.0006,
      "reward": 1.757142923772335,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428772062063,
      "rewards/format_reward_func": 1.0,
      "step": 11292
    },
    {
      "completion_length": 249.0491189956665,
      "epoch": 1.8935831342470346,
      "grad_norm": 0.23547687690667424,
      "kl": 0.344940185546875,
      "learning_rate": 4.962339268206081e-07,
      "loss": 0.0003,
      "reward": 1.7875000536441803,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7919643074274063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11294
    },
    {
      "completion_length": 250.0669755935669,
      "epoch": 1.8939184374869023,
      "grad_norm": 0.44210786853010475,
      "kl": 0.396240234375,
      "learning_rate": 4.962318522196274e-07,
      "loss": 0.0004,
      "reward": 1.7125000953674316,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7169643193483353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11296
    },
    {
      "completion_length": 264.73215198516846,
      "epoch": 1.8942537407267697,
      "grad_norm": 0.2360487677899644,
      "kl": 0.343963623046875,
      "learning_rate": 4.962297770517296e-07,
      "loss": 0.0003,
      "reward": 1.801785759627819,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.8151785954833031,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11298
    },
    {
      "completion_length": 253.4285831451416,
      "epoch": 1.8945890439666373,
      "grad_norm": 0.19174211335333957,
      "kl": 0.182464599609375,
      "learning_rate": 4.962277013169197e-07,
      "loss": 0.0002,
      "reward": 1.7089286521077156,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7133928909897804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11300
    },
    {
      "completion_length": 262.34376335144043,
      "epoch": 1.894924347206505,
      "grad_norm": 0.4497420045476892,
      "kl": 0.34619140625,
      "learning_rate": 4.962256250152022e-07,
      "loss": 0.0003,
      "reward": 1.6500000730156898,
      "reward_std": 0.07071067858487368,
      "rewards/equation_reward_func": 0.6678571738302708,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 11302
    },
    {
      "completion_length": 266.2901906967163,
      "epoch": 1.8952596504463726,
      "grad_norm": 0.19696851735656268,
      "kl": 0.706573486328125,
      "learning_rate": 4.962235481465821e-07,
      "loss": 0.0007,
      "reward": 1.7504464909434319,
      "reward_std": 0.0700793326832354,
      "rewards/equation_reward_func": 0.7651786096394062,
      "rewards/format_reward_func": 0.9852678664028645,
      "step": 11304
    },
    {
      "completion_length": 257.4821538925171,
      "epoch": 1.89559495368624,
      "grad_norm": 0.2895527404607826,
      "kl": 0.30169677734375,
      "learning_rate": 4.962214707110641e-07,
      "loss": 0.0003,
      "reward": 1.7000000849366188,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.708928607404232,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11306
    },
    {
      "completion_length": 245.58929824829102,
      "epoch": 1.8959302569261074,
      "grad_norm": 0.3727980145309879,
      "kl": 1.33642578125,
      "learning_rate": 4.96219392708653e-07,
      "loss": 0.0013,
      "reward": 1.7535715103149414,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7625000365078449,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11308
    },
    {
      "completion_length": 251.3259048461914,
      "epoch": 1.896265560165975,
      "grad_norm": 0.08608763982420042,
      "kl": 1.362823486328125,
      "learning_rate": 4.962173141393535e-07,
      "loss": 0.0014,
      "reward": 1.801785759627819,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8062500320374966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11310
    },
    {
      "completion_length": 251.4330472946167,
      "epoch": 1.8966008634058427,
      "grad_norm": 0.2428648181228268,
      "kl": 0.157745361328125,
      "learning_rate": 4.962152350031704e-07,
      "loss": 0.0002,
      "reward": 1.7303572297096252,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7348214462399483,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11312
    },
    {
      "completion_length": 260.3080472946167,
      "epoch": 1.8969361666457103,
      "grad_norm": 0.2209283834755941,
      "kl": 0.304290771484375,
      "learning_rate": 4.962131553001086e-07,
      "loss": 0.0003,
      "reward": 1.7446429133415222,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7491071820259094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11314
    },
    {
      "completion_length": 248.6696548461914,
      "epoch": 1.897271469885578,
      "grad_norm": 0.159877997058151,
      "kl": 0.5746612548828125,
      "learning_rate": 4.962110750301729e-07,
      "loss": 0.0006,
      "reward": 1.759375050663948,
      "reward_std": 0.017046324210241437,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 11316
    },
    {
      "completion_length": 260.1384038925171,
      "epoch": 1.8976067731254453,
      "grad_norm": 0.17812440090711842,
      "kl": 0.610321044921875,
      "learning_rate": 4.96208994193368e-07,
      "loss": 0.0006,
      "reward": 1.7660714983940125,
      "reward_std": 0.07828682288527489,
      "rewards/equation_reward_func": 0.7794643212109804,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11318
    },
    {
      "completion_length": 248.71875858306885,
      "epoch": 1.8979420763653128,
      "grad_norm": 0.22698536374050526,
      "kl": 0.59393310546875,
      "learning_rate": 4.962069127896987e-07,
      "loss": 0.0006,
      "reward": 1.841071456670761,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.8455357290804386,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11320
    },
    {
      "completion_length": 253.8259038925171,
      "epoch": 1.8982773796051804,
      "grad_norm": 0.21589363012833795,
      "kl": 0.1286773681640625,
      "learning_rate": 4.962048308191698e-07,
      "loss": 0.0001,
      "reward": 1.7660714983940125,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7705357484519482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11322
    },
    {
      "completion_length": 256.321439743042,
      "epoch": 1.898612682845048,
      "grad_norm": 0.13809367925350216,
      "kl": 1.128448486328125,
      "learning_rate": 4.96202748281786e-07,
      "loss": 0.0011,
      "reward": 1.7625000551342964,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643051922321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11324
    },
    {
      "completion_length": 255.62054538726807,
      "epoch": 1.8989479860849157,
      "grad_norm": 0.1571339857944827,
      "kl": 0.396942138671875,
      "learning_rate": 4.962006651775522e-07,
      "loss": 0.0004,
      "reward": 1.7535714954137802,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714730620384,
      "rewards/format_reward_func": 1.0,
      "step": 11326
    },
    {
      "completion_length": 263.977689743042,
      "epoch": 1.899283289324783,
      "grad_norm": 0.3801393073801637,
      "kl": 0.1540679931640625,
      "learning_rate": 4.961985815064732e-07,
      "loss": 0.0002,
      "reward": 1.7803572043776512,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7848214581608772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11328
    },
    {
      "completion_length": 259.214298248291,
      "epoch": 1.8996185925646507,
      "grad_norm": 0.007573744361931633,
      "kl": 0.76910400390625,
      "learning_rate": 4.961964972685539e-07,
      "loss": 0.0008,
      "reward": 1.78035718947649,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11330
    },
    {
      "completion_length": 266.71875953674316,
      "epoch": 1.8999538958045181,
      "grad_norm": 0.18791643353272014,
      "kl": 0.676361083984375,
      "learning_rate": 4.961944124637989e-07,
      "loss": 0.0007,
      "reward": 1.723214365541935,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7276786081492901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11332
    },
    {
      "completion_length": 257.7366199493408,
      "epoch": 1.9002891990443858,
      "grad_norm": 0.20574448512918195,
      "kl": 0.811431884765625,
      "learning_rate": 4.96192327092213e-07,
      "loss": 0.0008,
      "reward": 1.730357214808464,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7348214723169804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11334
    },
    {
      "completion_length": 266.5669765472412,
      "epoch": 1.9006245022842534,
      "grad_norm": 0.12088067703855519,
      "kl": 0.2629852294921875,
      "learning_rate": 4.961902411538013e-07,
      "loss": 0.0003,
      "reward": 1.739285759627819,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7482143305242062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11336
    },
    {
      "completion_length": 275.45537090301514,
      "epoch": 1.900959805524121,
      "grad_norm": 0.19524772069955354,
      "kl": 0.3109130859375,
      "learning_rate": 4.961881546485682e-07,
      "loss": 0.0003,
      "reward": 1.7625000849366188,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11338
    },
    {
      "completion_length": 271.44644355773926,
      "epoch": 1.9012951087639884,
      "grad_norm": 0.1591140863477574,
      "kl": 0.1329803466796875,
      "learning_rate": 4.961860675765188e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.769642885774374,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11340
    },
    {
      "completion_length": 278.3035821914673,
      "epoch": 1.9016304120038559,
      "grad_norm": 0.2581112163246211,
      "kl": 0.2618408203125,
      "learning_rate": 4.961839799376576e-07,
      "loss": 0.0003,
      "reward": 1.8392857685685158,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8392857275903225,
      "rewards/format_reward_func": 1.0,
      "step": 11342
    },
    {
      "completion_length": 263.99554920196533,
      "epoch": 1.9019657152437235,
      "grad_norm": 0.18510317852809857,
      "kl": 0.1702880859375,
      "learning_rate": 4.961818917319897e-07,
      "loss": 0.0002,
      "reward": 1.7267857939004898,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7312500365078449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11344
    },
    {
      "completion_length": 265.2634048461914,
      "epoch": 1.9023010184835911,
      "grad_norm": 0.1454532985720656,
      "kl": 0.371673583984375,
      "learning_rate": 4.961798029595199e-07,
      "loss": 0.0004,
      "reward": 1.76071435213089,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143130153418,
      "rewards/format_reward_func": 1.0,
      "step": 11346
    },
    {
      "completion_length": 271.0044765472412,
      "epoch": 1.9026363217234588,
      "grad_norm": 0.11359424624294441,
      "kl": 0.22344970703125,
      "learning_rate": 4.961777136202528e-07,
      "loss": 0.0002,
      "reward": 1.7410714998841286,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7455357424914837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11348
    },
    {
      "completion_length": 269.70536708831787,
      "epoch": 1.9029716249633262,
      "grad_norm": 0.14220671747259186,
      "kl": 0.181732177734375,
      "learning_rate": 4.961756237141934e-07,
      "loss": 0.0002,
      "reward": 1.728571504354477,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7285714633762836,
      "rewards/format_reward_func": 1.0,
      "step": 11350
    },
    {
      "completion_length": 269.3705530166626,
      "epoch": 1.9033069282031938,
      "grad_norm": 0.1606879487831985,
      "kl": 0.114715576171875,
      "learning_rate": 4.961735332413465e-07,
      "loss": 0.0001,
      "reward": 1.751785770058632,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7562500443309546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11352
    },
    {
      "completion_length": 269.37054538726807,
      "epoch": 1.9036422314430612,
      "grad_norm": 0.3928058113985592,
      "kl": 0.1796112060546875,
      "learning_rate": 4.961714422017167e-07,
      "loss": 0.0002,
      "reward": 1.7678571790456772,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 11354
    },
    {
      "completion_length": 267.2500123977661,
      "epoch": 1.9039775346829289,
      "grad_norm": 0.10755558405850703,
      "kl": 0.14447021484375,
      "learning_rate": 4.96169350595309e-07,
      "loss": 0.0001,
      "reward": 1.8464286103844643,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8464285954833031,
      "rewards/format_reward_func": 1.0,
      "step": 11356
    },
    {
      "completion_length": 264.2812623977661,
      "epoch": 1.9043128379227965,
      "grad_norm": 0.28424929901275836,
      "kl": 0.12335205078125,
      "learning_rate": 4.961672584221282e-07,
      "loss": 0.0001,
      "reward": 1.8000000640749931,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.8089285865426064,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11358
    },
    {
      "completion_length": 264.68751525878906,
      "epoch": 1.9046481411626641,
      "grad_norm": 0.14374849264193768,
      "kl": 0.1146240234375,
      "learning_rate": 4.961651656821791e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 11360
    },
    {
      "completion_length": 273.4866180419922,
      "epoch": 1.9049834444025315,
      "grad_norm": 0.227732210229808,
      "kl": 0.21929931640625,
      "learning_rate": 4.961630723754666e-07,
      "loss": 0.0002,
      "reward": 1.7267857640981674,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7312500309199095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11362
    },
    {
      "completion_length": 270.5803689956665,
      "epoch": 1.905318747642399,
      "grad_norm": 0.17363189254935005,
      "kl": 0.14501953125,
      "learning_rate": 4.961609785019954e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 11364
    },
    {
      "completion_length": 271.4330472946167,
      "epoch": 1.9056540508822666,
      "grad_norm": 0.23595968121958819,
      "kl": 0.207855224609375,
      "learning_rate": 4.961588840617703e-07,
      "loss": 0.0002,
      "reward": 1.7142857983708382,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7232143171131611,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11366
    },
    {
      "completion_length": 264.21876430511475,
      "epoch": 1.9059893541221342,
      "grad_norm": 0.14596728600368417,
      "kl": 0.11456298828125,
      "learning_rate": 4.961567890547962e-07,
      "loss": 0.0001,
      "reward": 1.8017857521772385,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.8062500208616257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11368
    },
    {
      "completion_length": 270.11162090301514,
      "epoch": 1.9063246573620019,
      "grad_norm": 0.23307965611049494,
      "kl": 0.19744873046875,
      "learning_rate": 4.96154693481078e-07,
      "loss": 0.0002,
      "reward": 1.7053572162985802,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.709821468219161,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11370
    },
    {
      "completion_length": 265.9509029388428,
      "epoch": 1.9066599606018693,
      "grad_norm": 0.32637737542984435,
      "kl": 0.22430419921875,
      "learning_rate": 4.961525973406203e-07,
      "loss": 0.0002,
      "reward": 1.8000000640749931,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.8089285865426064,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11372
    },
    {
      "completion_length": 261.34376335144043,
      "epoch": 1.906995263841737,
      "grad_norm": 0.2971004762919588,
      "kl": 0.1528472900390625,
      "learning_rate": 4.961505006334281e-07,
      "loss": 0.0002,
      "reward": 1.8000000566244125,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.8089285790920258,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11374
    },
    {
      "completion_length": 266.3169775009155,
      "epoch": 1.9073305670816043,
      "grad_norm": 0.0754146216793177,
      "kl": 0.145721435546875,
      "learning_rate": 4.961484033595061e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7660714499652386,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11376
    },
    {
      "completion_length": 264.87055110931396,
      "epoch": 1.907665870321472,
      "grad_norm": 0.14408567338438105,
      "kl": 0.123016357421875,
      "learning_rate": 4.961463055188593e-07,
      "loss": 0.0001,
      "reward": 1.7500000521540642,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7500000428408384,
      "rewards/format_reward_func": 1.0,
      "step": 11378
    },
    {
      "completion_length": 260.95983505249023,
      "epoch": 1.9080011735613396,
      "grad_norm": 0.21066553363979487,
      "kl": 0.201416015625,
      "learning_rate": 4.961442071114925e-07,
      "loss": 0.0002,
      "reward": 1.796428605914116,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286040514708,
      "rewards/format_reward_func": 1.0,
      "step": 11380
    },
    {
      "completion_length": 262.84822845458984,
      "epoch": 1.9083364768012072,
      "grad_norm": 0.25315756086610974,
      "kl": 0.1490478515625,
      "learning_rate": 4.961421081374104e-07,
      "loss": 0.0001,
      "reward": 1.7857143580913544,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7946428880095482,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11382
    },
    {
      "completion_length": 263.6250114440918,
      "epoch": 1.9086717800410746,
      "grad_norm": 0.2453006777873615,
      "kl": 0.224761962890625,
      "learning_rate": 4.961400085966179e-07,
      "loss": 0.0002,
      "reward": 1.7285714894533157,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7375000324100256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11384
    },
    {
      "completion_length": 259.40179538726807,
      "epoch": 1.909007083280942,
      "grad_norm": 0.39350043991505274,
      "kl": 0.12457275390625,
      "learning_rate": 4.961379084891199e-07,
      "loss": 0.0001,
      "reward": 1.7946429252624512,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7991071790456772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11386
    },
    {
      "completion_length": 270.33930015563965,
      "epoch": 1.9093423865208097,
      "grad_norm": 0.2779319904231321,
      "kl": 0.175750732421875,
      "learning_rate": 4.961358078149211e-07,
      "loss": 0.0002,
      "reward": 1.735714353621006,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7446428835391998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11388
    },
    {
      "completion_length": 257.08036708831787,
      "epoch": 1.9096776897606773,
      "grad_norm": 0.8968730072056478,
      "kl": 0.12542724609375,
      "learning_rate": 4.961337065740263e-07,
      "loss": 0.0001,
      "reward": 1.8107143267989159,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.8196428827941418,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11390
    },
    {
      "completion_length": 258.45090770721436,
      "epoch": 1.910012993000545,
      "grad_norm": 0.273157945393228,
      "kl": 0.130096435546875,
      "learning_rate": 4.961316047664406e-07,
      "loss": 0.0001,
      "reward": 1.7589286342263222,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7723214477300644,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11392
    },
    {
      "completion_length": 263.4241180419922,
      "epoch": 1.9103482962404126,
      "grad_norm": 0.13818252928656122,
      "kl": 0.154266357421875,
      "learning_rate": 4.961295023921687e-07,
      "loss": 0.0002,
      "reward": 1.7714286297559738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 11394
    },
    {
      "completion_length": 265.8214406967163,
      "epoch": 1.91068359948028,
      "grad_norm": 0.27011759053789397,
      "kl": 0.226043701171875,
      "learning_rate": 4.961273994512154e-07,
      "loss": 0.0002,
      "reward": 1.7500000670552254,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7500000447034836,
      "rewards/format_reward_func": 1.0,
      "step": 11396
    },
    {
      "completion_length": 250.55358409881592,
      "epoch": 1.9110189027201474,
      "grad_norm": 0.6885092609424,
      "kl": 0.2010955810546875,
      "learning_rate": 4.961252959435856e-07,
      "loss": 0.0002,
      "reward": 1.7464286610484123,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7553571779280901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11398
    },
    {
      "completion_length": 268.08929443359375,
      "epoch": 1.911354205960015,
      "grad_norm": 0.2028304503555617,
      "kl": 0.15380859375,
      "learning_rate": 4.961231918692839e-07,
      "loss": 0.0002,
      "reward": 1.7964286282658577,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7964286096394062,
      "rewards/format_reward_func": 1.0,
      "step": 11400
    },
    {
      "completion_length": 268.0803680419922,
      "epoch": 1.9116895091998827,
      "grad_norm": 0.8293108782393289,
      "kl": 0.17572021484375,
      "learning_rate": 4.961210872283157e-07,
      "loss": 0.0002,
      "reward": 1.7678572237491608,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7767857313156128,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11402
    },
    {
      "completion_length": 264.0044765472412,
      "epoch": 1.9120248124397503,
      "grad_norm": 0.3747041193333945,
      "kl": 0.146636962890625,
      "learning_rate": 4.961189820206852e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7803571745753288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11404
    },
    {
      "completion_length": 278.9464387893677,
      "epoch": 1.9123601156796177,
      "grad_norm": 0.13538982335741828,
      "kl": 0.15130615234375,
      "learning_rate": 4.961168762463978e-07,
      "loss": 0.0002,
      "reward": 1.774107202887535,
      "reward_std": 0.056821079924702644,
      "rewards/equation_reward_func": 0.7803571783006191,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 11406
    },
    {
      "completion_length": 266.1651906967163,
      "epoch": 1.9126954189194854,
      "grad_norm": 0.8431514704194816,
      "kl": 0.159820556640625,
      "learning_rate": 4.961147699054579e-07,
      "loss": 0.0002,
      "reward": 1.757142923772335,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7660714611411095,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11408
    },
    {
      "completion_length": 277.15180110931396,
      "epoch": 1.9130307221593528,
      "grad_norm": 1.5579412370998411,
      "kl": 0.156585693359375,
      "learning_rate": 4.961126629978707e-07,
      "loss": 0.0002,
      "reward": 1.7482143342494965,
      "reward_std": 0.0732360603287816,
      "rewards/equation_reward_func": 0.7705357447266579,
      "rewards/format_reward_func": 0.9776785783469677,
      "step": 11410
    },
    {
      "completion_length": 285.42858505249023,
      "epoch": 1.9133660253992204,
      "grad_norm": 0.41450893660091553,
      "kl": 0.1715240478515625,
      "learning_rate": 4.961105555236408e-07,
      "loss": 0.0002,
      "reward": 1.7928572073578835,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.8017857484519482,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11412
    },
    {
      "completion_length": 274.0803689956665,
      "epoch": 1.913701328639088,
      "grad_norm": 0.24263481536803763,
      "kl": 0.14422607421875,
      "learning_rate": 4.961084474827731e-07,
      "loss": 0.0001,
      "reward": 1.7446429133415222,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7580357491970062,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11414
    },
    {
      "completion_length": 280.65179443359375,
      "epoch": 1.9140366318789557,
      "grad_norm": 1.6763596841012791,
      "kl": 2.72528076171875,
      "learning_rate": 4.961063388752726e-07,
      "loss": 0.0027,
      "reward": 1.6982143819332123,
      "reward_std": 0.11364216078072786,
      "rewards/equation_reward_func": 0.7383928745985031,
      "rewards/format_reward_func": 0.9598214477300644,
      "step": 11416
    },
    {
      "completion_length": 286.12054538726807,
      "epoch": 1.914371935118823,
      "grad_norm": 0.6866111926368569,
      "kl": 0.192596435546875,
      "learning_rate": 4.961042297011441e-07,
      "loss": 0.0002,
      "reward": 1.6571429297327995,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.6928571723401546,
      "rewards/format_reward_func": 0.9642857275903225,
      "step": 11418
    },
    {
      "completion_length": 268.3839387893677,
      "epoch": 1.9147072383586905,
      "grad_norm": 0.2820457558191292,
      "kl": 0.11151123046875,
      "learning_rate": 4.961021199603923e-07,
      "loss": 0.0001,
      "reward": 1.769642911851406,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7830357402563095,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11420
    },
    {
      "completion_length": 270.7812623977661,
      "epoch": 1.9150425415985581,
      "grad_norm": 0.13665542251726445,
      "kl": 0.1313629150390625,
      "learning_rate": 4.961000096530222e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.04545686487108469,
      "rewards/equation_reward_func": 0.7839285954833031,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11422
    },
    {
      "completion_length": 268.5669765472412,
      "epoch": 1.9153778448384258,
      "grad_norm": 0.39813810964567975,
      "kl": 0.13531494140625,
      "learning_rate": 4.960978987790386e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428809314966,
      "rewards/format_reward_func": 1.0,
      "step": 11424
    },
    {
      "completion_length": 251.665189743042,
      "epoch": 1.9157131480782934,
      "grad_norm": 0.10712173856582168,
      "kl": 0.0943603515625,
      "learning_rate": 4.960957873384465e-07,
      "loss": 0.0001,
      "reward": 1.8392857536673546,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8392857164144516,
      "rewards/format_reward_func": 1.0,
      "step": 11426
    },
    {
      "completion_length": 255.4062623977661,
      "epoch": 1.9160484513181608,
      "grad_norm": 0.14126711549786453,
      "kl": 0.1662445068359375,
      "learning_rate": 4.960936753312506e-07,
      "loss": 0.0002,
      "reward": 1.807142898440361,
      "reward_std": 0.04040610231459141,
      "rewards/equation_reward_func": 0.8160714581608772,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11428
    },
    {
      "completion_length": 248.08037185668945,
      "epoch": 1.9163837545580285,
      "grad_norm": 0.3138137135372601,
      "kl": 0.113006591796875,
      "learning_rate": 4.960915627574558e-07,
      "loss": 0.0001,
      "reward": 1.7303572297096252,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.7348214536905289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11430
    },
    {
      "completion_length": 247.3169765472412,
      "epoch": 1.9167190577978959,
      "grad_norm": 0.04930311204874395,
      "kl": 0.1185302734375,
      "learning_rate": 4.96089449617067e-07,
      "loss": 0.0001,
      "reward": 1.785714328289032,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143245637417,
      "rewards/format_reward_func": 1.0,
      "step": 11432
    },
    {
      "completion_length": 253.94197368621826,
      "epoch": 1.9170543610377635,
      "grad_norm": 0.34253288139320753,
      "kl": 0.177154541015625,
      "learning_rate": 4.96087335910089e-07,
      "loss": 0.0002,
      "reward": 1.760714329779148,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 11434
    },
    {
      "completion_length": 248.2812614440918,
      "epoch": 1.9173896642776311,
      "grad_norm": 0.15253380622826307,
      "kl": 0.120941162109375,
      "learning_rate": 4.960852216365268e-07,
      "loss": 0.0001,
      "reward": 1.7642857655882835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 11436
    },
    {
      "completion_length": 259.6741180419922,
      "epoch": 1.9177249675174988,
      "grad_norm": 0.16595620257856564,
      "kl": 0.131561279296875,
      "learning_rate": 4.960831067963851e-07,
      "loss": 0.0001,
      "reward": 1.7392857670783997,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.748214315623045,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11438
    },
    {
      "completion_length": 242.8125114440918,
      "epoch": 1.9180602707573662,
      "grad_norm": 0.13210774905736541,
      "kl": 0.104461669921875,
      "learning_rate": 4.960809913896689e-07,
      "loss": 0.0001,
      "reward": 1.7571429014205933,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 11440
    },
    {
      "completion_length": 250.49554920196533,
      "epoch": 1.9183955739972336,
      "grad_norm": 0.004442574314263273,
      "kl": 0.148345947265625,
      "learning_rate": 4.960788754163829e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7821428962051868,
      "rewards/format_reward_func": 1.0,
      "step": 11442
    },
    {
      "completion_length": 235.95090103149414,
      "epoch": 1.9187308772371012,
      "grad_norm": 0.13640183603938238,
      "kl": 0.125457763671875,
      "learning_rate": 4.960767588765322e-07,
      "loss": 0.0001,
      "reward": 1.7464286610484123,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 11444
    },
    {
      "completion_length": 242.54911613464355,
      "epoch": 1.9190661804769689,
      "grad_norm": 0.3164099307381995,
      "kl": 0.1171875,
      "learning_rate": 4.960746417701215e-07,
      "loss": 0.0001,
      "reward": 1.8607143312692642,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8607143051922321,
      "rewards/format_reward_func": 1.0,
      "step": 11446
    },
    {
      "completion_length": 258.3526906967163,
      "epoch": 1.9194014837168365,
      "grad_norm": 0.24606226713899274,
      "kl": 0.09759521484375,
      "learning_rate": 4.960725240971558e-07,
      "loss": 0.0001,
      "reward": 1.7446429133415222,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7491071633994579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11448
    },
    {
      "completion_length": 249.64733219146729,
      "epoch": 1.9197367869567041,
      "grad_norm": 0.21903366414071526,
      "kl": 0.117919921875,
      "learning_rate": 4.960704058576399e-07,
      "loss": 0.0001,
      "reward": 1.789285771548748,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857566475868,
      "rewards/format_reward_func": 1.0,
      "step": 11450
    },
    {
      "completion_length": 250.1428680419922,
      "epoch": 1.9200720901965715,
      "grad_norm": 0.11739844578988058,
      "kl": 0.15765380859375,
      "learning_rate": 4.960682870515786e-07,
      "loss": 0.0002,
      "reward": 1.700000062584877,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7000000365078449,
      "rewards/format_reward_func": 1.0,
      "step": 11452
    },
    {
      "completion_length": 252.415189743042,
      "epoch": 1.920407393436439,
      "grad_norm": 0.23504962226583487,
      "kl": 0.130767822265625,
      "learning_rate": 4.96066167678977e-07,
      "loss": 0.0001,
      "reward": 1.8035715073347092,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.8035714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 11454
    },
    {
      "completion_length": 250.7500114440918,
      "epoch": 1.9207426966763066,
      "grad_norm": 0.17987810570866028,
      "kl": 0.1189422607421875,
      "learning_rate": 4.960640477398398e-07,
      "loss": 0.0001,
      "reward": 1.735714390873909,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 11456
    },
    {
      "completion_length": 251.9062614440918,
      "epoch": 1.9210779999161742,
      "grad_norm": 0.5340453361696048,
      "kl": 0.1201019287109375,
      "learning_rate": 4.96061927234172e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.757142897695303,
      "rewards/format_reward_func": 1.0,
      "step": 11458
    },
    {
      "completion_length": 257.8660840988159,
      "epoch": 1.9214133031560419,
      "grad_norm": 0.06368641751747849,
      "kl": 0.11151123046875,
      "learning_rate": 4.960598061619782e-07,
      "loss": 0.0001,
      "reward": 1.826785758137703,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8401785865426064,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11460
    },
    {
      "completion_length": 260.2812614440918,
      "epoch": 1.9217486063959093,
      "grad_norm": 0.20438380505560297,
      "kl": 0.2633209228515625,
      "learning_rate": 4.960576845232637e-07,
      "loss": 0.0003,
      "reward": 1.7714286297559738,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 11462
    },
    {
      "completion_length": 272.7500123977661,
      "epoch": 1.922083909635777,
      "grad_norm": 0.4689894616913192,
      "kl": 0.259124755859375,
      "learning_rate": 4.960555623180331e-07,
      "loss": 0.0003,
      "reward": 1.712500087916851,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7169643212109804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11464
    },
    {
      "completion_length": 262.7098321914673,
      "epoch": 1.9224192128756443,
      "grad_norm": 0.198931744948235,
      "kl": 0.163665771484375,
      "learning_rate": 4.960534395462913e-07,
      "loss": 0.0002,
      "reward": 1.7660714909434319,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11466
    },
    {
      "completion_length": 262.0714406967163,
      "epoch": 1.922754516115512,
      "grad_norm": 0.16364250738570876,
      "kl": 0.1119537353515625,
      "learning_rate": 4.960513162080434e-07,
      "loss": 0.0001,
      "reward": 1.7392857819795609,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857540398836,
      "rewards/format_reward_func": 1.0,
      "step": 11468
    },
    {
      "completion_length": 254.85715007781982,
      "epoch": 1.9230898193553796,
      "grad_norm": 0.17039801150252207,
      "kl": 0.227020263671875,
      "learning_rate": 4.96049192303294e-07,
      "loss": 0.0002,
      "reward": 1.7178572341799736,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7178571671247482,
      "rewards/format_reward_func": 1.0,
      "step": 11470
    },
    {
      "completion_length": 248.5312614440918,
      "epoch": 1.9234251225952472,
      "grad_norm": 0.2090763004651719,
      "kl": 0.159637451171875,
      "learning_rate": 4.960470678320482e-07,
      "loss": 0.0002,
      "reward": 1.7642857655882835,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7732143104076385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11472
    },
    {
      "completion_length": 260.42412090301514,
      "epoch": 1.9237604258351146,
      "grad_norm": 0.23593082654275854,
      "kl": 0.112701416015625,
      "learning_rate": 4.960449427943108e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714752972126,
      "rewards/format_reward_func": 1.0,
      "step": 11474
    },
    {
      "completion_length": 275.6696548461914,
      "epoch": 1.924095729074982,
      "grad_norm": 0.20935209515541264,
      "kl": 0.1348876953125,
      "learning_rate": 4.960428171900868e-07,
      "loss": 0.0001,
      "reward": 1.7446429133415222,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.7580357491970062,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11476
    },
    {
      "completion_length": 262.495548248291,
      "epoch": 1.9244310323148497,
      "grad_norm": 0.21949802358359427,
      "kl": 0.119415283203125,
      "learning_rate": 4.960406910193809e-07,
      "loss": 0.0001,
      "reward": 1.7875000461935997,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7919643148779869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11478
    },
    {
      "completion_length": 269.5134038925171,
      "epoch": 1.9247663355547173,
      "grad_norm": 0.15969970131972752,
      "kl": 0.154083251953125,
      "learning_rate": 4.960385642821982e-07,
      "loss": 0.0002,
      "reward": 1.8035714775323868,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 11480
    },
    {
      "completion_length": 271.7321557998657,
      "epoch": 1.925101638794585,
      "grad_norm": 0.22003888136350389,
      "kl": 0.22259521484375,
      "learning_rate": 4.960364369785433e-07,
      "loss": 0.0002,
      "reward": 1.760714329779148,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.769642885774374,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11482
    },
    {
      "completion_length": 273.16965770721436,
      "epoch": 1.9254369420344524,
      "grad_norm": 0.4368958131528556,
      "kl": 0.192474365234375,
      "learning_rate": 4.960343091084215e-07,
      "loss": 0.0002,
      "reward": 1.7357143610715866,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7446428947150707,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11484
    },
    {
      "completion_length": 275.43750953674316,
      "epoch": 1.92577224527432,
      "grad_norm": 0.1728484515807659,
      "kl": 0.2697906494140625,
      "learning_rate": 4.960321806718375e-07,
      "loss": 0.0003,
      "reward": 1.8178571835160255,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8178571723401546,
      "rewards/format_reward_func": 1.0,
      "step": 11486
    },
    {
      "completion_length": 276.66965675354004,
      "epoch": 1.9261075485141874,
      "grad_norm": 0.5261602344874771,
      "kl": 0.2050933837890625,
      "learning_rate": 4.960300516687961e-07,
      "loss": 0.0002,
      "reward": 1.7750000581145287,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7928571626543999,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 11488
    },
    {
      "completion_length": 269.5848340988159,
      "epoch": 1.926442851754055,
      "grad_norm": 0.19601843506053898,
      "kl": 0.2016143798828125,
      "learning_rate": 4.960279220993023e-07,
      "loss": 0.0002,
      "reward": 1.7053572311997414,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7098214700818062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11490
    },
    {
      "completion_length": 270.68751335144043,
      "epoch": 1.9267781549939227,
      "grad_norm": 0.17990204744519162,
      "kl": 0.17333984375,
      "learning_rate": 4.960257919633611e-07,
      "loss": 0.0002,
      "reward": 1.7857143133878708,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7946428991854191,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11492
    },
    {
      "completion_length": 262.0982265472412,
      "epoch": 1.9271134582337903,
      "grad_norm": 0.27930705227496755,
      "kl": 0.2054595947265625,
      "learning_rate": 4.960236612609773e-07,
      "loss": 0.0002,
      "reward": 1.733928643167019,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7383928876370192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11494
    },
    {
      "completion_length": 263.8571557998657,
      "epoch": 1.9274487614736577,
      "grad_norm": 0.23885652033263852,
      "kl": 0.25677490234375,
      "learning_rate": 4.960215299921557e-07,
      "loss": 0.0003,
      "reward": 1.7821429148316383,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 11496
    },
    {
      "completion_length": 261.25447273254395,
      "epoch": 1.9277840647135251,
      "grad_norm": 0.11426829648277192,
      "kl": 0.13201904296875,
      "learning_rate": 4.960193981569014e-07,
      "loss": 0.0001,
      "reward": 1.773214340209961,
      "reward_std": 0.017677669413387775,
      "rewards/equation_reward_func": 0.7776786088943481,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11498
    },
    {
      "completion_length": 257.1428689956665,
      "epoch": 1.9281193679533928,
      "grad_norm": 0.14128657839377187,
      "kl": 0.23486328125,
      "learning_rate": 4.960172657552192e-07,
      "loss": 0.0002,
      "reward": 1.7535715028643608,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 11500
    },
    {
      "completion_length": 269.8214387893677,
      "epoch": 1.9284546711932604,
      "grad_norm": 0.04353958717156185,
      "kl": 0.460693359375,
      "learning_rate": 4.960151327871141e-07,
      "loss": 0.0005,
      "reward": 1.7785714864730835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 11502
    },
    {
      "completion_length": 252.66072750091553,
      "epoch": 1.928789974433128,
      "grad_norm": 0.15968991693381757,
      "kl": 0.16265869140625,
      "learning_rate": 4.960129992525909e-07,
      "loss": 0.0002,
      "reward": 1.76607146859169,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7705357577651739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11504
    },
    {
      "completion_length": 263.5044775009155,
      "epoch": 1.9291252776729955,
      "grad_norm": 0.2532534123405873,
      "kl": 0.210540771484375,
      "learning_rate": 4.960108651516545e-07,
      "loss": 0.0002,
      "reward": 1.7571429535746574,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 11506
    },
    {
      "completion_length": 267.4821548461914,
      "epoch": 1.929460580912863,
      "grad_norm": 0.14032272575220447,
      "kl": 0.331695556640625,
      "learning_rate": 4.960087304843099e-07,
      "loss": 0.0003,
      "reward": 1.8178571835160255,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8178571611642838,
      "rewards/format_reward_func": 1.0,
      "step": 11508
    },
    {
      "completion_length": 254.3035831451416,
      "epoch": 1.9297958841527305,
      "grad_norm": 0.14579727584127364,
      "kl": 0.116943359375,
      "learning_rate": 4.96006595250562e-07,
      "loss": 0.0001,
      "reward": 1.8250000402331352,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.825000025331974,
      "rewards/format_reward_func": 1.0,
      "step": 11510
    },
    {
      "completion_length": 260.56251335144043,
      "epoch": 1.9301311873925981,
      "grad_norm": 0.2295536831446175,
      "kl": 0.098663330078125,
      "learning_rate": 4.960044594504158e-07,
      "loss": 0.0001,
      "reward": 1.79464291036129,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7991071678698063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11512
    },
    {
      "completion_length": 272.0000104904175,
      "epoch": 1.9304664906324658,
      "grad_norm": 0.1013315839791849,
      "kl": 0.131591796875,
      "learning_rate": 4.96002323083876e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 11514
    },
    {
      "completion_length": 267.3169775009155,
      "epoch": 1.9308017938723334,
      "grad_norm": 0.2789793491517081,
      "kl": 0.112640380859375,
      "learning_rate": 4.960001861509477e-07,
      "loss": 0.0001,
      "reward": 1.744642935693264,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7491071708500385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11516
    },
    {
      "completion_length": 262.5625114440918,
      "epoch": 1.9311370971122008,
      "grad_norm": 0.23336605649958678,
      "kl": 0.1778564453125,
      "learning_rate": 4.959980486516358e-07,
      "loss": 0.0002,
      "reward": 1.725000061094761,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7250000387430191,
      "rewards/format_reward_func": 1.0,
      "step": 11518
    },
    {
      "completion_length": 275.1562662124634,
      "epoch": 1.9314724003520682,
      "grad_norm": 0.19566252414623606,
      "kl": 0.120880126953125,
      "learning_rate": 4.959959105859451e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714656114578,
      "rewards/format_reward_func": 1.0,
      "step": 11520
    },
    {
      "completion_length": 273.3035821914673,
      "epoch": 1.9318077035919359,
      "grad_norm": 0.26468819573733504,
      "kl": 0.131500244140625,
      "learning_rate": 4.959937719538806e-07,
      "loss": 0.0001,
      "reward": 1.728571504354477,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7375000342726707,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11522
    },
    {
      "completion_length": 278.30804920196533,
      "epoch": 1.9321430068318035,
      "grad_norm": 0.16803667829067487,
      "kl": 0.1395416259765625,
      "learning_rate": 4.959916327554473e-07,
      "loss": 0.0001,
      "reward": 1.7107143476605415,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7107143253087997,
      "rewards/format_reward_func": 1.0,
      "step": 11524
    },
    {
      "completion_length": 282.00000953674316,
      "epoch": 1.9324783100716711,
      "grad_norm": 0.436746719978823,
      "kl": 0.212310791015625,
      "learning_rate": 4.959894929906499e-07,
      "loss": 0.0002,
      "reward": 1.7464286535978317,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 11526
    },
    {
      "completion_length": 274.15180110931396,
      "epoch": 1.9328136133115388,
      "grad_norm": 0.15089911493730374,
      "kl": 0.23822021484375,
      "learning_rate": 4.959873526594937e-07,
      "loss": 0.0002,
      "reward": 1.7196429148316383,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7241071872413158,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11528
    },
    {
      "completion_length": 268.16519355773926,
      "epoch": 1.9331489165514062,
      "grad_norm": 0.24695749023732816,
      "kl": 0.290924072265625,
      "learning_rate": 4.959852117619834e-07,
      "loss": 0.0003,
      "reward": 1.7410714849829674,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7455357518047094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11530
    },
    {
      "completion_length": 267.61608505249023,
      "epoch": 1.9334842197912736,
      "grad_norm": 0.15763417332067145,
      "kl": 0.22369384765625,
      "learning_rate": 4.959830702981237e-07,
      "loss": 0.0002,
      "reward": 1.7392857894301414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7392857298254967,
      "rewards/format_reward_func": 1.0,
      "step": 11532
    },
    {
      "completion_length": 281.9687623977661,
      "epoch": 1.9338195230311412,
      "grad_norm": 0.22574997557198315,
      "kl": 0.16107177734375,
      "learning_rate": 4.9598092826792e-07,
      "loss": 0.0002,
      "reward": 1.737500086426735,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7419643141329288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11534
    },
    {
      "completion_length": 272.75447845458984,
      "epoch": 1.9341548262710089,
      "grad_norm": 0.20676245907015658,
      "kl": 0.257080078125,
      "learning_rate": 4.95978785671377e-07,
      "loss": 0.0003,
      "reward": 1.7642858028411865,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 11536
    },
    {
      "completion_length": 273.80358695983887,
      "epoch": 1.9344901295108765,
      "grad_norm": 0.11482451783793798,
      "kl": 0.14068603515625,
      "learning_rate": 4.959766425084995e-07,
      "loss": 0.0001,
      "reward": 1.776785746216774,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7901786006987095,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11538
    },
    {
      "completion_length": 283.5178699493408,
      "epoch": 1.934825432750744,
      "grad_norm": 0.42962985392694314,
      "kl": 0.11175537109375,
      "learning_rate": 4.959744987792926e-07,
      "loss": 0.0001,
      "reward": 1.816071480512619,
      "reward_std": 0.07828682009130716,
      "rewards/equation_reward_func": 0.8205357491970062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11540
    },
    {
      "completion_length": 262.7500114440918,
      "epoch": 1.9351607359906116,
      "grad_norm": 0.26195593090194097,
      "kl": 0.10430908203125,
      "learning_rate": 4.959723544837611e-07,
      "loss": 0.0001,
      "reward": 1.8785714581608772,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8785714469850063,
      "rewards/format_reward_func": 1.0,
      "step": 11542
    },
    {
      "completion_length": 265.85269260406494,
      "epoch": 1.935496039230479,
      "grad_norm": 0.22136712208141024,
      "kl": 0.103271484375,
      "learning_rate": 4.959702096219103e-07,
      "loss": 0.0001,
      "reward": 1.8178571835160255,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.8267857432365417,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11544
    },
    {
      "completion_length": 276.78572845458984,
      "epoch": 1.9358313424703466,
      "grad_norm": 0.3523477652786812,
      "kl": 0.125274658203125,
      "learning_rate": 4.959680641937447e-07,
      "loss": 0.0001,
      "reward": 1.7428572326898575,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 11546
    },
    {
      "completion_length": 276.51340675354004,
      "epoch": 1.9361666457102142,
      "grad_norm": 0.12306280651202639,
      "kl": 0.116790771484375,
      "learning_rate": 4.959659181992695e-07,
      "loss": 0.0001,
      "reward": 1.7892857566475868,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7982143126428127,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11548
    },
    {
      "completion_length": 268.9509048461914,
      "epoch": 1.9365019489500819,
      "grad_norm": 0.1913487796137698,
      "kl": 0.123565673828125,
      "learning_rate": 4.959637716384895e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 11550
    },
    {
      "completion_length": 271.3392972946167,
      "epoch": 1.9368372521899493,
      "grad_norm": 0.14629526246249022,
      "kl": 0.1084747314453125,
      "learning_rate": 4.959616245114097e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 11552
    },
    {
      "completion_length": 279.4196548461914,
      "epoch": 1.9371725554298167,
      "grad_norm": 0.09694533133624259,
      "kl": 0.129730224609375,
      "learning_rate": 4.95959476818035e-07,
      "loss": 0.0001,
      "reward": 1.7660714760422707,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7705357521772385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11554
    },
    {
      "completion_length": 271.45983123779297,
      "epoch": 1.9375078586696843,
      "grad_norm": 0.2217887064929058,
      "kl": 0.11297607421875,
      "learning_rate": 4.959573285583706e-07,
      "loss": 0.0001,
      "reward": 1.8071429133415222,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8071428686380386,
      "rewards/format_reward_func": 1.0,
      "step": 11556
    },
    {
      "completion_length": 272.4732275009155,
      "epoch": 1.937843161909552,
      "grad_norm": 0.23558904806238534,
      "kl": 0.132354736328125,
      "learning_rate": 4.959551797324211e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964286059141159,
      "rewards/format_reward_func": 1.0,
      "step": 11558
    },
    {
      "completion_length": 266.68305110931396,
      "epoch": 1.9381784651494196,
      "grad_norm": 0.22151610735532148,
      "kl": 0.1597137451171875,
      "learning_rate": 4.959530303401915e-07,
      "loss": 0.0002,
      "reward": 1.8169643357396126,
      "reward_std": 0.05682107899338007,
      "rewards/equation_reward_func": 0.8232143074274063,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 11560
    },
    {
      "completion_length": 267.44644355773926,
      "epoch": 1.938513768389287,
      "grad_norm": 0.1748234891363921,
      "kl": 0.1086578369140625,
      "learning_rate": 4.959508803816868e-07,
      "loss": 0.0001,
      "reward": 1.7589286342263222,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7633928842842579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11562
    },
    {
      "completion_length": 284.33930015563965,
      "epoch": 1.9388490716291547,
      "grad_norm": 0.189758110151058,
      "kl": 0.1158447265625,
      "learning_rate": 4.959487298569121e-07,
      "loss": 0.0001,
      "reward": 1.7875000536441803,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.8008928745985031,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11564
    },
    {
      "completion_length": 273.1964416503906,
      "epoch": 1.939184374869022,
      "grad_norm": 0.22127814925587067,
      "kl": 0.1219329833984375,
      "learning_rate": 4.959465787658723e-07,
      "loss": 0.0001,
      "reward": 1.7464286237955093,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7553571723401546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11566
    },
    {
      "completion_length": 270.5446538925171,
      "epoch": 1.9395196781088897,
      "grad_norm": 0.2793978109461439,
      "kl": 0.2030029296875,
      "learning_rate": 4.959444271085723e-07,
      "loss": 0.0002,
      "reward": 1.7214286029338837,
      "reward_std": 0.0656599160283804,
      "rewards/equation_reward_func": 0.7482143081724644,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 11568
    },
    {
      "completion_length": 273.8259086608887,
      "epoch": 1.9398549813487573,
      "grad_norm": 0.550426955461273,
      "kl": 0.2730712890625,
      "learning_rate": 4.959422748850168e-07,
      "loss": 0.0003,
      "reward": 1.7321429327130318,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7410714421421289,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11570
    },
    {
      "completion_length": 277.6205472946167,
      "epoch": 1.940190284588625,
      "grad_norm": 0.19508560241066944,
      "kl": 0.1059417724609375,
      "learning_rate": 4.959401220952112e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7910714522004128,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11572
    },
    {
      "completion_length": 277.46429443359375,
      "epoch": 1.9405255878284924,
      "grad_norm": 0.44544147580496724,
      "kl": 0.1553955078125,
      "learning_rate": 4.959379687391602e-07,
      "loss": 0.0002,
      "reward": 1.803571492433548,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 11574
    },
    {
      "completion_length": 272.8750123977661,
      "epoch": 1.9408608910683598,
      "grad_norm": 0.20959783360587844,
      "kl": 0.17010498046875,
      "learning_rate": 4.959358148168687e-07,
      "loss": 0.0002,
      "reward": 1.7678572088479996,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7767857536673546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11576
    },
    {
      "completion_length": 267.2009057998657,
      "epoch": 1.9411961943082274,
      "grad_norm": 0.23572829651922414,
      "kl": 0.221099853515625,
      "learning_rate": 4.95933660328342e-07,
      "loss": 0.0002,
      "reward": 1.8000000715255737,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.8089285977184772,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11578
    },
    {
      "completion_length": 276.2991180419922,
      "epoch": 1.941531497548095,
      "grad_norm": 0.1705477545589577,
      "kl": 0.1273193359375,
      "learning_rate": 4.959315052735846e-07,
      "loss": 0.0001,
      "reward": 1.7767858058214188,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7812500409781933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11580
    },
    {
      "completion_length": 261.55358600616455,
      "epoch": 1.9418668007879627,
      "grad_norm": 0.4312032446580683,
      "kl": 0.151031494140625,
      "learning_rate": 4.959293496526018e-07,
      "loss": 0.0002,
      "reward": 1.7875000536441803,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7919643111526966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11582
    },
    {
      "completion_length": 269.62054443359375,
      "epoch": 1.9422021040278303,
      "grad_norm": 0.17382506989411386,
      "kl": 0.1229248046875,
      "learning_rate": 4.959271934653985e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 11584
    },
    {
      "completion_length": 271.15179538726807,
      "epoch": 1.9425374072676977,
      "grad_norm": 0.22970344709163384,
      "kl": 0.3343658447265625,
      "learning_rate": 4.959250367119795e-07,
      "loss": 0.0003,
      "reward": 1.73214291036129,
      "reward_std": 0.10606601648032665,
      "rewards/equation_reward_func": 0.7500000204890966,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 11586
    },
    {
      "completion_length": 277.8303680419922,
      "epoch": 1.9428727105075652,
      "grad_norm": 0.24053185195149798,
      "kl": 0.14495849609375,
      "learning_rate": 4.9592287939235e-07,
      "loss": 0.0001,
      "reward": 1.7357143387198448,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7446428835391998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11588
    },
    {
      "completion_length": 265.2723340988159,
      "epoch": 1.9432080137474328,
      "grad_norm": 0.2155865915403086,
      "kl": 0.13531494140625,
      "learning_rate": 4.959207215065148e-07,
      "loss": 0.0001,
      "reward": 1.7232143729925156,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7276785895228386,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11590
    },
    {
      "completion_length": 263.0491199493408,
      "epoch": 1.9435433169873004,
      "grad_norm": 0.19655043391107452,
      "kl": 0.6219482421875,
      "learning_rate": 4.959185630544788e-07,
      "loss": 0.0006,
      "reward": 1.760714367032051,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 11592
    },
    {
      "completion_length": 264.5535831451416,
      "epoch": 1.943878620227168,
      "grad_norm": 0.32884497758459014,
      "kl": 0.231201171875,
      "learning_rate": 4.959164040362473e-07,
      "loss": 0.0002,
      "reward": 1.7375000789761543,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7419643327593803,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11594
    },
    {
      "completion_length": 263.308048248291,
      "epoch": 1.9442139234670355,
      "grad_norm": 0.12803094395008807,
      "kl": 0.25775146484375,
      "learning_rate": 4.959142444518249e-07,
      "loss": 0.0003,
      "reward": 1.7196429371833801,
      "reward_std": 0.06313453428447247,
      "rewards/equation_reward_func": 0.7330357320606709,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11596
    },
    {
      "completion_length": 263.7901906967163,
      "epoch": 1.944549226706903,
      "grad_norm": 0.2162849149079578,
      "kl": 0.13641357421875,
      "learning_rate": 4.959120843012168e-07,
      "loss": 0.0001,
      "reward": 1.8196428939700127,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8241071589291096,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11598
    },
    {
      "completion_length": 253.55804538726807,
      "epoch": 1.9448845299467705,
      "grad_norm": 0.27344404414924084,
      "kl": 0.5733642578125,
      "learning_rate": 4.959099235844278e-07,
      "loss": 0.0006,
      "reward": 1.7982143461704254,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.8026785999536514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11600
    },
    {
      "completion_length": 255.9464406967163,
      "epoch": 1.9452198331866382,
      "grad_norm": 0.19967799496439764,
      "kl": 0.13580322265625,
      "learning_rate": 4.959077623014631e-07,
      "loss": 0.0001,
      "reward": 1.742857202887535,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 11602
    },
    {
      "completion_length": 255.94197750091553,
      "epoch": 1.9455551364265058,
      "grad_norm": 0.21747384010928997,
      "kl": 0.6826171875,
      "learning_rate": 4.959056004523275e-07,
      "loss": 0.0007,
      "reward": 1.7642857506871223,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7732143178582191,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11604
    },
    {
      "completion_length": 252.40179920196533,
      "epoch": 1.9458904396663734,
      "grad_norm": 0.13116440361104098,
      "kl": 0.253143310546875,
      "learning_rate": 4.959034380370261e-07,
      "loss": 0.0003,
      "reward": 1.7142857983708382,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7142857424914837,
      "rewards/format_reward_func": 1.0,
      "step": 11606
    },
    {
      "completion_length": 258.94643783569336,
      "epoch": 1.9462257429062408,
      "grad_norm": 0.17006587057823047,
      "kl": 0.4481964111328125,
      "learning_rate": 4.959012750555638e-07,
      "loss": 0.0004,
      "reward": 1.7553572058677673,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7598214708268642,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11608
    },
    {
      "completion_length": 251.8482255935669,
      "epoch": 1.9465610461461083,
      "grad_norm": 0.17390186294610058,
      "kl": 0.245513916015625,
      "learning_rate": 4.958991115079455e-07,
      "loss": 0.0002,
      "reward": 1.7785714864730835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 11610
    },
    {
      "completion_length": 252.4509048461914,
      "epoch": 1.9468963493859759,
      "grad_norm": 0.13355463954568156,
      "kl": 0.18939208984375,
      "learning_rate": 4.958969473941763e-07,
      "loss": 0.0002,
      "reward": 1.6750000789761543,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.6750000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 11612
    },
    {
      "completion_length": 263.87055015563965,
      "epoch": 1.9472316526258435,
      "grad_norm": 0.30431565753795486,
      "kl": 0.47369384765625,
      "learning_rate": 4.958947827142612e-07,
      "loss": 0.0005,
      "reward": 1.7303571999073029,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7348214630037546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11614
    },
    {
      "completion_length": 260.1562614440918,
      "epoch": 1.9475669558657112,
      "grad_norm": 0.2853115135319929,
      "kl": 0.405059814453125,
      "learning_rate": 4.958926174682052e-07,
      "loss": 0.0004,
      "reward": 1.7303572222590446,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.743750024586916,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11616
    },
    {
      "completion_length": 252.3392972946167,
      "epoch": 1.9479022591055786,
      "grad_norm": 0.15608169259880164,
      "kl": 0.25421142578125,
      "learning_rate": 4.958904516560132e-07,
      "loss": 0.0003,
      "reward": 1.771428607404232,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286148548126,
      "rewards/format_reward_func": 1.0,
      "step": 11618
    },
    {
      "completion_length": 247.1339406967163,
      "epoch": 1.9482375623454462,
      "grad_norm": 0.17523372472153195,
      "kl": 0.204986572265625,
      "learning_rate": 4.958882852776901e-07,
      "loss": 0.0002,
      "reward": 1.7196429297327995,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7241071816533804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11620
    },
    {
      "completion_length": 241.5134048461914,
      "epoch": 1.9485728655853136,
      "grad_norm": 0.2508027085559755,
      "kl": 0.18426513671875,
      "learning_rate": 4.958861183332411e-07,
      "loss": 0.0002,
      "reward": 1.7321429252624512,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428824216127,
      "rewards/format_reward_func": 1.0,
      "step": 11622
    },
    {
      "completion_length": 240.7187614440918,
      "epoch": 1.9489081688251813,
      "grad_norm": 0.19603250128713348,
      "kl": 0.172637939453125,
      "learning_rate": 4.95883950822671e-07,
      "loss": 0.0002,
      "reward": 1.8178571835160255,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8178571630269289,
      "rewards/format_reward_func": 1.0,
      "step": 11624
    },
    {
      "completion_length": 245.19197368621826,
      "epoch": 1.9492434720650489,
      "grad_norm": 0.32498217250259853,
      "kl": 0.6026763916015625,
      "learning_rate": 4.95881782745985e-07,
      "loss": 0.0006,
      "reward": 1.7982143610715866,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8026785850524902,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11626
    },
    {
      "completion_length": 244.44643878936768,
      "epoch": 1.9495787753049165,
      "grad_norm": 0.08157777296903064,
      "kl": 0.193695068359375,
      "learning_rate": 4.958796141031878e-07,
      "loss": 0.0002,
      "reward": 1.775000087916851,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 11628
    },
    {
      "completion_length": 246.67858409881592,
      "epoch": 1.949914078544784,
      "grad_norm": 0.24126868290716924,
      "kl": 0.1125946044921875,
      "learning_rate": 4.958774448942847e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 11630
    },
    {
      "completion_length": 250.96429634094238,
      "epoch": 1.9502493817846513,
      "grad_norm": 0.058257893302917085,
      "kl": 0.1483154296875,
      "learning_rate": 4.958752751192805e-07,
      "loss": 0.0001,
      "reward": 1.7982143461704254,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8026785999536514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11632
    },
    {
      "completion_length": 262.72768783569336,
      "epoch": 1.950584685024519,
      "grad_norm": 0.19758306784761007,
      "kl": 3.353057861328125,
      "learning_rate": 4.958731047781803e-07,
      "loss": 0.0034,
      "reward": 1.7053572162985802,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.718750037252903,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11634
    },
    {
      "completion_length": 250.64287185668945,
      "epoch": 1.9509199882643866,
      "grad_norm": 0.26712866657750123,
      "kl": 0.165252685546875,
      "learning_rate": 4.958709338709889e-07,
      "loss": 0.0002,
      "reward": 1.730357214808464,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7348214685916901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11636
    },
    {
      "completion_length": 261.1651916503906,
      "epoch": 1.9512552915042543,
      "grad_norm": 0.19153741964710466,
      "kl": 1.4565277099609375,
      "learning_rate": 4.958687623977117e-07,
      "loss": 0.0015,
      "reward": 1.767857201397419,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.776785746216774,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11638
    },
    {
      "completion_length": 258.1116199493408,
      "epoch": 1.9515905947441217,
      "grad_norm": 0.1725781835403997,
      "kl": 0.469635009765625,
      "learning_rate": 4.958665903583533e-07,
      "loss": 0.0005,
      "reward": 1.8053571954369545,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8098214603960514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11640
    },
    {
      "completion_length": 266.96876335144043,
      "epoch": 1.9519258979839893,
      "grad_norm": 0.3686855380317879,
      "kl": 0.53802490234375,
      "learning_rate": 4.958644177529189e-07,
      "loss": 0.0005,
      "reward": 1.7535715028643608,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 11642
    },
    {
      "completion_length": 244.67858123779297,
      "epoch": 1.9522612012238567,
      "grad_norm": 0.08202584755809787,
      "kl": 0.1743621826171875,
      "learning_rate": 4.958622445814133e-07,
      "loss": 0.0002,
      "reward": 1.8017857670783997,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.8062500320374966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11644
    },
    {
      "completion_length": 253.4553680419922,
      "epoch": 1.9525965044637243,
      "grad_norm": 0.28201809123002675,
      "kl": 0.7519073486328125,
      "learning_rate": 4.958600708438418e-07,
      "loss": 0.0008,
      "reward": 1.7910714894533157,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.8044643066823483,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11646
    },
    {
      "completion_length": 258.1875104904175,
      "epoch": 1.952931807703592,
      "grad_norm": 0.45200193199739935,
      "kl": 1.96221923828125,
      "learning_rate": 4.958578965402093e-07,
      "loss": 0.002,
      "reward": 1.742857187986374,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.7607143074274063,
      "rewards/format_reward_func": 0.9821428619325161,
      "step": 11648
    },
    {
      "completion_length": 256.8973321914673,
      "epoch": 1.9532671109434596,
      "grad_norm": 0.22485174876966724,
      "kl": 1.655181884765625,
      "learning_rate": 4.958557216705207e-07,
      "loss": 0.0017,
      "reward": 1.7232143580913544,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7366071715950966,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11650
    },
    {
      "completion_length": 252.80358409881592,
      "epoch": 1.953602414183327,
      "grad_norm": 0.3054215109505365,
      "kl": 0.42816162109375,
      "learning_rate": 4.95853546234781e-07,
      "loss": 0.0004,
      "reward": 1.7571429014205933,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7660714574158192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11652
    },
    {
      "completion_length": 279.90180110931396,
      "epoch": 1.9539377174231944,
      "grad_norm": 0.13204873486477667,
      "kl": 0.4475860595703125,
      "learning_rate": 4.958513702329953e-07,
      "loss": 0.0004,
      "reward": 1.7357143312692642,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7535714469850063,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 11654
    },
    {
      "completion_length": 260.99555015563965,
      "epoch": 1.954273020663062,
      "grad_norm": 0.2930258195465054,
      "kl": 0.1341094970703125,
      "learning_rate": 4.958491936651687e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.09091372694820166,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 11656
    },
    {
      "completion_length": 250.63840293884277,
      "epoch": 1.9546083239029297,
      "grad_norm": 0.1968308325811053,
      "kl": 0.443206787109375,
      "learning_rate": 4.95847016531306e-07,
      "loss": 0.0004,
      "reward": 1.7857143506407738,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7946428842842579,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11658
    },
    {
      "completion_length": 254.2053689956665,
      "epoch": 1.9549436271427973,
      "grad_norm": 0.21657157782427686,
      "kl": 0.255615234375,
      "learning_rate": 4.958448388314124e-07,
      "loss": 0.0003,
      "reward": 1.725000075995922,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7428571619093418,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 11660
    },
    {
      "completion_length": 259.3526906967163,
      "epoch": 1.955278930382665,
      "grad_norm": 0.1188084083595016,
      "kl": 0.2137908935546875,
      "learning_rate": 4.958426605654928e-07,
      "loss": 0.0002,
      "reward": 1.8000000342726707,
      "reward_std": 0.05050762835890055,
      "rewards/equation_reward_func": 0.808928607031703,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11662
    },
    {
      "completion_length": 250.0803680419922,
      "epoch": 1.9556142336225324,
      "grad_norm": 0.45858462645071885,
      "kl": 0.361175537109375,
      "learning_rate": 4.958404817335522e-07,
      "loss": 0.0004,
      "reward": 1.7071429342031479,
      "reward_std": 0.0909137288108468,
      "rewards/equation_reward_func": 0.7339286170899868,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 11664
    },
    {
      "completion_length": 238.53126049041748,
      "epoch": 1.9559495368623998,
      "grad_norm": 0.16654047526987906,
      "kl": 0.10797119140625,
      "learning_rate": 4.958383023355957e-07,
      "loss": 0.0001,
      "reward": 1.7553572058677673,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.768750037997961,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11666
    },
    {
      "completion_length": 247.0223331451416,
      "epoch": 1.9562848401022674,
      "grad_norm": 0.21473295137029416,
      "kl": 0.17388916015625,
      "learning_rate": 4.958361223716282e-07,
      "loss": 0.0002,
      "reward": 1.728571504354477,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7285714615136385,
      "rewards/format_reward_func": 1.0,
      "step": 11668
    },
    {
      "completion_length": 236.5491180419922,
      "epoch": 1.956620143342135,
      "grad_norm": 0.1944749028128767,
      "kl": 0.14154052734375,
      "learning_rate": 4.95833941841655e-07,
      "loss": 0.0001,
      "reward": 1.744642935693264,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7491071708500385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11670
    },
    {
      "completion_length": 237.00893878936768,
      "epoch": 1.9569554465820027,
      "grad_norm": 0.07876787095959105,
      "kl": 0.208038330078125,
      "learning_rate": 4.958317607456807e-07,
      "loss": 0.0002,
      "reward": 1.7517857775092125,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7562500275671482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11672
    },
    {
      "completion_length": 245.40625953674316,
      "epoch": 1.9572907498218701,
      "grad_norm": 0.11458006446040274,
      "kl": 0.266632080078125,
      "learning_rate": 4.958295790837106e-07,
      "loss": 0.0003,
      "reward": 1.719642922282219,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7330357432365417,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11674
    },
    {
      "completion_length": 238.54018783569336,
      "epoch": 1.9576260530617378,
      "grad_norm": 0.3717038714285689,
      "kl": 0.412078857421875,
      "learning_rate": 4.958273968557497e-07,
      "loss": 0.0004,
      "reward": 1.726785771548748,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.740178607404232,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11676
    },
    {
      "completion_length": 231.80358028411865,
      "epoch": 1.9579613563016052,
      "grad_norm": 0.20329722813535972,
      "kl": 0.27691650390625,
      "learning_rate": 4.95825214061803e-07,
      "loss": 0.0003,
      "reward": 1.746428668498993,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 11678
    },
    {
      "completion_length": 233.9241189956665,
      "epoch": 1.9582966595414728,
      "grad_norm": 0.12961171035557645,
      "kl": 0.254791259765625,
      "learning_rate": 4.958230307018755e-07,
      "loss": 0.0003,
      "reward": 1.7196429371833801,
      "reward_std": 0.0530330091714859,
      "rewards/equation_reward_func": 0.7330357432365417,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11680
    },
    {
      "completion_length": 231.82143878936768,
      "epoch": 1.9586319627813404,
      "grad_norm": 0.2286560781940869,
      "kl": 0.150177001953125,
      "learning_rate": 4.958208467759722e-07,
      "loss": 0.0002,
      "reward": 1.6982143595814705,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.702678607776761,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11682
    },
    {
      "completion_length": 234.32590293884277,
      "epoch": 1.958967266021208,
      "grad_norm": 0.1759078841403251,
      "kl": 0.170379638671875,
      "learning_rate": 4.958186622840982e-07,
      "loss": 0.0002,
      "reward": 1.7500000670552254,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000204890966,
      "rewards/format_reward_func": 1.0,
      "step": 11684
    },
    {
      "completion_length": 228.33483505249023,
      "epoch": 1.9593025692610755,
      "grad_norm": 0.25031878839867766,
      "kl": 0.27630615234375,
      "learning_rate": 4.958164772262584e-07,
      "loss": 0.0003,
      "reward": 1.7571429312229156,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428846567869,
      "rewards/format_reward_func": 1.0,
      "step": 11686
    },
    {
      "completion_length": 221.42411613464355,
      "epoch": 1.959637872500943,
      "grad_norm": 0.2143225747696588,
      "kl": 0.216766357421875,
      "learning_rate": 4.95814291602458e-07,
      "loss": 0.0002,
      "reward": 1.832142911851406,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8321428894996643,
      "rewards/format_reward_func": 1.0,
      "step": 11688
    },
    {
      "completion_length": 223.12500953674316,
      "epoch": 1.9599731757408105,
      "grad_norm": 0.11656490838104483,
      "kl": 0.2996826171875,
      "learning_rate": 4.95812105412702e-07,
      "loss": 0.0003,
      "reward": 1.7857143431901932,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143059372902,
      "rewards/format_reward_func": 1.0,
      "step": 11690
    },
    {
      "completion_length": 222.45090579986572,
      "epoch": 1.9603084789806782,
      "grad_norm": 0.20553337543261221,
      "kl": 0.117279052734375,
      "learning_rate": 4.958099186569953e-07,
      "loss": 0.0001,
      "reward": 1.7303571850061417,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.734821455553174,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11692
    },
    {
      "completion_length": 223.79465293884277,
      "epoch": 1.9606437822205458,
      "grad_norm": 0.09753404057868746,
      "kl": 0.360626220703125,
      "learning_rate": 4.958077313353432e-07,
      "loss": 0.0004,
      "reward": 1.7500000447034836,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 11694
    },
    {
      "completion_length": 216.57590293884277,
      "epoch": 1.9609790854604132,
      "grad_norm": 0.159109331582923,
      "kl": 0.179901123046875,
      "learning_rate": 4.958055434477504e-07,
      "loss": 0.0002,
      "reward": 1.796428620815277,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 11696
    },
    {
      "completion_length": 231.446439743042,
      "epoch": 1.9613143887002809,
      "grad_norm": 0.2844154985682583,
      "kl": 0.23492431640625,
      "learning_rate": 4.958033549942222e-07,
      "loss": 0.0002,
      "reward": 1.825000062584877,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8250000141561031,
      "rewards/format_reward_func": 1.0,
      "step": 11698
    },
    {
      "completion_length": 219.102689743042,
      "epoch": 1.9616496919401483,
      "grad_norm": 0.28877537376523416,
      "kl": 0.134918212890625,
      "learning_rate": 4.958011659747635e-07,
      "loss": 0.0001,
      "reward": 1.7142857909202576,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.714285746216774,
      "rewards/format_reward_func": 1.0,
      "step": 11700
    },
    {
      "completion_length": 226.44643783569336,
      "epoch": 1.961984995180016,
      "grad_norm": 0.42817238326910484,
      "kl": 0.4429931640625,
      "learning_rate": 4.957989763893793e-07,
      "loss": 0.0004,
      "reward": 1.7750000581145287,
      "reward_std": 0.045456862077116966,
      "rewards/equation_reward_func": 0.7750000152736902,
      "rewards/format_reward_func": 1.0,
      "step": 11702
    },
    {
      "completion_length": 220.68304538726807,
      "epoch": 1.9623202984198835,
      "grad_norm": 0.09978868725868749,
      "kl": 0.122467041015625,
      "learning_rate": 4.957967862380749e-07,
      "loss": 0.0001,
      "reward": 1.739285796880722,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 1.0,
      "step": 11704
    },
    {
      "completion_length": 226.65625762939453,
      "epoch": 1.9626556016597512,
      "grad_norm": 0.11651728537059637,
      "kl": 0.1053619384765625,
      "learning_rate": 4.95794595520855e-07,
      "loss": 0.0001,
      "reward": 1.800000049173832,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 11706
    },
    {
      "completion_length": 224.50893783569336,
      "epoch": 1.9629909048996186,
      "grad_norm": 0.2761169766878691,
      "kl": 0.214324951171875,
      "learning_rate": 4.957924042377248e-07,
      "loss": 0.0002,
      "reward": 1.753571480512619,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 11708
    },
    {
      "completion_length": 225.88393592834473,
      "epoch": 1.963326208139486,
      "grad_norm": 0.19632426609063403,
      "kl": 0.237152099609375,
      "learning_rate": 4.957902123886895e-07,
      "loss": 0.0002,
      "reward": 1.7857143506407738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 11710
    },
    {
      "completion_length": 213.43304634094238,
      "epoch": 1.9636615113793536,
      "grad_norm": 0.18476990547780098,
      "kl": 0.12530517578125,
      "learning_rate": 4.957880199737539e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714622586966,
      "rewards/format_reward_func": 1.0,
      "step": 11712
    },
    {
      "completion_length": 220.67858123779297,
      "epoch": 1.9639968146192213,
      "grad_norm": 0.20954613469586167,
      "kl": 0.3189544677734375,
      "learning_rate": 4.957858269929232e-07,
      "loss": 0.0003,
      "reward": 1.8035714849829674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 11714
    },
    {
      "completion_length": 230.14733123779297,
      "epoch": 1.964332117859089,
      "grad_norm": 0.18032852907919253,
      "kl": 0.217559814453125,
      "learning_rate": 4.957836334462024e-07,
      "loss": 0.0002,
      "reward": 1.7500000596046448,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000391155481,
      "rewards/format_reward_func": 1.0,
      "step": 11716
    },
    {
      "completion_length": 247.68304920196533,
      "epoch": 1.9646674210989565,
      "grad_norm": 0.2596767575265767,
      "kl": 0.3243408203125,
      "learning_rate": 4.957814393335964e-07,
      "loss": 0.0003,
      "reward": 1.7017857879400253,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.706250037997961,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11718
    },
    {
      "completion_length": 223.7321538925171,
      "epoch": 1.965002724338824,
      "grad_norm": 0.17295896277556655,
      "kl": 0.103607177734375,
      "learning_rate": 4.957792446551105e-07,
      "loss": 0.0001,
      "reward": 1.787500061094761,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7919643186032772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11720
    },
    {
      "completion_length": 244.13393878936768,
      "epoch": 1.9653380275786914,
      "grad_norm": 0.12359927873014928,
      "kl": 0.1367950439453125,
      "learning_rate": 4.957770494107496e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 11722
    },
    {
      "completion_length": 238.42858409881592,
      "epoch": 1.965673330818559,
      "grad_norm": 0.2447817104012549,
      "kl": 0.10650634765625,
      "learning_rate": 4.957748536005189e-07,
      "loss": 0.0001,
      "reward": 1.7732143476605415,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7776786014437675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11724
    },
    {
      "completion_length": 244.68304824829102,
      "epoch": 1.9660086340584266,
      "grad_norm": 0.25763781640983036,
      "kl": 0.107635498046875,
      "learning_rate": 4.957726572244233e-07,
      "loss": 0.0001,
      "reward": 1.7375000640749931,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643234461546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11726
    },
    {
      "completion_length": 256.0982275009155,
      "epoch": 1.9663439372982943,
      "grad_norm": 0.17076517137928504,
      "kl": 0.104156494140625,
      "learning_rate": 4.95770460282468e-07,
      "loss": 0.0001,
      "reward": 1.7803571969270706,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7848214656114578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11728
    },
    {
      "completion_length": 250.0982255935669,
      "epoch": 1.9666792405381617,
      "grad_norm": 0.28394623429905763,
      "kl": 0.131378173828125,
      "learning_rate": 4.957682627746578e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7589286044239998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11730
    },
    {
      "completion_length": 245.6473331451416,
      "epoch": 1.967014543778029,
      "grad_norm": 0.25643459126742696,
      "kl": 0.1122894287109375,
      "learning_rate": 4.957660647009981e-07,
      "loss": 0.0001,
      "reward": 1.7875000685453415,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7919643074274063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11732
    },
    {
      "completion_length": 244.6428680419922,
      "epoch": 1.9673498470178967,
      "grad_norm": 0.18632279817106337,
      "kl": 0.1124420166015625,
      "learning_rate": 4.957638660614938e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857544124126,
      "rewards/format_reward_func": 1.0,
      "step": 11734
    },
    {
      "completion_length": 256.9151916503906,
      "epoch": 1.9676851502577644,
      "grad_norm": 0.22765448085255816,
      "kl": 0.1229400634765625,
      "learning_rate": 4.957616668561498e-07,
      "loss": 0.0001,
      "reward": 1.7446429282426834,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7491071857511997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11736
    },
    {
      "completion_length": 246.9017972946167,
      "epoch": 1.968020453497632,
      "grad_norm": 0.6592077745590738,
      "kl": 0.13031005859375,
      "learning_rate": 4.957594670849715e-07,
      "loss": 0.0001,
      "reward": 1.8107143342494965,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.8196428827941418,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11738
    },
    {
      "completion_length": 255.5178689956665,
      "epoch": 1.9683557567374996,
      "grad_norm": 0.26862816593064337,
      "kl": 0.117919921875,
      "learning_rate": 4.957572667479637e-07,
      "loss": 0.0001,
      "reward": 1.7303572073578835,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7348214536905289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11740
    },
    {
      "completion_length": 246.37947368621826,
      "epoch": 1.968691059977367,
      "grad_norm": 0.2317197207042673,
      "kl": 0.115509033203125,
      "learning_rate": 4.957550658451315e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 11742
    },
    {
      "completion_length": 257.39286708831787,
      "epoch": 1.9690263632172345,
      "grad_norm": 0.1188256513109012,
      "kl": 0.1306610107421875,
      "learning_rate": 4.957528643764801e-07,
      "loss": 0.0001,
      "reward": 1.7500000894069672,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 11744
    },
    {
      "completion_length": 254.32144260406494,
      "epoch": 1.969361666457102,
      "grad_norm": 0.13580848106302223,
      "kl": 0.150390625,
      "learning_rate": 4.957506623420145e-07,
      "loss": 0.0002,
      "reward": 1.7446429133415222,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7580357491970062,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11746
    },
    {
      "completion_length": 249.18304824829102,
      "epoch": 1.9696969696969697,
      "grad_norm": 0.17056617952178113,
      "kl": 0.143218994140625,
      "learning_rate": 4.957484597417398e-07,
      "loss": 0.0001,
      "reward": 1.773214340209961,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7866071611642838,
      "rewards/format_reward_func": 0.9866071455180645,
      "step": 11748
    },
    {
      "completion_length": 257.696439743042,
      "epoch": 1.9700322729368374,
      "grad_norm": 0.05453721168213884,
      "kl": 0.112518310546875,
      "learning_rate": 4.957462565756609e-07,
      "loss": 0.0001,
      "reward": 1.7767857760190964,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7812500298023224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11750
    },
    {
      "completion_length": 258.0759057998657,
      "epoch": 1.9703675761767048,
      "grad_norm": 0.1605862489015661,
      "kl": 0.10565185546875,
      "learning_rate": 4.95744052843783e-07,
      "loss": 0.0001,
      "reward": 1.798214353621006,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8026786111295223,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11752
    },
    {
      "completion_length": 254.1607265472412,
      "epoch": 1.9707028794165724,
      "grad_norm": 0.19646260246464867,
      "kl": 0.1075439453125,
      "learning_rate": 4.957418485461112e-07,
      "loss": 0.0001,
      "reward": 1.7464286461472511,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 11754
    },
    {
      "completion_length": 252.89733505249023,
      "epoch": 1.9710381826564398,
      "grad_norm": 0.3499983244450021,
      "kl": 0.2943115234375,
      "learning_rate": 4.957396436826506e-07,
      "loss": 0.0003,
      "reward": 1.8098214641213417,
      "reward_std": 0.056821079924702644,
      "rewards/equation_reward_func": 0.816071443259716,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 11756
    },
    {
      "completion_length": 257.3348340988159,
      "epoch": 1.9713734858963075,
      "grad_norm": 0.21213137164595536,
      "kl": 0.149383544921875,
      "learning_rate": 4.957374382534063e-07,
      "loss": 0.0001,
      "reward": 1.7482143491506577,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526786085218191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11758
    },
    {
      "completion_length": 254.6785831451416,
      "epoch": 1.971708789136175,
      "grad_norm": 0.16187445714713122,
      "kl": 0.0967864990234375,
      "learning_rate": 4.957352322583831e-07,
      "loss": 0.0001,
      "reward": 1.8214286267757416,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8214285895228386,
      "rewards/format_reward_func": 1.0,
      "step": 11760
    },
    {
      "completion_length": 255.33929538726807,
      "epoch": 1.9720440923760427,
      "grad_norm": 0.17196609458920475,
      "kl": 0.127349853515625,
      "learning_rate": 4.957330256975865e-07,
      "loss": 0.0001,
      "reward": 1.7446429431438446,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.749107176437974,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11762
    },
    {
      "completion_length": 248.13840579986572,
      "epoch": 1.9723793956159101,
      "grad_norm": 0.29006952980499107,
      "kl": 0.0870361328125,
      "learning_rate": 4.957308185710212e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.785714328289032,
      "rewards/format_reward_func": 1.0,
      "step": 11764
    },
    {
      "completion_length": 249.27679634094238,
      "epoch": 1.9727146988557775,
      "grad_norm": 0.10785375158854305,
      "kl": 0.157928466796875,
      "learning_rate": 4.957286108786925e-07,
      "loss": 0.0002,
      "reward": 1.8107143342494965,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8107143193483353,
      "rewards/format_reward_func": 1.0,
      "step": 11766
    },
    {
      "completion_length": 247.71429920196533,
      "epoch": 1.9730500020956452,
      "grad_norm": 0.20807208197803714,
      "kl": 0.229705810546875,
      "learning_rate": 4.957264026206054e-07,
      "loss": 0.0002,
      "reward": 1.8000000789761543,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 11768
    },
    {
      "completion_length": 246.93304920196533,
      "epoch": 1.9733853053355128,
      "grad_norm": 0.17862243170744227,
      "kl": 0.180389404296875,
      "learning_rate": 4.957241937967651e-07,
      "loss": 0.0002,
      "reward": 1.8089286237955093,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8133928701281548,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11770
    },
    {
      "completion_length": 252.46876049041748,
      "epoch": 1.9737206085753805,
      "grad_norm": 0.22827912919286747,
      "kl": 0.3739166259765625,
      "learning_rate": 4.957219844071765e-07,
      "loss": 0.0004,
      "reward": 1.7250000908970833,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7250000350177288,
      "rewards/format_reward_func": 1.0,
      "step": 11772
    },
    {
      "completion_length": 236.30358409881592,
      "epoch": 1.9740559118152479,
      "grad_norm": 0.18547557891243543,
      "kl": 0.1163330078125,
      "learning_rate": 4.957197744518449e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857529222965,
      "rewards/format_reward_func": 1.0,
      "step": 11774
    },
    {
      "completion_length": 251.08483409881592,
      "epoch": 1.9743912150551155,
      "grad_norm": 0.5494083041560429,
      "kl": 0.337066650390625,
      "learning_rate": 4.957175639307751e-07,
      "loss": 0.0003,
      "reward": 1.7678572237491608,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 11776
    },
    {
      "completion_length": 252.4866189956665,
      "epoch": 1.974726518294983,
      "grad_norm": 0.1988050122697657,
      "kl": 0.1005859375,
      "learning_rate": 4.957153528439725e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.757142897695303,
      "rewards/format_reward_func": 1.0,
      "step": 11778
    },
    {
      "completion_length": 250.71429634094238,
      "epoch": 1.9750618215348505,
      "grad_norm": 0.17271351151488437,
      "kl": 0.090301513671875,
      "learning_rate": 4.95713141191442e-07,
      "loss": 0.0001,
      "reward": 1.7142857983708382,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7142857406288385,
      "rewards/format_reward_func": 1.0,
      "step": 11780
    },
    {
      "completion_length": 247.45983123779297,
      "epoch": 1.9753971247747182,
      "grad_norm": 0.1843118865578671,
      "kl": 0.389495849609375,
      "learning_rate": 4.957109289731888e-07,
      "loss": 0.0004,
      "reward": 1.7821429371833801,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 11782
    },
    {
      "completion_length": 236.5982255935669,
      "epoch": 1.9757324280145858,
      "grad_norm": 0.21773081688597665,
      "kl": 0.3225250244140625,
      "learning_rate": 4.957087161892179e-07,
      "loss": 0.0003,
      "reward": 1.7750000655651093,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 11784
    },
    {
      "completion_length": 241.3973331451416,
      "epoch": 1.9760677312544532,
      "grad_norm": 0.26868741544036107,
      "kl": 0.7315216064453125,
      "learning_rate": 4.957065028395344e-07,
      "loss": 0.0007,
      "reward": 1.8214285969734192,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8214285969734192,
      "rewards/format_reward_func": 1.0,
      "step": 11786
    },
    {
      "completion_length": 248.35715579986572,
      "epoch": 1.9764030344943206,
      "grad_norm": 0.20328298548363283,
      "kl": 0.1676788330078125,
      "learning_rate": 4.957042889241435e-07,
      "loss": 0.0002,
      "reward": 1.7750000655651093,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000394880772,
      "rewards/format_reward_func": 1.0,
      "step": 11788
    },
    {
      "completion_length": 243.7544755935669,
      "epoch": 1.9767383377341883,
      "grad_norm": 0.1907004951600171,
      "kl": 0.1032867431640625,
      "learning_rate": 4.957020744430502e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 11790
    },
    {
      "completion_length": 233.78572368621826,
      "epoch": 1.977073640974056,
      "grad_norm": 0.2596566984727189,
      "kl": 0.09918212890625,
      "learning_rate": 4.956998593962596e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 11792
    },
    {
      "completion_length": 236.5089406967163,
      "epoch": 1.9774089442139235,
      "grad_norm": 0.30107402027910873,
      "kl": 0.442413330078125,
      "learning_rate": 4.956976437837768e-07,
      "loss": 0.0004,
      "reward": 1.8000000789761543,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000193715096,
      "rewards/format_reward_func": 1.0,
      "step": 11794
    },
    {
      "completion_length": 242.1071538925171,
      "epoch": 1.9777442474537912,
      "grad_norm": 0.22727364686394377,
      "kl": 0.29437255859375,
      "learning_rate": 4.95695427605607e-07,
      "loss": 0.0003,
      "reward": 1.803571492433548,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 11796
    },
    {
      "completion_length": 243.19643878936768,
      "epoch": 1.9780795506936586,
      "grad_norm": 0.25835029992767844,
      "kl": 0.1036529541015625,
      "learning_rate": 4.956932108617552e-07,
      "loss": 0.0001,
      "reward": 1.760714367032051,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 11798
    },
    {
      "completion_length": 243.7991180419922,
      "epoch": 1.978414853933526,
      "grad_norm": 0.20062119433370976,
      "kl": 0.1900177001953125,
      "learning_rate": 4.956909935522265e-07,
      "loss": 0.0002,
      "reward": 1.74642863124609,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464285902678967,
      "rewards/format_reward_func": 1.0,
      "step": 11800
    },
    {
      "completion_length": 237.56251049041748,
      "epoch": 1.9787501571733936,
      "grad_norm": 0.19331615368744728,
      "kl": 0.183807373046875,
      "learning_rate": 4.95688775677026e-07,
      "loss": 0.0002,
      "reward": 1.7892857939004898,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 11802
    },
    {
      "completion_length": 243.19643878936768,
      "epoch": 1.9790854604132613,
      "grad_norm": 0.26249176823337705,
      "kl": 0.14044189453125,
      "learning_rate": 4.956865572361589e-07,
      "loss": 0.0001,
      "reward": 1.7000000849366188,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7000000309199095,
      "rewards/format_reward_func": 1.0,
      "step": 11804
    },
    {
      "completion_length": 231.95090198516846,
      "epoch": 1.979420763653129,
      "grad_norm": 0.11336370989884989,
      "kl": 0.0840301513671875,
      "learning_rate": 4.956843382296303e-07,
      "loss": 0.0001,
      "reward": 1.82857146859169,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8285714462399483,
      "rewards/format_reward_func": 1.0,
      "step": 11806
    },
    {
      "completion_length": 234.89733123779297,
      "epoch": 1.9797560668929963,
      "grad_norm": 0.16502533057972615,
      "kl": 0.1882171630859375,
      "learning_rate": 4.956821186574453e-07,
      "loss": 0.0002,
      "reward": 1.8000000417232513,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 11808
    },
    {
      "completion_length": 250.22768783569336,
      "epoch": 1.980091370132864,
      "grad_norm": 0.2793167309232402,
      "kl": 0.33477783203125,
      "learning_rate": 4.956798985196089e-07,
      "loss": 0.0003,
      "reward": 1.7535714954137802,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 11810
    },
    {
      "completion_length": 219.44643878936768,
      "epoch": 1.9804266733727314,
      "grad_norm": 0.1115267623950336,
      "kl": 0.0941162109375,
      "learning_rate": 4.956776778161262e-07,
      "loss": 0.0001,
      "reward": 1.8107143193483353,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.810714315623045,
      "rewards/format_reward_func": 1.0,
      "step": 11812
    },
    {
      "completion_length": 239.5982255935669,
      "epoch": 1.980761976612599,
      "grad_norm": 0.23300894138513203,
      "kl": 0.304840087890625,
      "learning_rate": 4.956754565470025e-07,
      "loss": 0.0003,
      "reward": 1.762946493923664,
      "reward_std": 0.03219861118122935,
      "rewards/equation_reward_func": 0.7642857488244772,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 11814
    },
    {
      "completion_length": 246.20983505249023,
      "epoch": 1.9810972798524666,
      "grad_norm": 0.24004711572000476,
      "kl": 0.123046875,
      "learning_rate": 4.956732347122428e-07,
      "loss": 0.0001,
      "reward": 1.8107143566012383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8107143193483353,
      "rewards/format_reward_func": 1.0,
      "step": 11816
    },
    {
      "completion_length": 243.9241180419922,
      "epoch": 1.9814325830923343,
      "grad_norm": 0.26924112990635035,
      "kl": 0.195953369140625,
      "learning_rate": 4.956710123118522e-07,
      "loss": 0.0002,
      "reward": 1.7714286223053932,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 11818
    },
    {
      "completion_length": 235.758939743042,
      "epoch": 1.9817678863322017,
      "grad_norm": 0.19173483626836108,
      "kl": 0.0982208251953125,
      "learning_rate": 4.956687893458359e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464286107569933,
      "rewards/format_reward_func": 1.0,
      "step": 11820
    },
    {
      "completion_length": 238.8259038925171,
      "epoch": 1.982103189572069,
      "grad_norm": 0.2064395106648162,
      "kl": 0.138946533203125,
      "learning_rate": 4.956665658141989e-07,
      "loss": 0.0001,
      "reward": 1.7000000849366188,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7000000290572643,
      "rewards/format_reward_func": 1.0,
      "step": 11822
    },
    {
      "completion_length": 241.92858123779297,
      "epoch": 1.9824384928119367,
      "grad_norm": 0.21597189485682244,
      "kl": 0.0922088623046875,
      "learning_rate": 4.956643417169464e-07,
      "loss": 0.0001,
      "reward": 1.750000037252903,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7589286100119352,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11824
    },
    {
      "completion_length": 241.89733505249023,
      "epoch": 1.9827737960518044,
      "grad_norm": 0.1910625884590803,
      "kl": 0.1044464111328125,
      "learning_rate": 4.956621170540834e-07,
      "loss": 0.0001,
      "reward": 1.7803572118282318,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7848214469850063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11826
    },
    {
      "completion_length": 231.9866180419922,
      "epoch": 1.983109099291672,
      "grad_norm": 0.13525587988326038,
      "kl": 0.090179443359375,
      "learning_rate": 4.956598918256151e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964286059141159,
      "rewards/format_reward_func": 1.0,
      "step": 11828
    },
    {
      "completion_length": 238.78572273254395,
      "epoch": 1.9834444025315394,
      "grad_norm": 0.21546699753134915,
      "kl": 0.71807861328125,
      "learning_rate": 4.956576660315468e-07,
      "loss": 0.0007,
      "reward": 1.8178571909666061,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8178571797907352,
      "rewards/format_reward_func": 1.0,
      "step": 11830
    },
    {
      "completion_length": 235.4910831451416,
      "epoch": 1.983779705771407,
      "grad_norm": 0.14166303451460713,
      "kl": 0.173187255859375,
      "learning_rate": 4.956554396718835e-07,
      "loss": 0.0002,
      "reward": 1.800000049173832,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 11832
    },
    {
      "completion_length": 246.1339406967163,
      "epoch": 1.9841150090112745,
      "grad_norm": 0.24250097169224633,
      "kl": 0.094482421875,
      "learning_rate": 4.956532127466302e-07,
      "loss": 0.0001,
      "reward": 1.733928643167019,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.738392885774374,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11834
    },
    {
      "completion_length": 240.32590293884277,
      "epoch": 1.984450312251142,
      "grad_norm": 0.23198980180167816,
      "kl": 0.0967254638671875,
      "learning_rate": 4.956509852557921e-07,
      "loss": 0.0001,
      "reward": 1.8178571686148643,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8178571723401546,
      "rewards/format_reward_func": 1.0,
      "step": 11836
    },
    {
      "completion_length": 247.0044755935669,
      "epoch": 1.9847856154910097,
      "grad_norm": 0.2247671504249133,
      "kl": 0.10345458984375,
      "learning_rate": 4.956487571993744e-07,
      "loss": 0.0001,
      "reward": 1.7785714715719223,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 11838
    },
    {
      "completion_length": 235.56697177886963,
      "epoch": 1.9851209187308774,
      "grad_norm": 0.22602112500128277,
      "kl": 0.2289276123046875,
      "learning_rate": 4.956465285773822e-07,
      "loss": 0.0002,
      "reward": 1.789285771548748,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 11840
    },
    {
      "completion_length": 238.22322368621826,
      "epoch": 1.9854562219707448,
      "grad_norm": 0.2934779266919726,
      "kl": 0.118408203125,
      "learning_rate": 4.956442993898206e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 11842
    },
    {
      "completion_length": 241.77679920196533,
      "epoch": 1.9857915252106122,
      "grad_norm": 0.25116354054121726,
      "kl": 0.3289794921875,
      "learning_rate": 4.956420696366947e-07,
      "loss": 0.0003,
      "reward": 1.8250000476837158,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8250000327825546,
      "rewards/format_reward_func": 1.0,
      "step": 11844
    },
    {
      "completion_length": 235.3169765472412,
      "epoch": 1.9861268284504798,
      "grad_norm": 0.13130801114902244,
      "kl": 0.2256011962890625,
      "learning_rate": 4.956398393180097e-07,
      "loss": 0.0002,
      "reward": 1.807142898440361,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428909897804,
      "rewards/format_reward_func": 1.0,
      "step": 11846
    },
    {
      "completion_length": 248.06250953674316,
      "epoch": 1.9864621316903475,
      "grad_norm": 0.22475071490244067,
      "kl": 0.15399169921875,
      "learning_rate": 4.956376084337707e-07,
      "loss": 0.0002,
      "reward": 1.8035714775323868,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 11848
    },
    {
      "completion_length": 230.65625858306885,
      "epoch": 1.986797434930215,
      "grad_norm": 0.21280194742861228,
      "kl": 0.1126708984375,
      "learning_rate": 4.956353769839829e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285925030708,
      "rewards/format_reward_func": 1.0,
      "step": 11850
    },
    {
      "completion_length": 245.8348331451416,
      "epoch": 1.9871327381700825,
      "grad_norm": 0.2820430557228894,
      "kl": 0.154571533203125,
      "learning_rate": 4.956331449686513e-07,
      "loss": 0.0002,
      "reward": 1.7035715207457542,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7035714648663998,
      "rewards/format_reward_func": 1.0,
      "step": 11852
    },
    {
      "completion_length": 240.2009038925171,
      "epoch": 1.9874680414099501,
      "grad_norm": 0.395312790500384,
      "kl": 0.4749603271484375,
      "learning_rate": 4.956309123877812e-07,
      "loss": 0.0005,
      "reward": 1.725000038743019,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7339286096394062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11854
    },
    {
      "completion_length": 239.75447368621826,
      "epoch": 1.9878033446498176,
      "grad_norm": 0.13378942220678655,
      "kl": 0.110565185546875,
      "learning_rate": 4.956286792413776e-07,
      "loss": 0.0001,
      "reward": 1.8142857626080513,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8142857402563095,
      "rewards/format_reward_func": 1.0,
      "step": 11856
    },
    {
      "completion_length": 231.61608123779297,
      "epoch": 1.9881386478896852,
      "grad_norm": 0.18475450589268128,
      "kl": 0.114471435546875,
      "learning_rate": 4.956264455294459e-07,
      "loss": 0.0001,
      "reward": 1.7607143372297287,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143260538578,
      "rewards/format_reward_func": 1.0,
      "step": 11858
    },
    {
      "completion_length": 238.84822463989258,
      "epoch": 1.9884739511295528,
      "grad_norm": 0.1555984953874135,
      "kl": 0.22589111328125,
      "learning_rate": 4.956242112519908e-07,
      "loss": 0.0002,
      "reward": 1.7785715013742447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 11860
    },
    {
      "completion_length": 232.9776906967163,
      "epoch": 1.9888092543694205,
      "grad_norm": 0.3377725470032938,
      "kl": 0.1129150390625,
      "learning_rate": 4.956219764090178e-07,
      "loss": 0.0001,
      "reward": 1.7875000685453415,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7919643260538578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11862
    },
    {
      "completion_length": 235.4509048461914,
      "epoch": 1.9891445576092879,
      "grad_norm": 0.2210363508356654,
      "kl": 0.12188720703125,
      "learning_rate": 4.956197410005319e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 11864
    },
    {
      "completion_length": 231.74108219146729,
      "epoch": 1.9894798608491553,
      "grad_norm": 0.1475095473252601,
      "kl": 0.1528472900390625,
      "learning_rate": 4.956175050265384e-07,
      "loss": 0.0002,
      "reward": 1.8267857804894447,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8312500193715096,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11866
    },
    {
      "completion_length": 245.79018878936768,
      "epoch": 1.989815164089023,
      "grad_norm": 0.15650601562466082,
      "kl": 0.136932373046875,
      "learning_rate": 4.956152684870422e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286059141159,
      "rewards/format_reward_func": 1.0,
      "step": 11868
    },
    {
      "completion_length": 246.821439743042,
      "epoch": 1.9901504673288906,
      "grad_norm": 0.1704402931331183,
      "kl": 0.185546875,
      "learning_rate": 4.956130313820487e-07,
      "loss": 0.0002,
      "reward": 1.7250000983476639,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7339286003261805,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11870
    },
    {
      "completion_length": 253.30358505249023,
      "epoch": 1.9904857705687582,
      "grad_norm": 0.38526863006362766,
      "kl": 0.310333251953125,
      "learning_rate": 4.956107937115629e-07,
      "loss": 0.0003,
      "reward": 1.7892857640981674,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 11872
    },
    {
      "completion_length": 247.47322368621826,
      "epoch": 1.9908210738086258,
      "grad_norm": 0.19867008633050648,
      "kl": 0.1563720703125,
      "learning_rate": 4.9560855547559e-07,
      "loss": 0.0002,
      "reward": 1.7232143506407738,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7276786006987095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11874
    },
    {
      "completion_length": 236.4866189956665,
      "epoch": 1.9911563770484932,
      "grad_norm": 0.1428311178265055,
      "kl": 0.11065673828125,
      "learning_rate": 4.956063166741351e-07,
      "loss": 0.0001,
      "reward": 1.7500000521540642,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000447034836,
      "rewards/format_reward_func": 1.0,
      "step": 11876
    },
    {
      "completion_length": 244.43751049041748,
      "epoch": 1.9914916802883607,
      "grad_norm": 0.2428530544891383,
      "kl": 0.1180877685546875,
      "learning_rate": 4.956040773072034e-07,
      "loss": 0.0001,
      "reward": 1.7696428894996643,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7741071805357933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11878
    },
    {
      "completion_length": 237.57590293884277,
      "epoch": 1.9918269835282283,
      "grad_norm": 1.1212923774941903,
      "kl": 0.1295166015625,
      "learning_rate": 4.956018373748001e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.06060915347188711,
      "rewards/equation_reward_func": 0.7732143066823483,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11880
    },
    {
      "completion_length": 230.1071538925171,
      "epoch": 1.992162286768096,
      "grad_norm": 0.1906099981958323,
      "kl": 0.11297607421875,
      "learning_rate": 4.955995968769302e-07,
      "loss": 0.0001,
      "reward": 1.7892857640981674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857491970062,
      "rewards/format_reward_func": 1.0,
      "step": 11882
    },
    {
      "completion_length": 234.17858219146729,
      "epoch": 1.9924975900079636,
      "grad_norm": 0.30340923600114456,
      "kl": 0.114898681640625,
      "learning_rate": 4.955973558135991e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.8089286014437675,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11884
    },
    {
      "completion_length": 241.93751049041748,
      "epoch": 1.992832893247831,
      "grad_norm": 0.24126169281411733,
      "kl": 0.104156494140625,
      "learning_rate": 4.955951141848117e-07,
      "loss": 0.0001,
      "reward": 1.7982143312692642,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8026785962283611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11886
    },
    {
      "completion_length": 242.1875114440918,
      "epoch": 1.9931681964876986,
      "grad_norm": 0.13715380355658038,
      "kl": 0.1492919921875,
      "learning_rate": 4.955928719905734e-07,
      "loss": 0.0001,
      "reward": 1.796428643167019,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.796428594738245,
      "rewards/format_reward_func": 1.0,
      "step": 11888
    },
    {
      "completion_length": 242.13393878936768,
      "epoch": 1.993503499727566,
      "grad_norm": 0.27291793247372775,
      "kl": 0.1129150390625,
      "learning_rate": 4.955906292308892e-07,
      "loss": 0.0001,
      "reward": 1.739285796880722,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7392857261002064,
      "rewards/format_reward_func": 1.0,
      "step": 11890
    },
    {
      "completion_length": 227.04465293884277,
      "epoch": 1.9938388029674337,
      "grad_norm": 0.23860127461571143,
      "kl": 0.0947265625,
      "learning_rate": 4.955883859057643e-07,
      "loss": 0.0001,
      "reward": 1.8285714611411095,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8285714630037546,
      "rewards/format_reward_func": 1.0,
      "step": 11892
    },
    {
      "completion_length": 234.54019165039062,
      "epoch": 1.9941741062073013,
      "grad_norm": 0.3011874190088839,
      "kl": 0.105133056640625,
      "learning_rate": 4.955861420152039e-07,
      "loss": 0.0001,
      "reward": 1.7892857939004898,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 11894
    },
    {
      "completion_length": 233.602689743042,
      "epoch": 1.994509409447169,
      "grad_norm": 0.24990488377435194,
      "kl": 0.118316650390625,
      "learning_rate": 4.955838975592131e-07,
      "loss": 0.0001,
      "reward": 1.7250000908970833,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7250000312924385,
      "rewards/format_reward_func": 1.0,
      "step": 11896
    },
    {
      "completion_length": 226.99554634094238,
      "epoch": 1.9948447126870363,
      "grad_norm": 0.37434620571171906,
      "kl": 0.120025634765625,
      "learning_rate": 4.955816525377971e-07,
      "loss": 0.0001,
      "reward": 1.8125000670552254,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8169643022119999,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11898
    },
    {
      "completion_length": 228.21876049041748,
      "epoch": 1.9951800159269037,
      "grad_norm": 0.1891353284554539,
      "kl": 0.113739013671875,
      "learning_rate": 4.955794069509611e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 11900
    },
    {
      "completion_length": 224.43304538726807,
      "epoch": 1.9955153191667714,
      "grad_norm": 0.6070216232285166,
      "kl": 0.18603515625,
      "learning_rate": 4.955771607987104e-07,
      "loss": 0.0002,
      "reward": 1.7303571924567223,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7348214685916901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11902
    },
    {
      "completion_length": 231.16072368621826,
      "epoch": 1.995850622406639,
      "grad_norm": 0.2760599269226852,
      "kl": 0.164215087890625,
      "learning_rate": 4.955749140810499e-07,
      "loss": 0.0002,
      "reward": 1.7589286342263222,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7633928768336773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11904
    },
    {
      "completion_length": 217.94197463989258,
      "epoch": 1.9961859256465067,
      "grad_norm": 0.2162571700265734,
      "kl": 0.114288330078125,
      "learning_rate": 4.955726667979848e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143044471741,
      "rewards/format_reward_func": 1.0,
      "step": 11906
    },
    {
      "completion_length": 233.70536518096924,
      "epoch": 1.996521228886374,
      "grad_norm": 0.17218754513613752,
      "kl": 0.125152587890625,
      "learning_rate": 4.955704189495205e-07,
      "loss": 0.0001,
      "reward": 1.7803572118282318,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214507102966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11908
    },
    {
      "completion_length": 229.40179634094238,
      "epoch": 1.9968565321262417,
      "grad_norm": 0.3804112945798657,
      "kl": 0.1534423828125,
      "learning_rate": 4.955681705356621e-07,
      "loss": 0.0002,
      "reward": 1.7053572237491608,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7098214775323868,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11910
    },
    {
      "completion_length": 230.25000953674316,
      "epoch": 1.9971918353661091,
      "grad_norm": 0.9168094662360421,
      "kl": 0.241485595703125,
      "learning_rate": 4.955659215564145e-07,
      "loss": 0.0002,
      "reward": 1.7750000655651093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 11912
    },
    {
      "completion_length": 224.8616189956665,
      "epoch": 1.9975271386059767,
      "grad_norm": 0.16848882583586025,
      "kl": 0.13287353515625,
      "learning_rate": 4.955636720117833e-07,
      "loss": 0.0001,
      "reward": 1.8107143267989159,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8107143081724644,
      "rewards/format_reward_func": 1.0,
      "step": 11914
    },
    {
      "completion_length": 230.39732933044434,
      "epoch": 1.9978624418458444,
      "grad_norm": 0.558286576121209,
      "kl": 0.13238525390625,
      "learning_rate": 4.955614219017734e-07,
      "loss": 0.0001,
      "reward": 1.7732143625617027,
      "reward_std": 0.08838834892958403,
      "rewards/equation_reward_func": 0.7866071723401546,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 11916
    },
    {
      "completion_length": 218.53572463989258,
      "epoch": 1.998197745085712,
      "grad_norm": 0.12048965951566463,
      "kl": 0.1212921142578125,
      "learning_rate": 4.9555917122639e-07,
      "loss": 0.0001,
      "reward": 1.798214316368103,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.8026786036789417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11918
    },
    {
      "completion_length": 229.7723331451416,
      "epoch": 1.9985330483255794,
      "grad_norm": 0.26405640247652984,
      "kl": 0.13812255859375,
      "learning_rate": 4.955569199856384e-07,
      "loss": 0.0001,
      "reward": 1.7732143327593803,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7776786014437675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11920
    },
    {
      "completion_length": 222.51340293884277,
      "epoch": 1.9988683515654468,
      "grad_norm": 0.20701527554064678,
      "kl": 0.119659423828125,
      "learning_rate": 4.955546681795238e-07,
      "loss": 0.0001,
      "reward": 1.753571517765522,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714730620384,
      "rewards/format_reward_func": 1.0,
      "step": 11922
    },
    {
      "completion_length": 233.08929634094238,
      "epoch": 1.9992036548053145,
      "grad_norm": 0.2992797604063413,
      "kl": 0.166656494140625,
      "learning_rate": 4.955524158080513e-07,
      "loss": 0.0002,
      "reward": 1.7267858013510704,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7312500365078449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11924
    },
    {
      "completion_length": 223.50447273254395,
      "epoch": 1.9995389580451821,
      "grad_norm": 0.21232811921496075,
      "kl": 0.116607666015625,
      "learning_rate": 4.955501628712259e-07,
      "loss": 0.0001,
      "reward": 1.783035770058632,
      "reward_std": 0.04419417306780815,
      "rewards/equation_reward_func": 0.7848214581608772,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 11926
    },
    {
      "completion_length": 224.19643783569336,
      "epoch": 1.9998742612850497,
      "grad_norm": 0.22939465547794297,
      "kl": 0.9582061767578125,
      "learning_rate": 4.955479093690532e-07,
      "loss": 0.001,
      "reward": 1.7303572073578835,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7348214648663998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11928
    },
    {
      "completion_length": 224.04546356201172,
      "epoch": 2.0003353032398676,
      "grad_norm": 0.20265008955204794,
      "kl": 0.11465731534090909,
      "learning_rate": 4.955456553015381e-07,
      "loss": 0.0002,
      "reward": 1.7909091440114109,
      "reward_std": 0.053262587298046456,
      "rewards/equation_reward_func": 0.7941558740355752,
      "rewards/format_reward_func": 0.9967532483014193,
      "step": 11930
    },
    {
      "completion_length": 219.13393878936768,
      "epoch": 2.0006706064797353,
      "grad_norm": 0.25991971602352476,
      "kl": 0.1035919189453125,
      "learning_rate": 4.955434006686859e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857566475868,
      "rewards/format_reward_func": 1.0,
      "step": 11932
    },
    {
      "completion_length": 217.21429634094238,
      "epoch": 2.0010059097196025,
      "grad_norm": 0.17602677588111465,
      "kl": 0.135345458984375,
      "learning_rate": 4.955411454705016e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964285835623741,
      "rewards/format_reward_func": 1.0,
      "step": 11934
    },
    {
      "completion_length": 226.6651906967163,
      "epoch": 2.00134121295947,
      "grad_norm": 0.23395105985453332,
      "kl": 0.1150360107421875,
      "learning_rate": 4.955388897069907e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 11936
    },
    {
      "completion_length": 221.41518783569336,
      "epoch": 2.0016765161993377,
      "grad_norm": 0.1894342197370109,
      "kl": 0.1356201171875,
      "learning_rate": 4.955366333781581e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 11938
    },
    {
      "completion_length": 224.43304634094238,
      "epoch": 2.0020118194392054,
      "grad_norm": 0.19172940809934583,
      "kl": 0.13311767578125,
      "learning_rate": 4.955343764840093e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8107143007218838,
      "rewards/format_reward_func": 1.0,
      "step": 11940
    },
    {
      "completion_length": 210.82143878936768,
      "epoch": 2.002347122679073,
      "grad_norm": 0.19378477540179337,
      "kl": 0.1277923583984375,
      "learning_rate": 4.955321190245491e-07,
      "loss": 0.0001,
      "reward": 1.7571429461240768,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.757142897695303,
      "rewards/format_reward_func": 1.0,
      "step": 11942
    },
    {
      "completion_length": 218.50000953674316,
      "epoch": 2.0026824259189406,
      "grad_norm": 0.20741615096888355,
      "kl": 0.123382568359375,
      "learning_rate": 4.955298609997831e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 11944
    },
    {
      "completion_length": 226.2321538925171,
      "epoch": 2.003017729158808,
      "grad_norm": 0.15846754537055674,
      "kl": 0.10369873046875,
      "learning_rate": 4.955276024097163e-07,
      "loss": 0.0001,
      "reward": 1.7607143446803093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 11946
    },
    {
      "completion_length": 228.56697463989258,
      "epoch": 2.0033530323986755,
      "grad_norm": 0.3061465529034817,
      "kl": 0.119293212890625,
      "learning_rate": 4.955253432543539e-07,
      "loss": 0.0001,
      "reward": 1.7696429267525673,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7741071600466967,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11948
    },
    {
      "completion_length": 217.4821538925171,
      "epoch": 2.003688335638543,
      "grad_norm": 0.373817116939758,
      "kl": 0.13421630859375,
      "learning_rate": 4.955230835337012e-07,
      "loss": 0.0001,
      "reward": 1.733928643167019,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.738392885774374,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11950
    },
    {
      "completion_length": 220.44643878936768,
      "epoch": 2.0040236388784107,
      "grad_norm": 0.2275983755411765,
      "kl": 0.130859375,
      "learning_rate": 4.955208232477633e-07,
      "loss": 0.0001,
      "reward": 1.74642863124609,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 11952
    },
    {
      "completion_length": 208.45536708831787,
      "epoch": 2.0043589421182784,
      "grad_norm": 0.1904971770533389,
      "kl": 0.129486083984375,
      "learning_rate": 4.955185623965454e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 11954
    },
    {
      "completion_length": 225.9821538925171,
      "epoch": 2.0046942453581456,
      "grad_norm": 0.17717864307604564,
      "kl": 0.13543701171875,
      "learning_rate": 4.955163009800527e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7410714738070965,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11956
    },
    {
      "completion_length": 214.18750953674316,
      "epoch": 2.005029548598013,
      "grad_norm": 0.16732658290309635,
      "kl": 0.115814208984375,
      "learning_rate": 4.955140389982904e-07,
      "loss": 0.0001,
      "reward": 1.8142857775092125,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8142857402563095,
      "rewards/format_reward_func": 1.0,
      "step": 11958
    },
    {
      "completion_length": 229.22322463989258,
      "epoch": 2.005364851837881,
      "grad_norm": 0.21928717715473264,
      "kl": 0.118377685546875,
      "learning_rate": 4.95511776451264e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7946428954601288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11960
    },
    {
      "completion_length": 219.00001049041748,
      "epoch": 2.0057001550777485,
      "grad_norm": 0.11188186472262818,
      "kl": 0.129486083984375,
      "learning_rate": 4.955095133389783e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 11962
    },
    {
      "completion_length": 216.20536613464355,
      "epoch": 2.006035458317616,
      "grad_norm": 0.287615123732697,
      "kl": 0.158966064453125,
      "learning_rate": 4.955072496614386e-07,
      "loss": 0.0002,
      "reward": 1.7839286401867867,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7883928753435612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11964
    },
    {
      "completion_length": 221.66965293884277,
      "epoch": 2.0063707615574837,
      "grad_norm": 0.13775078111321387,
      "kl": 0.1495361328125,
      "learning_rate": 4.955049854186503e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964285798370838,
      "rewards/format_reward_func": 1.0,
      "step": 11966
    },
    {
      "completion_length": 220.4285831451416,
      "epoch": 2.006706064797351,
      "grad_norm": 0.20963167919580555,
      "kl": 0.142425537109375,
      "learning_rate": 4.955027206106184e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143126428127,
      "rewards/format_reward_func": 1.0,
      "step": 11968
    },
    {
      "completion_length": 208.98661518096924,
      "epoch": 2.0070413680372186,
      "grad_norm": 0.048640389814782435,
      "kl": 0.104705810546875,
      "learning_rate": 4.955004552373483e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 11970
    },
    {
      "completion_length": 224.97322463989258,
      "epoch": 2.007376671277086,
      "grad_norm": 0.41609726327583996,
      "kl": 0.146697998046875,
      "learning_rate": 4.95498189298845e-07,
      "loss": 0.0001,
      "reward": 1.7232143878936768,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7276786044239998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11972
    },
    {
      "completion_length": 220.17858123779297,
      "epoch": 2.007711974516954,
      "grad_norm": 0.2719038569618615,
      "kl": 0.10687255859375,
      "learning_rate": 4.954959227951139e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 11974
    },
    {
      "completion_length": 215.39733123779297,
      "epoch": 2.0080472777568215,
      "grad_norm": 0.17050510753330186,
      "kl": 0.124847412109375,
      "learning_rate": 4.954936557261603e-07,
      "loss": 0.0001,
      "reward": 1.7196429297327995,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7241071779280901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11976
    },
    {
      "completion_length": 222.39286613464355,
      "epoch": 2.008382580996689,
      "grad_norm": 0.1061362122372038,
      "kl": 0.139892578125,
      "learning_rate": 4.954913880919892e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 11978
    },
    {
      "completion_length": 219.17858123779297,
      "epoch": 2.0087178842365563,
      "grad_norm": 0.4607873940896255,
      "kl": 0.220184326171875,
      "learning_rate": 4.954891198926058e-07,
      "loss": 0.0002,
      "reward": 1.7821429073810577,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428999304771,
      "rewards/format_reward_func": 1.0,
      "step": 11980
    },
    {
      "completion_length": 218.12054443359375,
      "epoch": 2.009053187476424,
      "grad_norm": 0.243853842105626,
      "kl": 0.13153076171875,
      "learning_rate": 4.954868511280156e-07,
      "loss": 0.0001,
      "reward": 1.8053572103381157,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8098214529454708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11982
    },
    {
      "completion_length": 219.4241180419922,
      "epoch": 2.0093884907162916,
      "grad_norm": 0.17020714244723772,
      "kl": 0.13604736328125,
      "learning_rate": 4.954845817982234e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 11984
    },
    {
      "completion_length": 213.47322368621826,
      "epoch": 2.009723793956159,
      "grad_norm": 0.09836840420537803,
      "kl": 0.15594482421875,
      "learning_rate": 4.95482311903235e-07,
      "loss": 0.0002,
      "reward": 1.733928643167019,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7383928932249546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 11986
    },
    {
      "completion_length": 212.58929538726807,
      "epoch": 2.010059097196027,
      "grad_norm": 0.275191574538384,
      "kl": 0.12689208984375,
      "learning_rate": 4.954800414430551e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 11988
    },
    {
      "completion_length": 232.5357265472412,
      "epoch": 2.010394400435894,
      "grad_norm": 0.1892092833782436,
      "kl": 0.17681884765625,
      "learning_rate": 4.954777704176891e-07,
      "loss": 0.0002,
      "reward": 1.7392857670783997,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857261002064,
      "rewards/format_reward_func": 1.0,
      "step": 11990
    },
    {
      "completion_length": 217.90179538726807,
      "epoch": 2.0107297036757616,
      "grad_norm": 0.1421242256000931,
      "kl": 0.14764404296875,
      "learning_rate": 4.954754988271423e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143227010965,
      "rewards/format_reward_func": 1.0,
      "step": 11992
    },
    {
      "completion_length": 220.9910831451416,
      "epoch": 2.0110650069156293,
      "grad_norm": 0.2826369696564157,
      "kl": 0.155670166015625,
      "learning_rate": 4.9547322667142e-07,
      "loss": 0.0002,
      "reward": 1.7919643595814705,
      "reward_std": 0.061871842946857214,
      "rewards/equation_reward_func": 0.798214316368103,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 11994
    },
    {
      "completion_length": 226.55358028411865,
      "epoch": 2.011400310155497,
      "grad_norm": 0.23081500085930848,
      "kl": 0.143829345703125,
      "learning_rate": 4.954709539505272e-07,
      "loss": 0.0001,
      "reward": 1.8250000551342964,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.8339286036789417,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 11996
    },
    {
      "completion_length": 225.42411708831787,
      "epoch": 2.0117356133953646,
      "grad_norm": 0.15678403010002545,
      "kl": 0.1483154296875,
      "learning_rate": 4.954686806644692e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714286148548126,
      "rewards/format_reward_func": 1.0,
      "step": 11998
    },
    {
      "completion_length": 234.8928689956665,
      "epoch": 2.012070916635232,
      "grad_norm": 0.41964272204763825,
      "kl": 0.121246337890625,
      "learning_rate": 4.954664068132514e-07,
      "loss": 0.0001,
      "reward": 1.8071429133415222,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.8160714618861675,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12000
    },
    {
      "completion_length": 230.38840293884277,
      "epoch": 2.0124062198750994,
      "grad_norm": 0.18041820981293524,
      "kl": 0.126129150390625,
      "learning_rate": 4.954641323968788e-07,
      "loss": 0.0001,
      "reward": 1.7553571909666061,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7598214633762836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12002
    },
    {
      "completion_length": 234.41965579986572,
      "epoch": 2.012741523114967,
      "grad_norm": 0.2307812936228628,
      "kl": 0.127593994140625,
      "learning_rate": 4.954618574153569e-07,
      "loss": 0.0001,
      "reward": 1.8107143342494965,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8107142932713032,
      "rewards/format_reward_func": 1.0,
      "step": 12004
    },
    {
      "completion_length": 247.77233695983887,
      "epoch": 2.0130768263548346,
      "grad_norm": 0.26071255191800646,
      "kl": 0.129608154296875,
      "learning_rate": 4.954595818686907e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.7625000216066837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12006
    },
    {
      "completion_length": 232.83929634094238,
      "epoch": 2.0134121295947023,
      "grad_norm": 0.20923252643628967,
      "kl": 0.13958740234375,
      "learning_rate": 4.954573057568856e-07,
      "loss": 0.0001,
      "reward": 1.69821435213089,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7026786059141159,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12008
    },
    {
      "completion_length": 237.15179634094238,
      "epoch": 2.01374743283457,
      "grad_norm": 0.25729457637255143,
      "kl": 0.141448974609375,
      "learning_rate": 4.954550290799468e-07,
      "loss": 0.0001,
      "reward": 1.7214286401867867,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7214285954833031,
      "rewards/format_reward_func": 1.0,
      "step": 12010
    },
    {
      "completion_length": 234.2946538925171,
      "epoch": 2.014082736074437,
      "grad_norm": 0.27225565276967606,
      "kl": 0.14306640625,
      "learning_rate": 4.954527518378794e-07,
      "loss": 0.0001,
      "reward": 1.7321429401636124,
      "reward_std": 0.08586296439170837,
      "rewards/equation_reward_func": 0.7410714589059353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12012
    },
    {
      "completion_length": 229.37500953674316,
      "epoch": 2.0144180393143047,
      "grad_norm": 0.14623596893714957,
      "kl": 0.12762451171875,
      "learning_rate": 4.954504740306887e-07,
      "loss": 0.0001,
      "reward": 1.8089286386966705,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.8133928813040257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12014
    },
    {
      "completion_length": 230.6339406967163,
      "epoch": 2.0147533425541724,
      "grad_norm": 0.2575949222882184,
      "kl": 0.12567138671875,
      "learning_rate": 4.954481956583802e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.785714328289032,
      "rewards/format_reward_func": 1.0,
      "step": 12016
    },
    {
      "completion_length": 232.95090293884277,
      "epoch": 2.01508864579404,
      "grad_norm": 0.2922948645964965,
      "kl": 0.154632568359375,
      "learning_rate": 4.954459167209588e-07,
      "loss": 0.0002,
      "reward": 1.80892863124609,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.8133928589522839,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12018
    },
    {
      "completion_length": 243.82590198516846,
      "epoch": 2.0154239490339076,
      "grad_norm": 0.30686562447010646,
      "kl": 0.15203857421875,
      "learning_rate": 4.9544363721843e-07,
      "loss": 0.0002,
      "reward": 1.7553571984171867,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7598214484751225,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12020
    },
    {
      "completion_length": 225.49107933044434,
      "epoch": 2.0157592522737753,
      "grad_norm": 0.20599917438686824,
      "kl": 0.16033935546875,
      "learning_rate": 4.954413571507988e-07,
      "loss": 0.0002,
      "reward": 1.7892857640981674,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 12022
    },
    {
      "completion_length": 236.26786708831787,
      "epoch": 2.0160945555136425,
      "grad_norm": 0.24825955537246205,
      "kl": 0.14947509765625,
      "learning_rate": 4.954390765180707e-07,
      "loss": 0.0001,
      "reward": 1.7696429267525673,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12024
    },
    {
      "completion_length": 233.32143878936768,
      "epoch": 2.01642985875351,
      "grad_norm": 0.10900224773752099,
      "kl": 0.155120849609375,
      "learning_rate": 4.954367953202509e-07,
      "loss": 0.0002,
      "reward": 1.7375000566244125,
      "reward_std": 0.05808377265930176,
      "rewards/equation_reward_func": 0.7508928813040257,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 12026
    },
    {
      "completion_length": 221.6964406967163,
      "epoch": 2.0167651619933777,
      "grad_norm": 0.2646779873755802,
      "kl": 0.149627685546875,
      "learning_rate": 4.954345135573445e-07,
      "loss": 0.0001,
      "reward": 1.812500037252903,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8169643133878708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12028
    },
    {
      "completion_length": 223.02679538726807,
      "epoch": 2.0171004652332454,
      "grad_norm": 0.20961948724422066,
      "kl": 0.15570068359375,
      "learning_rate": 4.954322312293568e-07,
      "loss": 0.0002,
      "reward": 1.753571480512619,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7535714656114578,
      "rewards/format_reward_func": 1.0,
      "step": 12030
    },
    {
      "completion_length": 233.98661708831787,
      "epoch": 2.017435768473113,
      "grad_norm": 0.44244999777054755,
      "kl": 0.162322998046875,
      "learning_rate": 4.954299483362932e-07,
      "loss": 0.0002,
      "reward": 1.787500061094761,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7919643111526966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12032
    },
    {
      "completion_length": 226.696439743042,
      "epoch": 2.0177710717129806,
      "grad_norm": 0.0029737839904434033,
      "kl": 0.13922119140625,
      "learning_rate": 4.954276648781588e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 12034
    },
    {
      "completion_length": 238.12054634094238,
      "epoch": 2.018106374952848,
      "grad_norm": 0.15109126581438292,
      "kl": 0.155914306640625,
      "learning_rate": 4.95425380854959e-07,
      "loss": 0.0002,
      "reward": 1.7232143580913544,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7276785969734192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12036
    },
    {
      "completion_length": 240.7991180419922,
      "epoch": 2.0184416781927155,
      "grad_norm": 0.20111466564363314,
      "kl": 0.168670654296875,
      "learning_rate": 4.954230962666989e-07,
      "loss": 0.0002,
      "reward": 1.771428644657135,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7714286111295223,
      "rewards/format_reward_func": 1.0,
      "step": 12038
    },
    {
      "completion_length": 239.30358219146729,
      "epoch": 2.018776981432583,
      "grad_norm": 0.1693994281368671,
      "kl": 0.142608642578125,
      "learning_rate": 4.954208111133839e-07,
      "loss": 0.0001,
      "reward": 1.8196429014205933,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8241071701049805,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12040
    },
    {
      "completion_length": 243.7812614440918,
      "epoch": 2.0191122846724507,
      "grad_norm": 0.3227749388676018,
      "kl": 0.171875,
      "learning_rate": 4.954185253950191e-07,
      "loss": 0.0002,
      "reward": 1.7339286729693413,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7383928839117289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12042
    },
    {
      "completion_length": 236.8660831451416,
      "epoch": 2.0194475879123184,
      "grad_norm": 0.17381341547292486,
      "kl": 0.1773681640625,
      "learning_rate": 4.9541623911161e-07,
      "loss": 0.0002,
      "reward": 1.796428643167019,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.8053571730852127,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12044
    },
    {
      "completion_length": 234.32590198516846,
      "epoch": 2.0197828911521856,
      "grad_norm": 0.21425360276716288,
      "kl": 0.141387939453125,
      "learning_rate": 4.954139522631617e-07,
      "loss": 0.0001,
      "reward": 1.7285715118050575,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7375000193715096,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12046
    },
    {
      "completion_length": 239.37500858306885,
      "epoch": 2.020118194392053,
      "grad_norm": 0.30991828406477046,
      "kl": 0.19769287109375,
      "learning_rate": 4.954116648496793e-07,
      "loss": 0.0002,
      "reward": 1.7428572252392769,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 12048
    },
    {
      "completion_length": 242.7366189956665,
      "epoch": 2.020453497631921,
      "grad_norm": 0.16492187085630522,
      "kl": 0.258819580078125,
      "learning_rate": 4.954093768711685e-07,
      "loss": 0.0003,
      "reward": 1.800000049173832,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000342726707,
      "rewards/format_reward_func": 1.0,
      "step": 12050
    },
    {
      "completion_length": 244.13393878936768,
      "epoch": 2.0207888008717885,
      "grad_norm": 0.36228155609406354,
      "kl": 0.3056640625,
      "learning_rate": 4.954070883276342e-07,
      "loss": 0.0003,
      "reward": 1.8142857626080513,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857477068901,
      "rewards/format_reward_func": 1.0,
      "step": 12052
    },
    {
      "completion_length": 239.1741189956665,
      "epoch": 2.021124104111656,
      "grad_norm": 0.23091231775013618,
      "kl": 0.128692626953125,
      "learning_rate": 4.954047992190818e-07,
      "loss": 0.0001,
      "reward": 1.742857202887535,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571712225676,
      "rewards/format_reward_func": 1.0,
      "step": 12054
    },
    {
      "completion_length": 232.46876335144043,
      "epoch": 2.0214594073515237,
      "grad_norm": 0.24644318616572938,
      "kl": 0.320404052734375,
      "learning_rate": 4.954025095455166e-07,
      "loss": 0.0003,
      "reward": 1.8142857402563095,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.8232143186032772,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12056
    },
    {
      "completion_length": 234.58036708831787,
      "epoch": 2.021794710591391,
      "grad_norm": 0.18374386788386235,
      "kl": 0.17694091796875,
      "learning_rate": 4.954002193069438e-07,
      "loss": 0.0002,
      "reward": 1.7500000670552254,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 12058
    },
    {
      "completion_length": 229.8794755935669,
      "epoch": 2.0221300138312586,
      "grad_norm": 0.11192113605103671,
      "kl": 0.22650146484375,
      "learning_rate": 4.953979285033687e-07,
      "loss": 0.0002,
      "reward": 1.767857201397419,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 12060
    },
    {
      "completion_length": 248.83929538726807,
      "epoch": 2.022465317071126,
      "grad_norm": 0.20104151765938574,
      "kl": 0.20709228515625,
      "learning_rate": 4.953956371347966e-07,
      "loss": 0.0002,
      "reward": 1.7000000700354576,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7089286129921675,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12062
    },
    {
      "completion_length": 228.36161708831787,
      "epoch": 2.022800620310994,
      "grad_norm": 0.27693609985790313,
      "kl": 0.66949462890625,
      "learning_rate": 4.953933452012327e-07,
      "loss": 0.0007,
      "reward": 1.7750000730156898,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 12064
    },
    {
      "completion_length": 225.81251049041748,
      "epoch": 2.0231359235508615,
      "grad_norm": 0.26251403881160484,
      "kl": 0.167205810546875,
      "learning_rate": 4.953910527026824e-07,
      "loss": 0.0002,
      "reward": 1.7517858073115349,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7562500387430191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12066
    },
    {
      "completion_length": 228.04018783569336,
      "epoch": 2.0234712267907287,
      "grad_norm": 0.29266835993005574,
      "kl": 0.438812255859375,
      "learning_rate": 4.95388759639151e-07,
      "loss": 0.0004,
      "reward": 1.7803572043776512,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7848214581608772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12068
    },
    {
      "completion_length": 226.27679443359375,
      "epoch": 2.0238065300305963,
      "grad_norm": 0.19577650905885083,
      "kl": 0.363250732421875,
      "learning_rate": 4.953864660106435e-07,
      "loss": 0.0004,
      "reward": 1.7946429178118706,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7991071566939354,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12070
    },
    {
      "completion_length": 230.3526906967163,
      "epoch": 2.024141833270464,
      "grad_norm": 0.14275428631407802,
      "kl": 0.262969970703125,
      "learning_rate": 4.953841718171655e-07,
      "loss": 0.0003,
      "reward": 1.7803572192788124,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7848214507102966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12072
    },
    {
      "completion_length": 225.99108123779297,
      "epoch": 2.0244771365103316,
      "grad_norm": 0.17367730450703955,
      "kl": 0.56787109375,
      "learning_rate": 4.953818770587221e-07,
      "loss": 0.0006,
      "reward": 1.7321429252624512,
      "reward_std": 0.08586296439170837,
      "rewards/equation_reward_func": 0.7410714514553547,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12074
    },
    {
      "completion_length": 230.7009038925171,
      "epoch": 2.024812439750199,
      "grad_norm": 0.25745718941039286,
      "kl": 0.167633056640625,
      "learning_rate": 4.953795817353187e-07,
      "loss": 0.0002,
      "reward": 1.7857143357396126,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143022119999,
      "rewards/format_reward_func": 1.0,
      "step": 12076
    },
    {
      "completion_length": 230.60715293884277,
      "epoch": 2.025147742990067,
      "grad_norm": 0.4703462047223365,
      "kl": 0.290985107421875,
      "learning_rate": 4.953772858469605e-07,
      "loss": 0.0003,
      "reward": 1.7857143729925156,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 12078
    },
    {
      "completion_length": 225.26340198516846,
      "epoch": 2.025483046229934,
      "grad_norm": 0.209617257104548,
      "kl": 0.496490478515625,
      "learning_rate": 4.953749893936527e-07,
      "loss": 0.0005,
      "reward": 1.78035718947649,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.784821467474103,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12080
    },
    {
      "completion_length": 222.3928680419922,
      "epoch": 2.0258183494698017,
      "grad_norm": 0.2109313296218021,
      "kl": 0.129119873046875,
      "learning_rate": 4.953726923754008e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.785714328289032,
      "rewards/format_reward_func": 1.0,
      "step": 12082
    },
    {
      "completion_length": 228.6473331451416,
      "epoch": 2.0261536527096693,
      "grad_norm": 0.7119368148754645,
      "kl": 0.433197021484375,
      "learning_rate": 4.9537039479221e-07,
      "loss": 0.0004,
      "reward": 1.7767857611179352,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7812500409781933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12084
    },
    {
      "completion_length": 228.45983409881592,
      "epoch": 2.026488955949537,
      "grad_norm": 0.20029418610940258,
      "kl": 0.416046142578125,
      "learning_rate": 4.953680966440855e-07,
      "loss": 0.0004,
      "reward": 1.7535715252161026,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 12086
    },
    {
      "completion_length": 227.72768783569336,
      "epoch": 2.0268242591894046,
      "grad_norm": 0.28945798184832394,
      "kl": 0.2611083984375,
      "learning_rate": 4.953657979310327e-07,
      "loss": 0.0003,
      "reward": 1.7125000730156898,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7169643230736256,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12088
    },
    {
      "completion_length": 217.47768783569336,
      "epoch": 2.0271595624292718,
      "grad_norm": 0.3697406638610568,
      "kl": 0.174652099609375,
      "learning_rate": 4.953634986530569e-07,
      "loss": 0.0002,
      "reward": 1.742857202887535,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571674972773,
      "rewards/format_reward_func": 1.0,
      "step": 12090
    },
    {
      "completion_length": 231.9598331451416,
      "epoch": 2.0274948656691394,
      "grad_norm": 0.19183056439734558,
      "kl": 0.171661376953125,
      "learning_rate": 4.953611988101633e-07,
      "loss": 0.0002,
      "reward": 1.7464286535978317,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 12092
    },
    {
      "completion_length": 223.97768783569336,
      "epoch": 2.027830168909007,
      "grad_norm": 0.2675128520542845,
      "kl": 0.1929931640625,
      "learning_rate": 4.953588984023573e-07,
      "loss": 0.0002,
      "reward": 1.7857143506407738,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7946428805589676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12094
    },
    {
      "completion_length": 225.57143783569336,
      "epoch": 2.0281654721488747,
      "grad_norm": 0.2959984170493704,
      "kl": 0.1334228515625,
      "learning_rate": 4.953565974296441e-07,
      "loss": 0.0001,
      "reward": 1.6714286804199219,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.6714286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 12096
    },
    {
      "completion_length": 216.91072463989258,
      "epoch": 2.0285007753887423,
      "grad_norm": 0.3055442488342892,
      "kl": 0.158233642578125,
      "learning_rate": 4.95354295892029e-07,
      "loss": 0.0002,
      "reward": 1.8035714998841286,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 12098
    },
    {
      "completion_length": 229.08483409881592,
      "epoch": 2.02883607862861,
      "grad_norm": 0.4134827691873043,
      "kl": 0.18817138671875,
      "learning_rate": 4.953519937895174e-07,
      "loss": 0.0002,
      "reward": 1.721428632736206,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7214286159723997,
      "rewards/format_reward_func": 1.0,
      "step": 12100
    },
    {
      "completion_length": 218.18304634094238,
      "epoch": 2.029171381868477,
      "grad_norm": 0.20157766541376793,
      "kl": 0.151275634765625,
      "learning_rate": 4.953496911221145e-07,
      "loss": 0.0002,
      "reward": 1.7642857804894447,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 12102
    },
    {
      "completion_length": 231.22768783569336,
      "epoch": 2.0295066851083448,
      "grad_norm": 0.2464018435854221,
      "kl": 0.17156982421875,
      "learning_rate": 4.953473878898258e-07,
      "loss": 0.0002,
      "reward": 1.810714341700077,
      "reward_std": 0.0656599123030901,
      "rewards/equation_reward_func": 0.8196428790688515,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12104
    },
    {
      "completion_length": 225.34375953674316,
      "epoch": 2.0298419883482124,
      "grad_norm": 0.18348432319917418,
      "kl": 0.15228271484375,
      "learning_rate": 4.953450840926563e-07,
      "loss": 0.0002,
      "reward": 1.7446429282426834,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7491071783006191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12106
    },
    {
      "completion_length": 228.602689743042,
      "epoch": 2.03017729158808,
      "grad_norm": 0.2064246228600626,
      "kl": 0.357513427734375,
      "learning_rate": 4.953427797306115e-07,
      "loss": 0.0004,
      "reward": 1.7803572043776512,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12108
    },
    {
      "completion_length": 236.37054634094238,
      "epoch": 2.0305125948279477,
      "grad_norm": 0.1549182499060302,
      "kl": 0.149200439453125,
      "learning_rate": 4.953404748036965e-07,
      "loss": 0.0001,
      "reward": 1.7553571909666061,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7598214782774448,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12110
    },
    {
      "completion_length": 237.93750953674316,
      "epoch": 2.0308478980678153,
      "grad_norm": 0.1820479868588014,
      "kl": 0.184326171875,
      "learning_rate": 4.953381693119169e-07,
      "loss": 0.0002,
      "reward": 1.74821437895298,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.752678606659174,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12112
    },
    {
      "completion_length": 236.83929824829102,
      "epoch": 2.0311832013076825,
      "grad_norm": 0.1271072757567212,
      "kl": 0.206817626953125,
      "learning_rate": 4.95335863255278e-07,
      "loss": 0.0002,
      "reward": 1.76071435213089,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7696428783237934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12114
    },
    {
      "completion_length": 237.74108409881592,
      "epoch": 2.03151850454755,
      "grad_norm": 0.30312186630702503,
      "kl": 0.359375,
      "learning_rate": 4.953335566337847e-07,
      "loss": 0.0004,
      "reward": 1.726785808801651,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7312500327825546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12116
    },
    {
      "completion_length": 237.87500953674316,
      "epoch": 2.0318538077874178,
      "grad_norm": 0.34231948362766396,
      "kl": 0.210174560546875,
      "learning_rate": 4.953312494474427e-07,
      "loss": 0.0002,
      "reward": 1.7683036476373672,
      "reward_std": 0.06502856919541955,
      "rewards/equation_reward_func": 0.7830357365310192,
      "rewards/format_reward_func": 0.9852678664028645,
      "step": 12118
    },
    {
      "completion_length": 247.51340293884277,
      "epoch": 2.0321891110272854,
      "grad_norm": 0.24182887804491468,
      "kl": 0.211639404296875,
      "learning_rate": 4.953289416962573e-07,
      "loss": 0.0002,
      "reward": 1.7375000789761543,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7419643215835094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12120
    },
    {
      "completion_length": 236.508939743042,
      "epoch": 2.032524414267153,
      "grad_norm": 0.18792396805814143,
      "kl": 0.15826416015625,
      "learning_rate": 4.953266333802336e-07,
      "loss": 0.0002,
      "reward": 1.7482143342494965,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7616071589291096,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 12122
    },
    {
      "completion_length": 234.99554634094238,
      "epoch": 2.03285971750702,
      "grad_norm": 0.23981471522060907,
      "kl": 0.146087646484375,
      "learning_rate": 4.953243244993771e-07,
      "loss": 0.0001,
      "reward": 1.7839286252856255,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7883928790688515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12124
    },
    {
      "completion_length": 230.20090293884277,
      "epoch": 2.033195020746888,
      "grad_norm": 0.5070715919217202,
      "kl": 0.173065185546875,
      "learning_rate": 4.953220150536931e-07,
      "loss": 0.0002,
      "reward": 1.8071429058909416,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.8071428947150707,
      "rewards/format_reward_func": 1.0,
      "step": 12126
    },
    {
      "completion_length": 235.46429538726807,
      "epoch": 2.0335303239867555,
      "grad_norm": 0.20365360632250812,
      "kl": 0.140228271484375,
      "learning_rate": 4.953197050431867e-07,
      "loss": 0.0001,
      "reward": 1.7625000551342964,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12128
    },
    {
      "completion_length": 228.5982255935669,
      "epoch": 2.033865627226623,
      "grad_norm": 0.3304221388416682,
      "kl": 0.15032958984375,
      "learning_rate": 4.953173944678635e-07,
      "loss": 0.0002,
      "reward": 1.8035714849829674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 12130
    },
    {
      "completion_length": 244.96429443359375,
      "epoch": 2.0342009304664908,
      "grad_norm": 0.48853927043600925,
      "kl": 0.200531005859375,
      "learning_rate": 4.953150833277286e-07,
      "loss": 0.0002,
      "reward": 1.7553571909666061,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.759821455925703,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12132
    },
    {
      "completion_length": 233.75000953674316,
      "epoch": 2.0345362337063584,
      "grad_norm": 0.5231974895770883,
      "kl": 0.131744384765625,
      "learning_rate": 4.953127716227875e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571939468384,
      "rewards/format_reward_func": 1.0,
      "step": 12134
    },
    {
      "completion_length": 221.73661708831787,
      "epoch": 2.0348715369462256,
      "grad_norm": 0.156355831696372,
      "kl": 0.13409423828125,
      "learning_rate": 4.953104593530455e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 12136
    },
    {
      "completion_length": 239.66072463989258,
      "epoch": 2.035206840186093,
      "grad_norm": 0.28143142463041965,
      "kl": 0.167388916015625,
      "learning_rate": 4.953081465185077e-07,
      "loss": 0.0002,
      "reward": 1.7321429178118706,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321428991854191,
      "rewards/format_reward_func": 1.0,
      "step": 12138
    },
    {
      "completion_length": 217.82590198516846,
      "epoch": 2.035542143425961,
      "grad_norm": 0.15327124267949874,
      "kl": 0.12646484375,
      "learning_rate": 4.953058331191797e-07,
      "loss": 0.0001,
      "reward": 1.7517857775092125,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500275671482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12140
    },
    {
      "completion_length": 220.13840198516846,
      "epoch": 2.0358774466658285,
      "grad_norm": 0.243204474048529,
      "kl": 0.13507080078125,
      "learning_rate": 4.953035191550667e-07,
      "loss": 0.0001,
      "reward": 1.775000050663948,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000413507223,
      "rewards/format_reward_func": 1.0,
      "step": 12142
    },
    {
      "completion_length": 240.28572845458984,
      "epoch": 2.036212749905696,
      "grad_norm": 0.3016617870258245,
      "kl": 0.14996337890625,
      "learning_rate": 4.95301204626174e-07,
      "loss": 0.0001,
      "reward": 1.800000049173832,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.800000011920929,
      "rewards/format_reward_func": 1.0,
      "step": 12144
    },
    {
      "completion_length": 238.83036613464355,
      "epoch": 2.0365480531455633,
      "grad_norm": 0.9705617575803153,
      "kl": 0.199249267578125,
      "learning_rate": 4.952988895325071e-07,
      "loss": 0.0002,
      "reward": 1.7464286461472511,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464285958558321,
      "rewards/format_reward_func": 1.0,
      "step": 12146
    },
    {
      "completion_length": 242.16518688201904,
      "epoch": 2.036883356385431,
      "grad_norm": 0.24141183406603373,
      "kl": 0.160888671875,
      "learning_rate": 4.952965738740712e-07,
      "loss": 0.0002,
      "reward": 1.764285758137703,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7642857581377029,
      "rewards/format_reward_func": 1.0,
      "step": 12148
    },
    {
      "completion_length": 222.35715198516846,
      "epoch": 2.0372186596252986,
      "grad_norm": 0.11416317860538366,
      "kl": 0.1640625,
      "learning_rate": 4.952942576508715e-07,
      "loss": 0.0002,
      "reward": 1.7892857566475868,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 12150
    },
    {
      "completion_length": 232.70536708831787,
      "epoch": 2.037553962865166,
      "grad_norm": 0.5464999282439347,
      "kl": 0.185272216796875,
      "learning_rate": 4.952919408629136e-07,
      "loss": 0.0002,
      "reward": 1.8321429267525673,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8321429006755352,
      "rewards/format_reward_func": 1.0,
      "step": 12152
    },
    {
      "completion_length": 231.72322463989258,
      "epoch": 2.037889266105034,
      "grad_norm": 0.19676459706372948,
      "kl": 0.166259765625,
      "learning_rate": 4.952896235102027e-07,
      "loss": 0.0002,
      "reward": 1.7178572118282318,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7178571820259094,
      "rewards/format_reward_func": 1.0,
      "step": 12154
    },
    {
      "completion_length": 233.97322463989258,
      "epoch": 2.0382245693449015,
      "grad_norm": 0.3356855438729618,
      "kl": 0.20611572265625,
      "learning_rate": 4.95287305592744e-07,
      "loss": 0.0002,
      "reward": 1.8214286267757416,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8214286006987095,
      "rewards/format_reward_func": 1.0,
      "step": 12156
    },
    {
      "completion_length": 238.13393783569336,
      "epoch": 2.0385598725847687,
      "grad_norm": 0.11742634321299134,
      "kl": 0.263336181640625,
      "learning_rate": 4.952849871105431e-07,
      "loss": 0.0003,
      "reward": 1.7575893327593803,
      "reward_std": 0.02967323106713593,
      "rewards/equation_reward_func": 0.7633928973227739,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 12158
    },
    {
      "completion_length": 234.12947273254395,
      "epoch": 2.0388951758246363,
      "grad_norm": 0.13895650350235308,
      "kl": 0.15374755859375,
      "learning_rate": 4.952826680636051e-07,
      "loss": 0.0002,
      "reward": 1.7785714864730835,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 12160
    },
    {
      "completion_length": 241.33483219146729,
      "epoch": 2.039230479064504,
      "grad_norm": 0.2290955370201332,
      "kl": 0.221832275390625,
      "learning_rate": 4.952803484519357e-07,
      "loss": 0.0002,
      "reward": 1.8214285895228386,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8214286044239998,
      "rewards/format_reward_func": 1.0,
      "step": 12162
    },
    {
      "completion_length": 248.62054824829102,
      "epoch": 2.0395657823043716,
      "grad_norm": 0.5380827733704721,
      "kl": 0.365081787109375,
      "learning_rate": 4.952780282755398e-07,
      "loss": 0.0004,
      "reward": 1.7196429148316383,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7241071630269289,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12164
    },
    {
      "completion_length": 244.63393783569336,
      "epoch": 2.039901085544239,
      "grad_norm": 0.21779016308612434,
      "kl": 0.276580810546875,
      "learning_rate": 4.95275707534423e-07,
      "loss": 0.0003,
      "reward": 1.785714365541935,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.785714328289032,
      "rewards/format_reward_func": 1.0,
      "step": 12166
    },
    {
      "completion_length": 237.87054347991943,
      "epoch": 2.0402363887841064,
      "grad_norm": 0.12616860556579165,
      "kl": 0.215911865234375,
      "learning_rate": 4.952733862285905e-07,
      "loss": 0.0002,
      "reward": 1.7821429297327995,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7910714633762836,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12168
    },
    {
      "completion_length": 262.1384057998657,
      "epoch": 2.040571692023974,
      "grad_norm": 0.19813592000547936,
      "kl": 0.26214599609375,
      "learning_rate": 4.952710643580478e-07,
      "loss": 0.0003,
      "reward": 1.7071429193019867,
      "reward_std": 0.08081220556050539,
      "rewards/equation_reward_func": 0.7250000312924385,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 12170
    },
    {
      "completion_length": 254.040189743042,
      "epoch": 2.0409069952638417,
      "grad_norm": 0.16061184685130958,
      "kl": 0.239654541015625,
      "learning_rate": 4.952687419228001e-07,
      "loss": 0.0002,
      "reward": 1.749553620815277,
      "reward_std": 0.06124049751088023,
      "rewards/equation_reward_func": 0.7616071701049805,
      "rewards/format_reward_func": 0.9879464358091354,
      "step": 12172
    },
    {
      "completion_length": 242.59375953674316,
      "epoch": 2.0412422985037093,
      "grad_norm": 0.2398218040558381,
      "kl": 0.1614990234375,
      "learning_rate": 4.952664189228529e-07,
      "loss": 0.0002,
      "reward": 1.7839286178350449,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12174
    },
    {
      "completion_length": 239.91965579986572,
      "epoch": 2.041577601743577,
      "grad_norm": 0.1620207173064198,
      "kl": 0.160888671875,
      "learning_rate": 4.952640953582114e-07,
      "loss": 0.0002,
      "reward": 1.7428572103381157,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571879863739,
      "rewards/format_reward_func": 1.0,
      "step": 12176
    },
    {
      "completion_length": 237.92858028411865,
      "epoch": 2.0419129049834446,
      "grad_norm": 0.21874752432843478,
      "kl": 0.17864990234375,
      "learning_rate": 4.95261771228881e-07,
      "loss": 0.0002,
      "reward": 1.767857201397419,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571827709675,
      "rewards/format_reward_func": 1.0,
      "step": 12178
    },
    {
      "completion_length": 251.3616180419922,
      "epoch": 2.0422482082233118,
      "grad_norm": 0.26590850068573935,
      "kl": 0.19781494140625,
      "learning_rate": 4.952594465348672e-07,
      "loss": 0.0002,
      "reward": 1.7482143491506577,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7526785936206579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12180
    },
    {
      "completion_length": 225.64286708831787,
      "epoch": 2.0425835114631794,
      "grad_norm": 0.20064903092848752,
      "kl": 0.1510009765625,
      "learning_rate": 4.95257121276175e-07,
      "loss": 0.0002,
      "reward": 1.8196429088711739,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.833035733550787,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 12182
    },
    {
      "completion_length": 246.2946538925171,
      "epoch": 2.042918814703047,
      "grad_norm": 0.2570804310225239,
      "kl": 0.225006103515625,
      "learning_rate": 4.952547954528101e-07,
      "loss": 0.0002,
      "reward": 1.7053572162985802,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7098214793950319,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12184
    },
    {
      "completion_length": 231.93304634094238,
      "epoch": 2.0432541179429147,
      "grad_norm": 0.24222254138144964,
      "kl": 0.16400146484375,
      "learning_rate": 4.952524690647778e-07,
      "loss": 0.0002,
      "reward": 1.7910714745521545,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7955357544124126,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12186
    },
    {
      "completion_length": 236.96876049041748,
      "epoch": 2.0435894211827823,
      "grad_norm": 0.24676297049278167,
      "kl": 0.234893798828125,
      "learning_rate": 4.952501421120832e-07,
      "loss": 0.0002,
      "reward": 1.7892857566475868,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.798214316368103,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12188
    },
    {
      "completion_length": 232.9687623977661,
      "epoch": 2.04392472442265,
      "grad_norm": 0.20240000669952848,
      "kl": 0.198699951171875,
      "learning_rate": 4.952478145947321e-07,
      "loss": 0.0002,
      "reward": 1.7482143566012383,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7616071663796902,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 12190
    },
    {
      "completion_length": 234.76340198516846,
      "epoch": 2.044260027662517,
      "grad_norm": 0.2367568215370992,
      "kl": 0.19390869140625,
      "learning_rate": 4.952454865127295e-07,
      "loss": 0.0002,
      "reward": 1.7642857879400253,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857320606709,
      "rewards/format_reward_func": 1.0,
      "step": 12192
    },
    {
      "completion_length": 226.4553680419922,
      "epoch": 2.0445953309023848,
      "grad_norm": 0.3957196761252101,
      "kl": 0.1790771484375,
      "learning_rate": 4.952431578660807e-07,
      "loss": 0.0002,
      "reward": 1.7803572118282318,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12194
    },
    {
      "completion_length": 238.0446548461914,
      "epoch": 2.0449306341422524,
      "grad_norm": 0.41923647210505366,
      "kl": 0.213134765625,
      "learning_rate": 4.952408286547913e-07,
      "loss": 0.0002,
      "reward": 1.753571480512619,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714767873287,
      "rewards/format_reward_func": 1.0,
      "step": 12196
    },
    {
      "completion_length": 230.87054824829102,
      "epoch": 2.04526593738212,
      "grad_norm": 0.16331890639992575,
      "kl": 0.154632568359375,
      "learning_rate": 4.952384988788666e-07,
      "loss": 0.0002,
      "reward": 1.6553572416305542,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.6598214600235224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12198
    },
    {
      "completion_length": 224.12500858306885,
      "epoch": 2.0456012406219877,
      "grad_norm": 0.4519803190382613,
      "kl": 0.20135498046875,
      "learning_rate": 4.95236168538312e-07,
      "loss": 0.0002,
      "reward": 1.7107143625617027,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7107143215835094,
      "rewards/format_reward_func": 1.0,
      "step": 12200
    },
    {
      "completion_length": 226.50000762939453,
      "epoch": 2.045936543861855,
      "grad_norm": 0.338038635536733,
      "kl": 0.194580078125,
      "learning_rate": 4.952338376331327e-07,
      "loss": 0.0002,
      "reward": 1.7857143431901932,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 12202
    },
    {
      "completion_length": 223.4464406967163,
      "epoch": 2.0462718471017225,
      "grad_norm": 0.10915545282494744,
      "kl": 0.13482666015625,
      "learning_rate": 4.952315061633343e-07,
      "loss": 0.0001,
      "reward": 1.8285714909434319,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8285714574158192,
      "rewards/format_reward_func": 1.0,
      "step": 12204
    },
    {
      "completion_length": 224.90179824829102,
      "epoch": 2.04660715034159,
      "grad_norm": 0.09504629135042848,
      "kl": 0.133575439453125,
      "learning_rate": 4.952291741289221e-07,
      "loss": 0.0001,
      "reward": 1.7767857760190964,
      "reward_std": 0.0328299580141902,
      "rewards/equation_reward_func": 0.781250037252903,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12206
    },
    {
      "completion_length": 225.60268878936768,
      "epoch": 2.0469424535814578,
      "grad_norm": 0.1865660206136097,
      "kl": 0.11761474609375,
      "learning_rate": 4.952268415299013e-07,
      "loss": 0.0001,
      "reward": 1.8250000402331352,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8250000104308128,
      "rewards/format_reward_func": 1.0,
      "step": 12208
    },
    {
      "completion_length": 228.60268878936768,
      "epoch": 2.0472777568213254,
      "grad_norm": 0.22629386598093604,
      "kl": 0.16473388671875,
      "learning_rate": 4.952245083662774e-07,
      "loss": 0.0002,
      "reward": 1.7535715252161026,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714507102966,
      "rewards/format_reward_func": 1.0,
      "step": 12210
    },
    {
      "completion_length": 230.33929634094238,
      "epoch": 2.047613060061193,
      "grad_norm": 0.27102016159457293,
      "kl": 0.1397705078125,
      "learning_rate": 4.952221746380557e-07,
      "loss": 0.0001,
      "reward": 1.760714367032051,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 12212
    },
    {
      "completion_length": 231.64286994934082,
      "epoch": 2.0479483633010602,
      "grad_norm": 0.29415841832577194,
      "kl": 0.14801025390625,
      "learning_rate": 4.952198403452417e-07,
      "loss": 0.0001,
      "reward": 1.7142857760190964,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.714285746216774,
      "rewards/format_reward_func": 1.0,
      "step": 12214
    },
    {
      "completion_length": 228.3616180419922,
      "epoch": 2.048283666540928,
      "grad_norm": 0.27608760539171945,
      "kl": 0.14324951171875,
      "learning_rate": 4.952175054878407e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 12216
    },
    {
      "completion_length": 221.0000057220459,
      "epoch": 2.0486189697807955,
      "grad_norm": 0.16831740463669606,
      "kl": 0.14044189453125,
      "learning_rate": 4.952151700658581e-07,
      "loss": 0.0001,
      "reward": 1.7767857611179352,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7812500298023224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12218
    },
    {
      "completion_length": 230.52233028411865,
      "epoch": 2.048954273020663,
      "grad_norm": 0.30209937383700664,
      "kl": 0.13238525390625,
      "learning_rate": 4.952128340792992e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857488244772,
      "rewards/format_reward_func": 1.0,
      "step": 12220
    },
    {
      "completion_length": 222.59375953674316,
      "epoch": 2.0492895762605308,
      "grad_norm": 0.3109690344702524,
      "kl": 0.151123046875,
      "learning_rate": 4.952104975281696e-07,
      "loss": 0.0002,
      "reward": 1.7535715252161026,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714693367481,
      "rewards/format_reward_func": 1.0,
      "step": 12222
    },
    {
      "completion_length": 232.5134038925171,
      "epoch": 2.049624879500398,
      "grad_norm": 0.2744549215687352,
      "kl": 0.1292724609375,
      "learning_rate": 4.952081604124743e-07,
      "loss": 0.0001,
      "reward": 1.6910715103149414,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.6955357603728771,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12224
    },
    {
      "completion_length": 235.00001049041748,
      "epoch": 2.0499601827402656,
      "grad_norm": 0.5163864123677877,
      "kl": 0.248382568359375,
      "learning_rate": 4.952058227322191e-07,
      "loss": 0.0002,
      "reward": 1.7910714820027351,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7955357506871223,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12226
    },
    {
      "completion_length": 220.38840293884277,
      "epoch": 2.0502954859801332,
      "grad_norm": 0.07724610768291482,
      "kl": 0.1690673828125,
      "learning_rate": 4.952034844874091e-07,
      "loss": 0.0002,
      "reward": 1.7964286133646965,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.796428594738245,
      "rewards/format_reward_func": 1.0,
      "step": 12228
    },
    {
      "completion_length": 242.69197463989258,
      "epoch": 2.050630789220001,
      "grad_norm": 0.16030103752196245,
      "kl": 0.132568359375,
      "learning_rate": 4.952011456780497e-07,
      "loss": 0.0001,
      "reward": 1.801785759627819,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8062500134110451,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12230
    },
    {
      "completion_length": 226.81251049041748,
      "epoch": 2.0509660924598685,
      "grad_norm": 0.17879382796345136,
      "kl": 0.123504638671875,
      "learning_rate": 4.951988063041464e-07,
      "loss": 0.0001,
      "reward": 1.7625000327825546,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12232
    },
    {
      "completion_length": 227.62500858306885,
      "epoch": 2.051301395699736,
      "grad_norm": 0.26533960090283815,
      "kl": 0.143218994140625,
      "learning_rate": 4.951964663657046e-07,
      "loss": 0.0001,
      "reward": 1.8035714849829674,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8035714700818062,
      "rewards/format_reward_func": 1.0,
      "step": 12234
    },
    {
      "completion_length": 235.46875953674316,
      "epoch": 2.0516366989396033,
      "grad_norm": 0.3753427425352448,
      "kl": 0.139190673828125,
      "learning_rate": 4.951941258627294e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 12236
    },
    {
      "completion_length": 230.11608028411865,
      "epoch": 2.051972002179471,
      "grad_norm": 0.35682220127991493,
      "kl": 0.129425048828125,
      "learning_rate": 4.951917847952266e-07,
      "loss": 0.0001,
      "reward": 1.7321429252624512,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321428693830967,
      "rewards/format_reward_func": 1.0,
      "step": 12238
    },
    {
      "completion_length": 222.79465198516846,
      "epoch": 2.0523073054193386,
      "grad_norm": 0.13520711911359942,
      "kl": 0.142059326171875,
      "learning_rate": 4.951894431632014e-07,
      "loss": 0.0001,
      "reward": 1.733928643167019,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7383928969502449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12240
    },
    {
      "completion_length": 218.290189743042,
      "epoch": 2.0526426086592062,
      "grad_norm": 0.13303137022439182,
      "kl": 0.121063232421875,
      "learning_rate": 4.951871009666591e-07,
      "loss": 0.0001,
      "reward": 1.7785715088248253,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 1.0,
      "step": 12242
    },
    {
      "completion_length": 223.2455472946167,
      "epoch": 2.052977911899074,
      "grad_norm": 0.2376218232090186,
      "kl": 0.28485107421875,
      "learning_rate": 4.951847582056053e-07,
      "loss": 0.0003,
      "reward": 1.8000000566244125,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 12244
    },
    {
      "completion_length": 207.01786708831787,
      "epoch": 2.0533132151389415,
      "grad_norm": 0.09265186879940798,
      "kl": 0.1336669921875,
      "learning_rate": 4.951824148800452e-07,
      "loss": 0.0001,
      "reward": 1.8071429058909416,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8071428723633289,
      "rewards/format_reward_func": 1.0,
      "step": 12246
    },
    {
      "completion_length": 219.26786613464355,
      "epoch": 2.0536485183788087,
      "grad_norm": 0.2203674172881168,
      "kl": 0.193115234375,
      "learning_rate": 4.951800709899843e-07,
      "loss": 0.0002,
      "reward": 1.7607143595814705,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 12248
    },
    {
      "completion_length": 219.40625762939453,
      "epoch": 2.0539838216186763,
      "grad_norm": 0.11510668979914045,
      "kl": 0.20257568359375,
      "learning_rate": 4.95177726535428e-07,
      "loss": 0.0002,
      "reward": 1.7500000819563866,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7500000447034836,
      "rewards/format_reward_func": 1.0,
      "step": 12250
    },
    {
      "completion_length": 217.38840198516846,
      "epoch": 2.054319124858544,
      "grad_norm": 0.16410520017656868,
      "kl": 0.1424102783203125,
      "learning_rate": 4.951753815163816e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 12252
    },
    {
      "completion_length": 211.48215198516846,
      "epoch": 2.0546544280984116,
      "grad_norm": 0.18802458515738016,
      "kl": 0.313232421875,
      "learning_rate": 4.951730359328507e-07,
      "loss": 0.0003,
      "reward": 1.7964286506175995,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.796428594738245,
      "rewards/format_reward_func": 1.0,
      "step": 12254
    },
    {
      "completion_length": 217.08483123779297,
      "epoch": 2.0549897313382792,
      "grad_norm": 0.3090721220893524,
      "kl": 0.201141357421875,
      "learning_rate": 4.951706897848404e-07,
      "loss": 0.0002,
      "reward": 1.7928571999073029,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 12256
    },
    {
      "completion_length": 214.23661518096924,
      "epoch": 2.0553250345781464,
      "grad_norm": 0.21218909376421832,
      "kl": 0.141326904296875,
      "learning_rate": 4.951683430723563e-07,
      "loss": 0.0001,
      "reward": 1.7642858102917671,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857357859612,
      "rewards/format_reward_func": 1.0,
      "step": 12258
    },
    {
      "completion_length": 216.852689743042,
      "epoch": 2.055660337818014,
      "grad_norm": 0.32296017850789377,
      "kl": 0.15814208984375,
      "learning_rate": 4.951659957954039e-07,
      "loss": 0.0002,
      "reward": 1.8017857819795609,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.8062500059604645,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12260
    },
    {
      "completion_length": 214.16072463989258,
      "epoch": 2.0559956410578817,
      "grad_norm": 0.5911557976325647,
      "kl": 0.276123046875,
      "learning_rate": 4.951636479539883e-07,
      "loss": 0.0003,
      "reward": 1.8267857432365417,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8312500342726707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12262
    },
    {
      "completion_length": 226.8125123977661,
      "epoch": 2.0563309442977493,
      "grad_norm": 0.3042559685781432,
      "kl": 0.15435791015625,
      "learning_rate": 4.951612995481152e-07,
      "loss": 0.0002,
      "reward": 1.7553572058677673,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7598214671015739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12264
    },
    {
      "completion_length": 213.71429634094238,
      "epoch": 2.056666247537617,
      "grad_norm": 0.2080678635967977,
      "kl": 0.33575439453125,
      "learning_rate": 4.951589505777899e-07,
      "loss": 0.0003,
      "reward": 1.7821428999304771,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 12266
    },
    {
      "completion_length": 217.05804443359375,
      "epoch": 2.0570015507774846,
      "grad_norm": 0.1594908542843629,
      "kl": 0.2147216796875,
      "learning_rate": 4.951566010430177e-07,
      "loss": 0.0002,
      "reward": 1.7607143372297287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143260538578,
      "rewards/format_reward_func": 1.0,
      "step": 12268
    },
    {
      "completion_length": 209.49554443359375,
      "epoch": 2.057336854017352,
      "grad_norm": 0.1559413315105943,
      "kl": 0.192535400390625,
      "learning_rate": 4.951542509438041e-07,
      "loss": 0.0002,
      "reward": 1.7589286342263222,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7633928954601288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12270
    },
    {
      "completion_length": 201.78125858306885,
      "epoch": 2.0576721572572194,
      "grad_norm": 0.271571152588572,
      "kl": 0.2626953125,
      "learning_rate": 4.951519002801546e-07,
      "loss": 0.0003,
      "reward": 1.7142858058214188,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7142857499420643,
      "rewards/format_reward_func": 1.0,
      "step": 12272
    },
    {
      "completion_length": 202.54911613464355,
      "epoch": 2.058007460497087,
      "grad_norm": 0.30856242436789477,
      "kl": 0.436859130859375,
      "learning_rate": 4.951495490520745e-07,
      "loss": 0.0004,
      "reward": 1.7571429386734962,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 12274
    },
    {
      "completion_length": 205.22322273254395,
      "epoch": 2.0583427637369547,
      "grad_norm": 0.21680426274120007,
      "kl": 0.1307373046875,
      "learning_rate": 4.951471972595694e-07,
      "loss": 0.0001,
      "reward": 1.800000049173832,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 12276
    },
    {
      "completion_length": 199.67858123779297,
      "epoch": 2.0586780669768223,
      "grad_norm": 0.10996703065060714,
      "kl": 0.391815185546875,
      "learning_rate": 4.951448449026443e-07,
      "loss": 0.0004,
      "reward": 1.7928571924567223,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571552038193,
      "rewards/format_reward_func": 1.0,
      "step": 12278
    },
    {
      "completion_length": 204.74108123779297,
      "epoch": 2.0590133702166895,
      "grad_norm": 0.09081324500665985,
      "kl": 1.105133056640625,
      "learning_rate": 4.95142491981305e-07,
      "loss": 0.0011,
      "reward": 1.8071429207921028,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8071428798139095,
      "rewards/format_reward_func": 1.0,
      "step": 12280
    },
    {
      "completion_length": 214.883939743042,
      "epoch": 2.059348673456557,
      "grad_norm": 0.34335413647230867,
      "kl": 0.147857666015625,
      "learning_rate": 4.951401384955568e-07,
      "loss": 0.0001,
      "reward": 1.7107143551111221,
      "reward_std": 0.035355339758098125,
      "rewards/equation_reward_func": 0.7196428906172514,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12282
    },
    {
      "completion_length": 211.67857933044434,
      "epoch": 2.059683976696425,
      "grad_norm": 0.42040633375445585,
      "kl": 0.171630859375,
      "learning_rate": 4.951377844454051e-07,
      "loss": 0.0002,
      "reward": 1.7160715013742447,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7205357477068901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12284
    },
    {
      "completion_length": 216.30804634094238,
      "epoch": 2.0600192799362924,
      "grad_norm": 0.03826716338430946,
      "kl": 1.191925048828125,
      "learning_rate": 4.951354298308554e-07,
      "loss": 0.0012,
      "reward": 1.7892857789993286,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.789285734295845,
      "rewards/format_reward_func": 1.0,
      "step": 12286
    },
    {
      "completion_length": 212.11161613464355,
      "epoch": 2.06035458317616,
      "grad_norm": 0.3257955722546039,
      "kl": 0.256622314453125,
      "learning_rate": 4.951330746519129e-07,
      "loss": 0.0003,
      "reward": 1.7892857789993286,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 12288
    },
    {
      "completion_length": 211.2232255935669,
      "epoch": 2.0606898864160277,
      "grad_norm": 0.1655865158957351,
      "kl": 0.50634765625,
      "learning_rate": 4.951307189085833e-07,
      "loss": 0.0005,
      "reward": 1.7321429178118706,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321428973227739,
      "rewards/format_reward_func": 1.0,
      "step": 12290
    },
    {
      "completion_length": 216.52679443359375,
      "epoch": 2.061025189655895,
      "grad_norm": 0.33884942243688243,
      "kl": 0.244232177734375,
      "learning_rate": 4.951283626008717e-07,
      "loss": 0.0002,
      "reward": 1.7321429401636124,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7321428917348385,
      "rewards/format_reward_func": 1.0,
      "step": 12292
    },
    {
      "completion_length": 208.66072463989258,
      "epoch": 2.0613604928957625,
      "grad_norm": 0.24913338205065425,
      "kl": 0.36474609375,
      "learning_rate": 4.951260057287839e-07,
      "loss": 0.0004,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 12294
    },
    {
      "completion_length": 208.3616180419922,
      "epoch": 2.06169579613563,
      "grad_norm": 0.19352248282013856,
      "kl": 0.12322998046875,
      "learning_rate": 4.951236482923252e-07,
      "loss": 0.0001,
      "reward": 1.7892857939004898,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857566475868,
      "rewards/format_reward_func": 1.0,
      "step": 12296
    },
    {
      "completion_length": 208.8303689956665,
      "epoch": 2.062031099375498,
      "grad_norm": 0.08871718127407249,
      "kl": 0.1153717041015625,
      "learning_rate": 4.951212902915009e-07,
      "loss": 0.0001,
      "reward": 1.8196429163217545,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8241071626543999,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12298
    },
    {
      "completion_length": 227.90625762939453,
      "epoch": 2.0623664026153654,
      "grad_norm": 0.1901467761656163,
      "kl": 0.38531494140625,
      "learning_rate": 4.951189317263164e-07,
      "loss": 0.0004,
      "reward": 1.7214286625385284,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7303571626543999,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12300
    },
    {
      "completion_length": 209.73661518096924,
      "epoch": 2.0627017058552326,
      "grad_norm": 0.2980171673664818,
      "kl": 0.2169189453125,
      "learning_rate": 4.951165725967774e-07,
      "loss": 0.0002,
      "reward": 1.767410784959793,
      "reward_std": 0.0460882093757391,
      "rewards/equation_reward_func": 0.7705357521772385,
      "rewards/format_reward_func": 0.9968750067055225,
      "step": 12302
    },
    {
      "completion_length": 205.17411518096924,
      "epoch": 2.0630370090951002,
      "grad_norm": 0.12847052840527082,
      "kl": 0.134552001953125,
      "learning_rate": 4.951142129028891e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 12304
    },
    {
      "completion_length": 215.54018783569336,
      "epoch": 2.063372312334968,
      "grad_norm": 0.1620875253578144,
      "kl": 0.37359619140625,
      "learning_rate": 4.951118526446569e-07,
      "loss": 0.0004,
      "reward": 1.7714286372065544,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 12306
    },
    {
      "completion_length": 225.3259038925171,
      "epoch": 2.0637076155748355,
      "grad_norm": 0.24326793036443256,
      "kl": 0.216827392578125,
      "learning_rate": 4.951094918220865e-07,
      "loss": 0.0002,
      "reward": 1.814285770058632,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8142857328057289,
      "rewards/format_reward_func": 1.0,
      "step": 12308
    },
    {
      "completion_length": 211.29018878936768,
      "epoch": 2.064042918814703,
      "grad_norm": 0.16845783070573056,
      "kl": 0.169158935546875,
      "learning_rate": 4.951071304351831e-07,
      "loss": 0.0002,
      "reward": 1.7928571999073029,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 12310
    },
    {
      "completion_length": 205.84822368621826,
      "epoch": 2.0643782220545708,
      "grad_norm": 0.4214904190744296,
      "kl": 0.1233978271484375,
      "learning_rate": 4.951047684839522e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714626312256,
      "rewards/format_reward_func": 1.0,
      "step": 12312
    },
    {
      "completion_length": 210.79465007781982,
      "epoch": 2.064713525294438,
      "grad_norm": 0.21442170307890102,
      "kl": 0.15289306640625,
      "learning_rate": 4.951024059683993e-07,
      "loss": 0.0002,
      "reward": 1.7580357789993286,
      "reward_std": 0.059346459805965424,
      "rewards/equation_reward_func": 0.7598214745521545,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 12314
    },
    {
      "completion_length": 211.85715103149414,
      "epoch": 2.0650488285343056,
      "grad_norm": 0.3091366156084564,
      "kl": 0.134857177734375,
      "learning_rate": 4.951000428885297e-07,
      "loss": 0.0001,
      "reward": 1.844642885029316,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.849107164889574,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12316
    },
    {
      "completion_length": 212.8571538925171,
      "epoch": 2.0653841317741732,
      "grad_norm": 0.45210288819457994,
      "kl": 0.677154541015625,
      "learning_rate": 4.95097679244349e-07,
      "loss": 0.0007,
      "reward": 1.7339286506175995,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7383928894996643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12318
    },
    {
      "completion_length": 227.17411708831787,
      "epoch": 2.065719435014041,
      "grad_norm": 0.2152976862107728,
      "kl": 0.2509307861328125,
      "learning_rate": 4.950953150358625e-07,
      "loss": 0.0003,
      "reward": 1.8232143223285675,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.8276785910129547,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12320
    },
    {
      "completion_length": 227.540189743042,
      "epoch": 2.0660547382539085,
      "grad_norm": 0.5459242095577989,
      "kl": 0.271026611328125,
      "learning_rate": 4.950929502630757e-07,
      "loss": 0.0003,
      "reward": 1.782142922282219,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 12322
    },
    {
      "completion_length": 220.99554347991943,
      "epoch": 2.066390041493776,
      "grad_norm": 0.425233638393935,
      "kl": 0.907958984375,
      "learning_rate": 4.950905849259942e-07,
      "loss": 0.0009,
      "reward": 1.7821429446339607,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 12324
    },
    {
      "completion_length": 234.6250114440918,
      "epoch": 2.0667253447336433,
      "grad_norm": 0.3270812138751066,
      "kl": 0.9796600341796875,
      "learning_rate": 4.950882190246232e-07,
      "loss": 0.001,
      "reward": 1.6357143744826317,
      "reward_std": 0.11111677903681993,
      "rewards/equation_reward_func": 0.6625000331550837,
      "rewards/format_reward_func": 0.9732142984867096,
      "step": 12326
    },
    {
      "completion_length": 221.82143878936768,
      "epoch": 2.067060647973511,
      "grad_norm": 0.2345927392282807,
      "kl": 0.71343994140625,
      "learning_rate": 4.950858525589682e-07,
      "loss": 0.0007,
      "reward": 1.7857143357396126,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7946428917348385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12328
    },
    {
      "completion_length": 222.04018783569336,
      "epoch": 2.0673959512133786,
      "grad_norm": 0.1827131215605797,
      "kl": 0.113525390625,
      "learning_rate": 4.950834855290347e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 12330
    },
    {
      "completion_length": 212.12500953674316,
      "epoch": 2.0677312544532462,
      "grad_norm": 0.17331603165530027,
      "kl": 0.10479736328125,
      "learning_rate": 4.950811179348282e-07,
      "loss": 0.0001,
      "reward": 1.8535714596509933,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8535714522004128,
      "rewards/format_reward_func": 1.0,
      "step": 12332
    },
    {
      "completion_length": 226.7901906967163,
      "epoch": 2.068066557693114,
      "grad_norm": 0.2783940467018383,
      "kl": 2.2259521484375,
      "learning_rate": 4.950787497763541e-07,
      "loss": 0.0022,
      "reward": 1.7232143580913544,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7357143126428127,
      "rewards/format_reward_func": 0.9875000044703484,
      "step": 12334
    },
    {
      "completion_length": 220.68751049041748,
      "epoch": 2.068401860932981,
      "grad_norm": 0.2831385527880495,
      "kl": 0.7620086669921875,
      "learning_rate": 4.950763810536178e-07,
      "loss": 0.0008,
      "reward": 1.8035714998841286,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 12336
    },
    {
      "completion_length": 225.80804634094238,
      "epoch": 2.0687371641728487,
      "grad_norm": 0.2587481213808961,
      "kl": 2.004638671875,
      "learning_rate": 4.950740117666248e-07,
      "loss": 0.002,
      "reward": 1.7678572311997414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 12338
    },
    {
      "completion_length": 225.87501049041748,
      "epoch": 2.0690724674127163,
      "grad_norm": 0.14320682134950544,
      "kl": 0.146728515625,
      "learning_rate": 4.950716419153806e-07,
      "loss": 0.0001,
      "reward": 1.7892857939004898,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 12340
    },
    {
      "completion_length": 226.38840198516846,
      "epoch": 2.069407770652584,
      "grad_norm": 0.1876906013637508,
      "kl": 0.278961181640625,
      "learning_rate": 4.950692714998906e-07,
      "loss": 0.0003,
      "reward": 1.7241072282195091,
      "reward_std": 0.06692260596901178,
      "rewards/equation_reward_func": 0.7303571663796902,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 12342
    },
    {
      "completion_length": 234.2946538925171,
      "epoch": 2.0697430738924516,
      "grad_norm": 0.37498717092004047,
      "kl": 0.311004638671875,
      "learning_rate": 4.950669005201603e-07,
      "loss": 0.0003,
      "reward": 1.716071493923664,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7205357626080513,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12344
    },
    {
      "completion_length": 218.04465579986572,
      "epoch": 2.0700783771323192,
      "grad_norm": 0.18627335088277233,
      "kl": 0.1470947265625,
      "learning_rate": 4.950645289761952e-07,
      "loss": 0.0001,
      "reward": 1.8250000476837158,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8250000178813934,
      "rewards/format_reward_func": 1.0,
      "step": 12346
    },
    {
      "completion_length": 228.9017972946167,
      "epoch": 2.0704136803721864,
      "grad_norm": 0.3317378232716388,
      "kl": 0.1026153564453125,
      "learning_rate": 4.950621568680006e-07,
      "loss": 0.0001,
      "reward": 1.7625000551342964,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12348
    },
    {
      "completion_length": 215.88840198516846,
      "epoch": 2.070748983612054,
      "grad_norm": 0.13501275717490083,
      "kl": 0.2255859375,
      "learning_rate": 4.950597841955821e-07,
      "loss": 0.0002,
      "reward": 1.767857201397419,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.7767857350409031,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12350
    },
    {
      "completion_length": 223.21429443359375,
      "epoch": 2.0710842868519217,
      "grad_norm": 0.1918316612561862,
      "kl": 0.6068115234375,
      "learning_rate": 4.95057410958945e-07,
      "loss": 0.0006,
      "reward": 1.7535714954137802,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 12352
    },
    {
      "completion_length": 222.90179538726807,
      "epoch": 2.0714195900917893,
      "grad_norm": 0.28909041151334275,
      "kl": 0.0965576171875,
      "learning_rate": 4.95055037158095e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857428640127,
      "rewards/format_reward_func": 1.0,
      "step": 12354
    },
    {
      "completion_length": 232.17411708831787,
      "epoch": 2.071754893331657,
      "grad_norm": 0.23212451175861568,
      "kl": 0.3805084228515625,
      "learning_rate": 4.950526627930374e-07,
      "loss": 0.0004,
      "reward": 1.7410715073347092,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7455357536673546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12356
    },
    {
      "completion_length": 229.99108028411865,
      "epoch": 2.0720901965715246,
      "grad_norm": 0.1909832712879006,
      "kl": 0.110687255859375,
      "learning_rate": 4.950502878637776e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 1.0,
      "step": 12358
    },
    {
      "completion_length": 220.53572463989258,
      "epoch": 2.072425499811392,
      "grad_norm": 0.10395956700858099,
      "kl": 0.14569091796875,
      "learning_rate": 4.950479123703213e-07,
      "loss": 0.0001,
      "reward": 1.807142898440361,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8071428742259741,
      "rewards/format_reward_func": 1.0,
      "step": 12360
    },
    {
      "completion_length": 219.85715103149414,
      "epoch": 2.0727608030512594,
      "grad_norm": 0.18317384805951137,
      "kl": 0.105224609375,
      "learning_rate": 4.950455363126739e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428868919611,
      "rewards/format_reward_func": 1.0,
      "step": 12362
    },
    {
      "completion_length": 223.12500953674316,
      "epoch": 2.073096106291127,
      "grad_norm": 0.1526248796039914,
      "kl": 0.239898681640625,
      "learning_rate": 4.950431596908408e-07,
      "loss": 0.0002,
      "reward": 1.7571429312229156,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428846567869,
      "rewards/format_reward_func": 1.0,
      "step": 12364
    },
    {
      "completion_length": 220.6250114440918,
      "epoch": 2.0734314095309947,
      "grad_norm": 0.25779972652365757,
      "kl": 0.1161346435546875,
      "learning_rate": 4.950407825048273e-07,
      "loss": 0.0001,
      "reward": 1.791071467101574,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7955357432365417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12366
    },
    {
      "completion_length": 214.35268688201904,
      "epoch": 2.0737667127708623,
      "grad_norm": 0.19846381452816875,
      "kl": 0.09454345703125,
      "learning_rate": 4.950384047546393e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 12368
    },
    {
      "completion_length": 220.6384038925171,
      "epoch": 2.0741020160107295,
      "grad_norm": 0.21754462554569812,
      "kl": 0.136505126953125,
      "learning_rate": 4.95036026440282e-07,
      "loss": 0.0001,
      "reward": 1.751785770058632,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7562500238418579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12370
    },
    {
      "completion_length": 225.25001049041748,
      "epoch": 2.074437319250597,
      "grad_norm": 0.18328997328817917,
      "kl": 0.206512451171875,
      "learning_rate": 4.950336475617608e-07,
      "loss": 0.0002,
      "reward": 1.7696429193019867,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7741071656346321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12372
    },
    {
      "completion_length": 221.01340293884277,
      "epoch": 2.074772622490465,
      "grad_norm": 0.2382505115459546,
      "kl": 0.163055419921875,
      "learning_rate": 4.950312681190813e-07,
      "loss": 0.0002,
      "reward": 1.7660714983940125,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.770535733550787,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12374
    },
    {
      "completion_length": 221.47322368621826,
      "epoch": 2.0751079257303324,
      "grad_norm": 0.36282886473131615,
      "kl": 0.253021240234375,
      "learning_rate": 4.950288881122491e-07,
      "loss": 0.0003,
      "reward": 1.8071429058909416,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428742259741,
      "rewards/format_reward_func": 1.0,
      "step": 12376
    },
    {
      "completion_length": 219.15179634094238,
      "epoch": 2.0754432289702,
      "grad_norm": 0.06999156876761552,
      "kl": 0.363494873046875,
      "learning_rate": 4.950265075412694e-07,
      "loss": 0.0004,
      "reward": 1.7750000581145287,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000171363354,
      "rewards/format_reward_func": 1.0,
      "step": 12378
    },
    {
      "completion_length": 207.70536613464355,
      "epoch": 2.0757785322100677,
      "grad_norm": 0.21250234133357188,
      "kl": 0.13916015625,
      "learning_rate": 4.95024126406148e-07,
      "loss": 0.0001,
      "reward": 1.817857176065445,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.817857176065445,
      "rewards/format_reward_func": 1.0,
      "step": 12380
    },
    {
      "completion_length": 221.18750953674316,
      "epoch": 2.076113835449935,
      "grad_norm": 0.21523313575949132,
      "kl": 0.14617919921875,
      "learning_rate": 4.950217447068901e-07,
      "loss": 0.0001,
      "reward": 1.7160715162754059,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.720535758882761,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12382
    },
    {
      "completion_length": 211.87500953674316,
      "epoch": 2.0764491386898025,
      "grad_norm": 0.1939850515633133,
      "kl": 0.119110107421875,
      "learning_rate": 4.950193624435013e-07,
      "loss": 0.0001,
      "reward": 1.7500000521540642,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000242143869,
      "rewards/format_reward_func": 1.0,
      "step": 12384
    },
    {
      "completion_length": 220.07590198516846,
      "epoch": 2.07678444192967,
      "grad_norm": 0.23250932473523991,
      "kl": 0.113616943359375,
      "learning_rate": 4.950169796159871e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 12386
    },
    {
      "completion_length": 217.35715293884277,
      "epoch": 2.077119745169538,
      "grad_norm": 0.21520081391524276,
      "kl": 0.161041259765625,
      "learning_rate": 4.950145962243529e-07,
      "loss": 0.0002,
      "reward": 1.785714328289032,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 12388
    },
    {
      "completion_length": 223.40179538726807,
      "epoch": 2.0774550484094054,
      "grad_norm": 0.751735218253887,
      "kl": 1.289947509765625,
      "learning_rate": 4.950122122686044e-07,
      "loss": 0.0013,
      "reward": 1.792857177555561,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571589291096,
      "rewards/format_reward_func": 1.0,
      "step": 12390
    },
    {
      "completion_length": 220.88840293884277,
      "epoch": 2.0777903516492726,
      "grad_norm": 0.2627694441752578,
      "kl": 0.12933349609375,
      "learning_rate": 4.950098277487468e-07,
      "loss": 0.0001,
      "reward": 1.7500000894069672,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 12392
    },
    {
      "completion_length": 217.602689743042,
      "epoch": 2.0781256548891403,
      "grad_norm": 0.09369122758849056,
      "kl": 0.142425537109375,
      "learning_rate": 4.950074426647858e-07,
      "loss": 0.0001,
      "reward": 1.7303572297096252,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7348214723169804,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12394
    },
    {
      "completion_length": 213.12947368621826,
      "epoch": 2.078460958129008,
      "grad_norm": 0.22262652267705993,
      "kl": 0.17474365234375,
      "learning_rate": 4.950050570167268e-07,
      "loss": 0.0002,
      "reward": 1.8000000640749931,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 12396
    },
    {
      "completion_length": 216.79911518096924,
      "epoch": 2.0787962613688755,
      "grad_norm": 0.252928583623105,
      "kl": 0.204071044921875,
      "learning_rate": 4.950026708045754e-07,
      "loss": 0.0002,
      "reward": 1.7803572416305542,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7848214581608772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12398
    },
    {
      "completion_length": 222.32590293884277,
      "epoch": 2.079131564608743,
      "grad_norm": 0.1992977674304534,
      "kl": 0.112945556640625,
      "learning_rate": 4.95000284028337e-07,
      "loss": 0.0001,
      "reward": 1.756696492433548,
      "reward_std": 0.04103744635358453,
      "rewards/equation_reward_func": 0.7598214615136385,
      "rewards/format_reward_func": 0.9968750029802322,
      "step": 12400
    },
    {
      "completion_length": 229.6071538925171,
      "epoch": 2.079466867848611,
      "grad_norm": 0.8789151158222025,
      "kl": 0.360076904296875,
      "learning_rate": 4.94997896688017e-07,
      "loss": 0.0004,
      "reward": 1.7660714909434319,
      "reward_std": 0.07828682195395231,
      "rewards/equation_reward_func": 0.7794643081724644,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 12402
    },
    {
      "completion_length": 228.88840103149414,
      "epoch": 2.079802171088478,
      "grad_norm": 0.1887117558689548,
      "kl": 0.135833740234375,
      "learning_rate": 4.94995508783621e-07,
      "loss": 0.0001,
      "reward": 1.7321429401636124,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7321428842842579,
      "rewards/format_reward_func": 1.0,
      "step": 12404
    },
    {
      "completion_length": 232.11608409881592,
      "epoch": 2.0801374743283456,
      "grad_norm": 0.19187944328862966,
      "kl": 0.108917236328125,
      "learning_rate": 4.949931203151547e-07,
      "loss": 0.0001,
      "reward": 1.7375000640749931,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7419643066823483,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12406
    },
    {
      "completion_length": 233.0803689956665,
      "epoch": 2.0804727775682132,
      "grad_norm": 0.21875754778227985,
      "kl": 0.12646484375,
      "learning_rate": 4.949907312826231e-07,
      "loss": 0.0001,
      "reward": 1.7625000551342964,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7758928798139095,
      "rewards/format_reward_func": 0.9866071455180645,
      "step": 12408
    },
    {
      "completion_length": 221.72768688201904,
      "epoch": 2.080808080808081,
      "grad_norm": 0.28006826916314953,
      "kl": 0.11962890625,
      "learning_rate": 4.949883416860322e-07,
      "loss": 0.0001,
      "reward": 1.7803571969270706,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7848214581608772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12410
    },
    {
      "completion_length": 231.99108028411865,
      "epoch": 2.0811433840479485,
      "grad_norm": 0.10164294385984807,
      "kl": 0.131683349609375,
      "learning_rate": 4.949859515253873e-07,
      "loss": 0.0001,
      "reward": 1.7607143372297287,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7607143260538578,
      "rewards/format_reward_func": 1.0,
      "step": 12412
    },
    {
      "completion_length": 226.0044755935669,
      "epoch": 2.0814786872878157,
      "grad_norm": 0.2045897020362465,
      "kl": 0.165740966796875,
      "learning_rate": 4.949835608006939e-07,
      "loss": 0.0002,
      "reward": 1.7821429073810577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 12414
    },
    {
      "completion_length": 236.52679443359375,
      "epoch": 2.0818139905276833,
      "grad_norm": 0.16084116411428384,
      "kl": 0.12066650390625,
      "learning_rate": 4.949811695119574e-07,
      "loss": 0.0001,
      "reward": 1.7428572177886963,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7517857365310192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12416
    },
    {
      "completion_length": 232.37947845458984,
      "epoch": 2.082149293767551,
      "grad_norm": 0.21511452457931682,
      "kl": 0.130126953125,
      "learning_rate": 4.949787776591836e-07,
      "loss": 0.0001,
      "reward": 1.7767857387661934,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7812500391155481,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12418
    },
    {
      "completion_length": 229.25447463989258,
      "epoch": 2.0824845970074186,
      "grad_norm": 0.21129750414378579,
      "kl": 0.1319732666015625,
      "learning_rate": 4.949763852423776e-07,
      "loss": 0.0001,
      "reward": 1.735714353621006,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143312692642,
      "rewards/format_reward_func": 1.0,
      "step": 12420
    },
    {
      "completion_length": 237.6964406967163,
      "epoch": 2.0828199002472862,
      "grad_norm": 0.40663412421997575,
      "kl": 0.1898193359375,
      "learning_rate": 4.949739922615454e-07,
      "loss": 0.0002,
      "reward": 1.7589286863803864,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7633928768336773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12422
    },
    {
      "completion_length": 230.7946538925171,
      "epoch": 2.083155203487154,
      "grad_norm": 0.16470876741545637,
      "kl": 0.1302337646484375,
      "learning_rate": 4.949715987166921e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7839285936206579,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12424
    },
    {
      "completion_length": 230.53572368621826,
      "epoch": 2.083490506727021,
      "grad_norm": 0.15797828640324696,
      "kl": 0.15777587890625,
      "learning_rate": 4.949692046078232e-07,
      "loss": 0.0002,
      "reward": 1.7857143580913544,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 12426
    },
    {
      "completion_length": 237.6875114440918,
      "epoch": 2.0838258099668887,
      "grad_norm": 0.22081453106368287,
      "kl": 0.161590576171875,
      "learning_rate": 4.949668099349446e-07,
      "loss": 0.0002,
      "reward": 1.782142922282219,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 12428
    },
    {
      "completion_length": 228.06250858306885,
      "epoch": 2.0841611132067563,
      "grad_norm": 0.22329552621134163,
      "kl": 0.1348876953125,
      "learning_rate": 4.949644146980615e-07,
      "loss": 0.0001,
      "reward": 1.7464286163449287,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 1.0,
      "step": 12430
    },
    {
      "completion_length": 230.2991180419922,
      "epoch": 2.084496416446624,
      "grad_norm": 0.22023536636649677,
      "kl": 0.164520263671875,
      "learning_rate": 4.949620188971794e-07,
      "loss": 0.0002,
      "reward": 1.764285795390606,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7732143178582191,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12432
    },
    {
      "completion_length": 224.04018878936768,
      "epoch": 2.0848317196864916,
      "grad_norm": 0.17440609622680234,
      "kl": 0.1301422119140625,
      "learning_rate": 4.94959622532304e-07,
      "loss": 0.0001,
      "reward": 1.8142857775092125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857402563095,
      "rewards/format_reward_func": 1.0,
      "step": 12434
    },
    {
      "completion_length": 232.14733219146729,
      "epoch": 2.085167022926359,
      "grad_norm": 0.17576071552016947,
      "kl": 0.13531494140625,
      "learning_rate": 4.949572256034406e-07,
      "loss": 0.0001,
      "reward": 1.783928632736206,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7883928753435612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12436
    },
    {
      "completion_length": 232.62947463989258,
      "epoch": 2.0855023261662264,
      "grad_norm": 0.19723789067097333,
      "kl": 0.14813232421875,
      "learning_rate": 4.94954828110595e-07,
      "loss": 0.0001,
      "reward": 1.7785714715719223,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7875000219792128,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12438
    },
    {
      "completion_length": 235.81697463989258,
      "epoch": 2.085837629406094,
      "grad_norm": 0.14950628149768866,
      "kl": 0.1369171142578125,
      "learning_rate": 4.949524300537726e-07,
      "loss": 0.0001,
      "reward": 1.7500000596046448,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7589286081492901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12440
    },
    {
      "completion_length": 230.64286708831787,
      "epoch": 2.0861729326459617,
      "grad_norm": 0.4378233230600497,
      "kl": 0.23809814453125,
      "learning_rate": 4.949500314329787e-07,
      "loss": 0.0002,
      "reward": 1.7089286223053932,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7133928835391998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12442
    },
    {
      "completion_length": 222.9107255935669,
      "epoch": 2.0865082358858293,
      "grad_norm": 0.46352014756844934,
      "kl": 0.178680419921875,
      "learning_rate": 4.949476322482191e-07,
      "loss": 0.0002,
      "reward": 1.782142922282219,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428775787354,
      "rewards/format_reward_func": 1.0,
      "step": 12444
    },
    {
      "completion_length": 215.74554443359375,
      "epoch": 2.086843539125697,
      "grad_norm": 0.24448623383945575,
      "kl": 0.13616943359375,
      "learning_rate": 4.949452324994992e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143081724644,
      "rewards/format_reward_func": 1.0,
      "step": 12446
    },
    {
      "completion_length": 226.6250123977661,
      "epoch": 2.087178842365564,
      "grad_norm": 0.2482961130975891,
      "kl": 0.15899658203125,
      "learning_rate": 4.949428321868246e-07,
      "loss": 0.0002,
      "reward": 1.7714286372065544,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 12448
    },
    {
      "completion_length": 232.67858409881592,
      "epoch": 2.087514145605432,
      "grad_norm": 0.31050673652855626,
      "kl": 0.240020751953125,
      "learning_rate": 4.949404313102008e-07,
      "loss": 0.0002,
      "reward": 1.7589286416769028,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7633928842842579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12450
    },
    {
      "completion_length": 234.43304443359375,
      "epoch": 2.0878494488452994,
      "grad_norm": 0.26036864450294417,
      "kl": 0.2906341552734375,
      "learning_rate": 4.949380298696333e-07,
      "loss": 0.0003,
      "reward": 1.7723214700818062,
      "reward_std": 0.049244935624301434,
      "rewards/equation_reward_func": 0.774107176810503,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 12452
    },
    {
      "completion_length": 229.3616180419922,
      "epoch": 2.088184752085167,
      "grad_norm": 0.09615753218038946,
      "kl": 0.14678955078125,
      "learning_rate": 4.949356278651277e-07,
      "loss": 0.0001,
      "reward": 1.74642863124609,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.746428582817316,
      "rewards/format_reward_func": 1.0,
      "step": 12454
    },
    {
      "completion_length": 237.2232255935669,
      "epoch": 2.0885200553250347,
      "grad_norm": 0.24540628173182102,
      "kl": 0.163970947265625,
      "learning_rate": 4.949332252966893e-07,
      "loss": 0.0002,
      "reward": 1.750000074505806,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7589285969734192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12456
    },
    {
      "completion_length": 221.55804538726807,
      "epoch": 2.0888553585649023,
      "grad_norm": 0.18356936840482607,
      "kl": 0.1224517822265625,
      "learning_rate": 4.949308221643239e-07,
      "loss": 0.0001,
      "reward": 1.731250062584877,
      "reward_std": 0.056821079924702644,
      "rewards/equation_reward_func": 0.7375000398606062,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 12458
    },
    {
      "completion_length": 229.7009038925171,
      "epoch": 2.0891906618047695,
      "grad_norm": 0.4034373111510341,
      "kl": 0.2835540771484375,
      "learning_rate": 4.949284184680369e-07,
      "loss": 0.0003,
      "reward": 1.8232143372297287,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8276785835623741,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12460
    },
    {
      "completion_length": 220.22768878936768,
      "epoch": 2.089525965044637,
      "grad_norm": 0.1536477528310614,
      "kl": 0.1710205078125,
      "learning_rate": 4.949260142078339e-07,
      "loss": 0.0002,
      "reward": 1.8178571909666061,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.817857164889574,
      "rewards/format_reward_func": 1.0,
      "step": 12462
    },
    {
      "completion_length": 231.63393878936768,
      "epoch": 2.089861268284505,
      "grad_norm": 0.244357653418181,
      "kl": 0.19512939453125,
      "learning_rate": 4.949236093837204e-07,
      "loss": 0.0002,
      "reward": 1.8071429133415222,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8071428835391998,
      "rewards/format_reward_func": 1.0,
      "step": 12464
    },
    {
      "completion_length": 225.87054538726807,
      "epoch": 2.0901965715243724,
      "grad_norm": 0.18870748260664152,
      "kl": 0.23162841796875,
      "learning_rate": 4.949212039957019e-07,
      "loss": 0.0002,
      "reward": 1.7629465013742447,
      "reward_std": 0.05240166233852506,
      "rewards/equation_reward_func": 0.764285733923316,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 12466
    },
    {
      "completion_length": 216.29911613464355,
      "epoch": 2.09053187476424,
      "grad_norm": 0.19688421926182828,
      "kl": 0.33929443359375,
      "learning_rate": 4.949187980437841e-07,
      "loss": 0.0003,
      "reward": 1.7857143506407738,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 12468
    },
    {
      "completion_length": 222.70983219146729,
      "epoch": 2.0908671780041073,
      "grad_norm": 0.5548509345150273,
      "kl": 0.38800048828125,
      "learning_rate": 4.949163915279722e-07,
      "loss": 0.0004,
      "reward": 1.717857226729393,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7178571708500385,
      "rewards/format_reward_func": 1.0,
      "step": 12470
    },
    {
      "completion_length": 210.02232933044434,
      "epoch": 2.091202481243975,
      "grad_norm": 0.20444495515084055,
      "kl": 0.30419921875,
      "learning_rate": 4.949139844482721e-07,
      "loss": 0.0003,
      "reward": 1.7696429193019867,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7741071805357933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12472
    },
    {
      "completion_length": 224.08483028411865,
      "epoch": 2.0915377844838425,
      "grad_norm": 0.22854212897739953,
      "kl": 0.231231689453125,
      "learning_rate": 4.949115768046893e-07,
      "loss": 0.0002,
      "reward": 1.8196429014205933,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.8241071663796902,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12474
    },
    {
      "completion_length": 218.58929824829102,
      "epoch": 2.09187308772371,
      "grad_norm": 0.22404496442844024,
      "kl": 0.5330810546875,
      "learning_rate": 4.949091685972291e-07,
      "loss": 0.0005,
      "reward": 1.7660714983940125,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7705357223749161,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12476
    },
    {
      "completion_length": 225.81251049041748,
      "epoch": 2.092208390963578,
      "grad_norm": 0.26550278680486267,
      "kl": 0.1566162109375,
      "learning_rate": 4.949067598258972e-07,
      "loss": 0.0002,
      "reward": 1.7589286491274834,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928805589676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12478
    },
    {
      "completion_length": 228.25894165039062,
      "epoch": 2.0925436942034454,
      "grad_norm": 0.20035787461152282,
      "kl": 0.33837890625,
      "learning_rate": 4.949043504906992e-07,
      "loss": 0.0003,
      "reward": 1.8053571954369545,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8098214566707611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12480
    },
    {
      "completion_length": 224.30358123779297,
      "epoch": 2.0928789974433126,
      "grad_norm": 0.09406973070022828,
      "kl": 0.242889404296875,
      "learning_rate": 4.949019405916405e-07,
      "loss": 0.0002,
      "reward": 1.82857146859169,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8285714648663998,
      "rewards/format_reward_func": 1.0,
      "step": 12482
    },
    {
      "completion_length": 229.59375858306885,
      "epoch": 2.0932143006831803,
      "grad_norm": 0.2354022784534689,
      "kl": 0.712188720703125,
      "learning_rate": 4.948995301287268e-07,
      "loss": 0.0007,
      "reward": 1.814285770058632,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857551574707,
      "rewards/format_reward_func": 1.0,
      "step": 12484
    },
    {
      "completion_length": 222.6607255935669,
      "epoch": 2.093549603923048,
      "grad_norm": 0.14301333677568714,
      "kl": 0.337249755859375,
      "learning_rate": 4.948971191019635e-07,
      "loss": 0.0003,
      "reward": 1.7839286401867867,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7883928865194321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12486
    },
    {
      "completion_length": 222.1116180419922,
      "epoch": 2.0938849071629155,
      "grad_norm": 0.09825610613763396,
      "kl": 0.35711669921875,
      "learning_rate": 4.948947075113563e-07,
      "loss": 0.0004,
      "reward": 1.7500000670552254,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 12488
    },
    {
      "completion_length": 230.32143783569336,
      "epoch": 2.094220210402783,
      "grad_norm": 0.01907351915692328,
      "kl": 0.244720458984375,
      "learning_rate": 4.948922953569107e-07,
      "loss": 0.0002,
      "reward": 1.757142923772335,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 12490
    },
    {
      "completion_length": 228.71429443359375,
      "epoch": 2.0945555136426504,
      "grad_norm": 0.19290737070434144,
      "kl": 0.406707763671875,
      "learning_rate": 4.948898826386322e-07,
      "loss": 0.0004,
      "reward": 1.735714353621006,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7357143275439739,
      "rewards/format_reward_func": 1.0,
      "step": 12492
    },
    {
      "completion_length": 230.04465293884277,
      "epoch": 2.094890816882518,
      "grad_norm": 0.19890650026038148,
      "kl": 0.1217041015625,
      "learning_rate": 4.948874693565263e-07,
      "loss": 0.0001,
      "reward": 1.7410715147852898,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7455357387661934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12494
    },
    {
      "completion_length": 232.09375953674316,
      "epoch": 2.0952261201223856,
      "grad_norm": 0.24585345275777434,
      "kl": 0.1280670166015625,
      "learning_rate": 4.948850555105988e-07,
      "loss": 0.0001,
      "reward": 1.7214286476373672,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7303571812808514,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12496
    },
    {
      "completion_length": 227.74554634094238,
      "epoch": 2.0955614233622533,
      "grad_norm": 0.2588696781419015,
      "kl": 0.45916748046875,
      "learning_rate": 4.948826411008551e-07,
      "loss": 0.0005,
      "reward": 1.757142923772335,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428697556257,
      "rewards/format_reward_func": 1.0,
      "step": 12498
    },
    {
      "completion_length": 245.97768878936768,
      "epoch": 2.095896726602121,
      "grad_norm": 0.428716342350395,
      "kl": 0.237884521484375,
      "learning_rate": 4.948802261273007e-07,
      "loss": 0.0002,
      "reward": 1.7392857894301414,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.748214315623045,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12500
    },
    {
      "completion_length": 225.5044755935669,
      "epoch": 2.0962320298419885,
      "grad_norm": 0.16797785673849946,
      "kl": 0.152435302734375,
      "learning_rate": 4.948778105899412e-07,
      "loss": 0.0002,
      "reward": 1.8142857775092125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857402563095,
      "rewards/format_reward_func": 1.0,
      "step": 12502
    },
    {
      "completion_length": 239.165189743042,
      "epoch": 2.0965673330818557,
      "grad_norm": 0.13361972927868213,
      "kl": 0.1374664306640625,
      "learning_rate": 4.948753944887823e-07,
      "loss": 0.0001,
      "reward": 1.814285770058632,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857439815998,
      "rewards/format_reward_func": 1.0,
      "step": 12504
    },
    {
      "completion_length": 223.86608219146729,
      "epoch": 2.0969026363217234,
      "grad_norm": 0.1260006794495531,
      "kl": 0.21466064453125,
      "learning_rate": 4.948729778238293e-07,
      "loss": 0.0002,
      "reward": 1.8000000268220901,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 12506
    },
    {
      "completion_length": 233.87947368621826,
      "epoch": 2.097237939561591,
      "grad_norm": 0.21277466514518686,
      "kl": 0.112701416015625,
      "learning_rate": 4.948705605950879e-07,
      "loss": 0.0001,
      "reward": 1.7750000357627869,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7839285973459482,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12508
    },
    {
      "completion_length": 237.5178689956665,
      "epoch": 2.0975732428014586,
      "grad_norm": 0.2227883614821665,
      "kl": 0.1185150146484375,
      "learning_rate": 4.948681428025638e-07,
      "loss": 0.0001,
      "reward": 1.775000050663948,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 12510
    },
    {
      "completion_length": 250.70983219146729,
      "epoch": 2.0979085460413263,
      "grad_norm": 0.20100054630349626,
      "kl": 0.16143798828125,
      "learning_rate": 4.948657244462624e-07,
      "loss": 0.0002,
      "reward": 1.719642922282219,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7241071797907352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12512
    },
    {
      "completion_length": 248.77233505249023,
      "epoch": 2.098243849281194,
      "grad_norm": 0.3901435180972612,
      "kl": 0.18560791015625,
      "learning_rate": 4.948633055261894e-07,
      "loss": 0.0002,
      "reward": 1.803571492433548,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.8035714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 12514
    },
    {
      "completion_length": 248.1026906967163,
      "epoch": 2.098579152521061,
      "grad_norm": 0.17118320801676853,
      "kl": 0.18536376953125,
      "learning_rate": 4.948608860423501e-07,
      "loss": 0.0002,
      "reward": 1.825000062584877,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8250000141561031,
      "rewards/format_reward_func": 1.0,
      "step": 12516
    },
    {
      "completion_length": 253.0580472946167,
      "epoch": 2.0989144557609287,
      "grad_norm": 0.1819693796045121,
      "kl": 0.36138916015625,
      "learning_rate": 4.948584659947504e-07,
      "loss": 0.0004,
      "reward": 1.8250000402331352,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8250000327825546,
      "rewards/format_reward_func": 1.0,
      "step": 12518
    },
    {
      "completion_length": 260.1785840988159,
      "epoch": 2.0992497590007964,
      "grad_norm": 0.3183018413425357,
      "kl": 0.29443359375,
      "learning_rate": 4.948560453833956e-07,
      "loss": 0.0003,
      "reward": 1.7267857939004898,
      "reward_std": 0.0530330091714859,
      "rewards/equation_reward_func": 0.7401785999536514,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 12520
    },
    {
      "completion_length": 256.9642972946167,
      "epoch": 2.099585062240664,
      "grad_norm": 0.3034727556135848,
      "kl": 0.21856689453125,
      "learning_rate": 4.948536242082915e-07,
      "loss": 0.0002,
      "reward": 1.7339286357164383,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.738392885774374,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12522
    },
    {
      "completion_length": 252.75447750091553,
      "epoch": 2.0999203654805316,
      "grad_norm": 0.2733843588864411,
      "kl": 0.21575927734375,
      "learning_rate": 4.948512024694436e-07,
      "loss": 0.0002,
      "reward": 1.783928632736206,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7883928883820772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12524
    },
    {
      "completion_length": 256.7901906967163,
      "epoch": 2.100255668720399,
      "grad_norm": 0.09788705115293139,
      "kl": 0.50946044921875,
      "learning_rate": 4.948487801668574e-07,
      "loss": 0.0005,
      "reward": 1.7785715013742447,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 12526
    },
    {
      "completion_length": 254.4241180419922,
      "epoch": 2.1005909719602665,
      "grad_norm": 0.4143288036401865,
      "kl": 0.23956298828125,
      "learning_rate": 4.948463573005384e-07,
      "loss": 0.0002,
      "reward": 1.7933036237955093,
      "reward_std": 0.05997780663892627,
      "rewards/equation_reward_func": 0.7991071790456772,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 12528
    },
    {
      "completion_length": 253.6160831451416,
      "epoch": 2.100926275200134,
      "grad_norm": 0.2037976901278312,
      "kl": 0.17803955078125,
      "learning_rate": 4.948439338704925e-07,
      "loss": 0.0002,
      "reward": 1.7428572103381157,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7517857495695353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12530
    },
    {
      "completion_length": 254.9375114440918,
      "epoch": 2.1012615784400017,
      "grad_norm": 0.1723439343144399,
      "kl": 0.4744873046875,
      "learning_rate": 4.94841509876725e-07,
      "loss": 0.0005,
      "reward": 1.7910715118050575,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7955357320606709,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12532
    },
    {
      "completion_length": 255.24108505249023,
      "epoch": 2.1015968816798694,
      "grad_norm": 0.555953331373012,
      "kl": 0.710052490234375,
      "learning_rate": 4.948390853192415e-07,
      "loss": 0.0007,
      "reward": 1.7714286297559738,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7803571745753288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12534
    },
    {
      "completion_length": 257.3660840988159,
      "epoch": 2.101932184919737,
      "grad_norm": 0.2409187824669436,
      "kl": 0.139373779296875,
      "learning_rate": 4.948366601980479e-07,
      "loss": 0.0001,
      "reward": 1.805357202887535,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8098214566707611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12536
    },
    {
      "completion_length": 257.7991199493408,
      "epoch": 2.102267488159604,
      "grad_norm": 0.2669260915173354,
      "kl": 0.519439697265625,
      "learning_rate": 4.948342345131492e-07,
      "loss": 0.0005,
      "reward": 1.7803571969270706,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214767873287,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12538
    },
    {
      "completion_length": 248.3526906967163,
      "epoch": 2.102602791399472,
      "grad_norm": 0.24818006289820269,
      "kl": 0.297607421875,
      "learning_rate": 4.948318082645515e-07,
      "loss": 0.0003,
      "reward": 1.7464286386966705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 12540
    },
    {
      "completion_length": 247.46429920196533,
      "epoch": 2.1029380946393395,
      "grad_norm": 0.16690281788865083,
      "kl": 0.3175048828125,
      "learning_rate": 4.948293814522602e-07,
      "loss": 0.0003,
      "reward": 1.787500038743019,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.8008929006755352,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 12542
    },
    {
      "completion_length": 249.2991180419922,
      "epoch": 2.103273397879207,
      "grad_norm": 0.3490837667156745,
      "kl": 0.246368408203125,
      "learning_rate": 4.948269540762809e-07,
      "loss": 0.0002,
      "reward": 1.789285771548748,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 12544
    },
    {
      "completion_length": 254.8884038925171,
      "epoch": 2.1036087011190747,
      "grad_norm": 0.13535564372346423,
      "kl": 0.2268829345703125,
      "learning_rate": 4.94824526136619e-07,
      "loss": 0.0002,
      "reward": 1.7232143729925156,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7276785969734192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12546
    },
    {
      "completion_length": 262.5759038925171,
      "epoch": 2.103944004358942,
      "grad_norm": 0.17458894142526152,
      "kl": 0.57427978515625,
      "learning_rate": 4.948220976332804e-07,
      "loss": 0.0006,
      "reward": 1.8000000715255737,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 12548
    },
    {
      "completion_length": 257.714298248291,
      "epoch": 2.1042793075988095,
      "grad_norm": 0.309491138362646,
      "kl": 0.987518310546875,
      "learning_rate": 4.948196685662705e-07,
      "loss": 0.001,
      "reward": 1.7714286521077156,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 12550
    },
    {
      "completion_length": 256.870548248291,
      "epoch": 2.104614610838677,
      "grad_norm": 0.1550566234114588,
      "kl": 0.277984619140625,
      "learning_rate": 4.948172389355951e-07,
      "loss": 0.0003,
      "reward": 1.778571493923664,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 12552
    },
    {
      "completion_length": 251.26787185668945,
      "epoch": 2.104949914078545,
      "grad_norm": 0.16131576931507266,
      "kl": 0.341156005859375,
      "learning_rate": 4.948148087412594e-07,
      "loss": 0.0003,
      "reward": 1.7928572073578835,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.8017857372760773,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12554
    },
    {
      "completion_length": 250.04911994934082,
      "epoch": 2.1052852173184124,
      "grad_norm": 0.21502764088900536,
      "kl": 0.14691162109375,
      "learning_rate": 4.948123779832694e-07,
      "loss": 0.0001,
      "reward": 1.7589286267757416,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7633928991854191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12556
    },
    {
      "completion_length": 282.6250104904175,
      "epoch": 2.10562052055828,
      "grad_norm": 0.3334944368615149,
      "kl": 1.33514404296875,
      "learning_rate": 4.948099466616307e-07,
      "loss": 0.0013,
      "reward": 1.750000037252903,
      "reward_std": 0.08081220276653767,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 12558
    },
    {
      "completion_length": 262.6875114440918,
      "epoch": 2.1059558237981473,
      "grad_norm": 0.0697285926398769,
      "kl": 0.35528564453125,
      "learning_rate": 4.948075147763484e-07,
      "loss": 0.0004,
      "reward": 1.7767857760190964,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7901786081492901,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 12560
    },
    {
      "completion_length": 270.2767963409424,
      "epoch": 2.106291127038015,
      "grad_norm": 0.18152372568138508,
      "kl": 0.301727294921875,
      "learning_rate": 4.948050823274287e-07,
      "loss": 0.0003,
      "reward": 1.7553572058677673,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214633762836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12562
    },
    {
      "completion_length": 267.56251335144043,
      "epoch": 2.1066264302778825,
      "grad_norm": 0.18210512754892727,
      "kl": 0.208160400390625,
      "learning_rate": 4.948026493148769e-07,
      "loss": 0.0002,
      "reward": 1.8107143566012383,
      "reward_std": 0.07576144021004438,
      "rewards/equation_reward_func": 0.8196428790688515,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12564
    },
    {
      "completion_length": 267.49554538726807,
      "epoch": 2.10696173351775,
      "grad_norm": 0.654924979260132,
      "kl": 0.542205810546875,
      "learning_rate": 4.948002157386984e-07,
      "loss": 0.0005,
      "reward": 1.7785714864730835,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7875000312924385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12566
    },
    {
      "completion_length": 265.4642963409424,
      "epoch": 2.107297036757618,
      "grad_norm": 0.10704312735958628,
      "kl": 0.801483154296875,
      "learning_rate": 4.947977815988992e-07,
      "loss": 0.0008,
      "reward": 1.7446429431438446,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7491071783006191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12568
    },
    {
      "completion_length": 278.745548248291,
      "epoch": 2.107632339997485,
      "grad_norm": 0.1477785650076975,
      "kl": 0.212066650390625,
      "learning_rate": 4.947953468954848e-07,
      "loss": 0.0002,
      "reward": 1.7196429148316383,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7419643215835094,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 12570
    },
    {
      "completion_length": 269.9241199493408,
      "epoch": 2.1079676432373526,
      "grad_norm": 0.2375696627186269,
      "kl": 0.45843505859375,
      "learning_rate": 4.947929116284606e-07,
      "loss": 0.0005,
      "reward": 1.7196429073810577,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7241071723401546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12572
    },
    {
      "completion_length": 263.5312604904175,
      "epoch": 2.1083029464772203,
      "grad_norm": 0.15276034133103347,
      "kl": 0.269683837890625,
      "learning_rate": 4.947904757978325e-07,
      "loss": 0.0003,
      "reward": 1.7464286610484123,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 12574
    },
    {
      "completion_length": 253.84375858306885,
      "epoch": 2.108638249717088,
      "grad_norm": 0.19924283727198752,
      "kl": 0.11798095703125,
      "learning_rate": 4.947880394036058e-07,
      "loss": 0.0001,
      "reward": 1.7571428939700127,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7660714443773031,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12576
    },
    {
      "completion_length": 266.6339387893677,
      "epoch": 2.1089735529569555,
      "grad_norm": 0.5955246532748683,
      "kl": 0.442657470703125,
      "learning_rate": 4.947856024457865e-07,
      "loss": 0.0004,
      "reward": 1.7410714998841286,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7455357443541288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12578
    },
    {
      "completion_length": 255.42411708831787,
      "epoch": 2.109308856196823,
      "grad_norm": 0.1900532254599884,
      "kl": 0.520721435546875,
      "learning_rate": 4.947831649243798e-07,
      "loss": 0.0005,
      "reward": 1.8107143193483353,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.8196428939700127,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12580
    },
    {
      "completion_length": 269.7991189956665,
      "epoch": 2.1096441594366904,
      "grad_norm": 0.31313371529861506,
      "kl": 0.828125,
      "learning_rate": 4.947807268393915e-07,
      "loss": 0.0008,
      "reward": 1.7093750685453415,
      "reward_std": 0.06755395070649683,
      "rewards/equation_reward_func": 0.7196428924798965,
      "rewards/format_reward_func": 0.9897321499884129,
      "step": 12582
    },
    {
      "completion_length": 256.04465198516846,
      "epoch": 2.109979462676558,
      "grad_norm": 0.2102414212533989,
      "kl": 1.605743408203125,
      "learning_rate": 4.947782881908273e-07,
      "loss": 0.0016,
      "reward": 1.8017857819795609,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8062500320374966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12584
    },
    {
      "completion_length": 251.70536613464355,
      "epoch": 2.1103147659164256,
      "grad_norm": 0.13884928172541447,
      "kl": 0.314422607421875,
      "learning_rate": 4.947758489786927e-07,
      "loss": 0.0003,
      "reward": 1.7464286386966705,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.755357176065445,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12586
    },
    {
      "completion_length": 254.7366180419922,
      "epoch": 2.1106500691562933,
      "grad_norm": 0.19675570563541028,
      "kl": 0.25933837890625,
      "learning_rate": 4.947734092029934e-07,
      "loss": 0.0003,
      "reward": 1.714285783469677,
      "reward_std": 0.09091372787952423,
      "rewards/equation_reward_func": 0.7232143357396126,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12588
    },
    {
      "completion_length": 251.1428680419922,
      "epoch": 2.110985372396161,
      "grad_norm": 0.24465264663289335,
      "kl": 0.3800048828125,
      "learning_rate": 4.947709688637348e-07,
      "loss": 0.0004,
      "reward": 1.7107143849134445,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7107143178582191,
      "rewards/format_reward_func": 1.0,
      "step": 12590
    },
    {
      "completion_length": 250.22322463989258,
      "epoch": 2.1113206756360285,
      "grad_norm": 0.18432661633018005,
      "kl": 0.72412109375,
      "learning_rate": 4.947685279609228e-07,
      "loss": 0.0007,
      "reward": 1.775000087916851,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 12592
    },
    {
      "completion_length": 253.7857265472412,
      "epoch": 2.1116559788758957,
      "grad_norm": 0.2395700514182744,
      "kl": 0.30462646484375,
      "learning_rate": 4.947660864945629e-07,
      "loss": 0.0003,
      "reward": 1.7535715028643608,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7535714656114578,
      "rewards/format_reward_func": 1.0,
      "step": 12594
    },
    {
      "completion_length": 243.6473331451416,
      "epoch": 2.1119912821157634,
      "grad_norm": 0.4434999353814348,
      "kl": 0.285491943359375,
      "learning_rate": 4.947636444646605e-07,
      "loss": 0.0003,
      "reward": 1.7285715267062187,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7285714596509933,
      "rewards/format_reward_func": 1.0,
      "step": 12596
    },
    {
      "completion_length": 253.28126430511475,
      "epoch": 2.112326585355631,
      "grad_norm": 0.20279855857111112,
      "kl": 0.1322021484375,
      "learning_rate": 4.947612018712218e-07,
      "loss": 0.0001,
      "reward": 1.7285715118050575,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714671015739,
      "rewards/format_reward_func": 1.0,
      "step": 12598
    },
    {
      "completion_length": 249.29465770721436,
      "epoch": 2.1126618885954986,
      "grad_norm": 0.28162706638188384,
      "kl": 0.419097900390625,
      "learning_rate": 4.947587587142518e-07,
      "loss": 0.0004,
      "reward": 1.7285715192556381,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7285714671015739,
      "rewards/format_reward_func": 1.0,
      "step": 12600
    },
    {
      "completion_length": 249.19643878936768,
      "epoch": 2.1129971918353663,
      "grad_norm": 0.18187601895711714,
      "kl": 0.145660400390625,
      "learning_rate": 4.947563149937565e-07,
      "loss": 0.0001,
      "reward": 1.8200893625617027,
      "reward_std": 0.05240166140720248,
      "rewards/equation_reward_func": 0.8214285932481289,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 12602
    },
    {
      "completion_length": 243.66965579986572,
      "epoch": 2.1133324950752335,
      "grad_norm": 0.23567424608565962,
      "kl": 0.263092041015625,
      "learning_rate": 4.947538707097413e-07,
      "loss": 0.0003,
      "reward": 1.8250000774860382,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.825000025331974,
      "rewards/format_reward_func": 1.0,
      "step": 12604
    },
    {
      "completion_length": 246.23215579986572,
      "epoch": 2.113667798315101,
      "grad_norm": 0.26916544402526654,
      "kl": 0.621246337890625,
      "learning_rate": 4.94751425862212e-07,
      "loss": 0.0006,
      "reward": 1.7607143446803093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143223285675,
      "rewards/format_reward_func": 1.0,
      "step": 12606
    },
    {
      "completion_length": 238.81697463989258,
      "epoch": 2.1140031015549687,
      "grad_norm": 0.18268104728073659,
      "kl": 0.4063720703125,
      "learning_rate": 4.947489804511742e-07,
      "loss": 0.0004,
      "reward": 1.757142923772335,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 12608
    },
    {
      "completion_length": 251.133939743042,
      "epoch": 2.1143384047948364,
      "grad_norm": 0.17624304213752912,
      "kl": 0.40753173828125,
      "learning_rate": 4.947465344766335e-07,
      "loss": 0.0004,
      "reward": 1.7535715103149414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714451223612,
      "rewards/format_reward_func": 1.0,
      "step": 12610
    },
    {
      "completion_length": 248.90625762939453,
      "epoch": 2.114673708034704,
      "grad_norm": 0.28952980847030996,
      "kl": 0.224822998046875,
      "learning_rate": 4.947440879385955e-07,
      "loss": 0.0002,
      "reward": 1.7214286476373672,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7214285992085934,
      "rewards/format_reward_func": 1.0,
      "step": 12612
    },
    {
      "completion_length": 231.34822463989258,
      "epoch": 2.1150090112745716,
      "grad_norm": 0.39774063211490374,
      "kl": 0.192047119140625,
      "learning_rate": 4.947416408370659e-07,
      "loss": 0.0002,
      "reward": 1.8589285984635353,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.8633928783237934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12614
    },
    {
      "completion_length": 242.60268783569336,
      "epoch": 2.115344314514439,
      "grad_norm": 0.24908888175240032,
      "kl": 0.17919921875,
      "learning_rate": 4.947391931720503e-07,
      "loss": 0.0002,
      "reward": 1.7357143461704254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143182307482,
      "rewards/format_reward_func": 1.0,
      "step": 12616
    },
    {
      "completion_length": 253.0669755935669,
      "epoch": 2.1156796177543065,
      "grad_norm": 0.19669393215519845,
      "kl": 0.18682861328125,
      "learning_rate": 4.947367449435542e-07,
      "loss": 0.0002,
      "reward": 1.7875000536441803,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7919643111526966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12618
    },
    {
      "completion_length": 242.2232255935669,
      "epoch": 2.116014920994174,
      "grad_norm": 0.1812298854708925,
      "kl": 0.133697509765625,
      "learning_rate": 4.947342961515835e-07,
      "loss": 0.0001,
      "reward": 1.785714365541935,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 12620
    },
    {
      "completion_length": 245.42858505249023,
      "epoch": 2.1163502242340417,
      "grad_norm": 0.6592117291790457,
      "kl": 0.193939208984375,
      "learning_rate": 4.947318467961437e-07,
      "loss": 0.0002,
      "reward": 1.7946429029107094,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7991071678698063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12622
    },
    {
      "completion_length": 254.62947750091553,
      "epoch": 2.1166855274739094,
      "grad_norm": 0.1700151949627714,
      "kl": 0.1432952880859375,
      "learning_rate": 4.947293968772405e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286111295223,
      "rewards/format_reward_func": 1.0,
      "step": 12624
    },
    {
      "completion_length": 242.79911708831787,
      "epoch": 2.1170208307137766,
      "grad_norm": 0.17733056326423086,
      "kl": 0.13336181640625,
      "learning_rate": 4.947269463948795e-07,
      "loss": 0.0001,
      "reward": 1.7964286133646965,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964286170899868,
      "rewards/format_reward_func": 1.0,
      "step": 12626
    },
    {
      "completion_length": 252.6250114440918,
      "epoch": 2.117356133953644,
      "grad_norm": 0.1922726528758622,
      "kl": 0.155487060546875,
      "learning_rate": 4.947244953490662e-07,
      "loss": 0.0002,
      "reward": 1.760714367032051,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.760714340955019,
      "rewards/format_reward_func": 1.0,
      "step": 12628
    },
    {
      "completion_length": 249.55804920196533,
      "epoch": 2.117691437193512,
      "grad_norm": 0.24193949809038998,
      "kl": 0.129150390625,
      "learning_rate": 4.947220437398064e-07,
      "loss": 0.0001,
      "reward": 1.7750000357627869,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000413507223,
      "rewards/format_reward_func": 1.0,
      "step": 12630
    },
    {
      "completion_length": 243.89286708831787,
      "epoch": 2.1180267404333795,
      "grad_norm": 0.1576165068164449,
      "kl": 0.19390869140625,
      "learning_rate": 4.947195915671058e-07,
      "loss": 0.0002,
      "reward": 1.8035714775323868,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8035714738070965,
      "rewards/format_reward_func": 1.0,
      "step": 12632
    },
    {
      "completion_length": 256.64733505249023,
      "epoch": 2.118362043673247,
      "grad_norm": 0.10762529730307481,
      "kl": 0.192108154296875,
      "learning_rate": 4.947171388309699e-07,
      "loss": 0.0002,
      "reward": 1.8285714536905289,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8285714499652386,
      "rewards/format_reward_func": 1.0,
      "step": 12634
    },
    {
      "completion_length": 255.4642972946167,
      "epoch": 2.1186973469131147,
      "grad_norm": 0.1776726750654768,
      "kl": 0.19140625,
      "learning_rate": 4.947146855314045e-07,
      "loss": 0.0002,
      "reward": 1.721428669989109,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7214286103844643,
      "rewards/format_reward_func": 1.0,
      "step": 12636
    },
    {
      "completion_length": 249.75893878936768,
      "epoch": 2.119032650152982,
      "grad_norm": 0.19551152845046466,
      "kl": 0.137237548828125,
      "learning_rate": 4.947122316684151e-07,
      "loss": 0.0001,
      "reward": 1.8178572058677673,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8178571723401546,
      "rewards/format_reward_func": 1.0,
      "step": 12638
    },
    {
      "completion_length": 263.6919755935669,
      "epoch": 2.1193679533928496,
      "grad_norm": 0.16631205617192682,
      "kl": 0.1456298828125,
      "learning_rate": 4.947097772420074e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 12640
    },
    {
      "completion_length": 262.1205463409424,
      "epoch": 2.119703256632717,
      "grad_norm": 0.25879339095547543,
      "kl": 0.1573333740234375,
      "learning_rate": 4.94707322252187e-07,
      "loss": 0.0002,
      "reward": 1.7571429461240768,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 1.0,
      "step": 12642
    },
    {
      "completion_length": 265.4285840988159,
      "epoch": 2.120038559872585,
      "grad_norm": 0.008186147312404166,
      "kl": 0.17401123046875,
      "learning_rate": 4.947048666989597e-07,
      "loss": 0.0002,
      "reward": 1.7107143551111221,
      "reward_std": 0.025253813713788986,
      "rewards/equation_reward_func": 0.7196428868919611,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12644
    },
    {
      "completion_length": 255.92858600616455,
      "epoch": 2.1203738631124525,
      "grad_norm": 0.1079018393352671,
      "kl": 0.259307861328125,
      "learning_rate": 4.947024105823311e-07,
      "loss": 0.0003,
      "reward": 1.769642896950245,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7830357346683741,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 12646
    },
    {
      "completion_length": 261.4285840988159,
      "epoch": 2.12070916635232,
      "grad_norm": 0.16324194777430936,
      "kl": 0.167999267578125,
      "learning_rate": 4.946999539023068e-07,
      "loss": 0.0002,
      "reward": 1.7410715147852898,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.745535746216774,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12648
    },
    {
      "completion_length": 266.54465770721436,
      "epoch": 2.1210444695921873,
      "grad_norm": 0.24910040332383881,
      "kl": 0.21734619140625,
      "learning_rate": 4.946974966588924e-07,
      "loss": 0.0002,
      "reward": 1.7464286237955093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464285958558321,
      "rewards/format_reward_func": 1.0,
      "step": 12650
    },
    {
      "completion_length": 273.5401954650879,
      "epoch": 2.121379772832055,
      "grad_norm": 0.24830042928049242,
      "kl": 0.1650390625,
      "learning_rate": 4.946950388520938e-07,
      "loss": 0.0002,
      "reward": 1.739285796880722,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7392857521772385,
      "rewards/format_reward_func": 1.0,
      "step": 12652
    },
    {
      "completion_length": 259.7500104904175,
      "epoch": 2.1217150760719226,
      "grad_norm": 0.1759505675423626,
      "kl": 0.13311767578125,
      "learning_rate": 4.946925804819163e-07,
      "loss": 0.0001,
      "reward": 1.8250000476837158,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8250000141561031,
      "rewards/format_reward_func": 1.0,
      "step": 12654
    },
    {
      "completion_length": 263.8080463409424,
      "epoch": 2.12205037931179,
      "grad_norm": 0.3763846045538354,
      "kl": 0.1600341796875,
      "learning_rate": 4.946901215483659e-07,
      "loss": 0.0002,
      "reward": 1.753571480512619,
      "reward_std": 0.08586296532303095,
      "rewards/equation_reward_func": 0.7625000216066837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12656
    },
    {
      "completion_length": 262.22768688201904,
      "epoch": 2.122385682551658,
      "grad_norm": 0.09794541085463028,
      "kl": 0.1695556640625,
      "learning_rate": 4.94687662051448e-07,
      "loss": 0.0002,
      "reward": 1.7892857566475868,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857529222965,
      "rewards/format_reward_func": 1.0,
      "step": 12658
    },
    {
      "completion_length": 261.977689743042,
      "epoch": 2.122720985791525,
      "grad_norm": 0.21298906750501434,
      "kl": 0.194488525390625,
      "learning_rate": 4.946852019911684e-07,
      "loss": 0.0002,
      "reward": 1.7750000357627869,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000301748514,
      "rewards/format_reward_func": 1.0,
      "step": 12660
    },
    {
      "completion_length": 253.3482255935669,
      "epoch": 2.1230562890313927,
      "grad_norm": 0.1329742623791657,
      "kl": 0.120574951171875,
      "learning_rate": 4.946827413675328e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 12662
    },
    {
      "completion_length": 256.8526906967163,
      "epoch": 2.1233915922712603,
      "grad_norm": 0.27845250518816944,
      "kl": 0.246185302734375,
      "learning_rate": 4.946802801805467e-07,
      "loss": 0.0002,
      "reward": 1.7392857819795609,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857410013676,
      "rewards/format_reward_func": 1.0,
      "step": 12664
    },
    {
      "completion_length": 247.48215293884277,
      "epoch": 2.123726895511128,
      "grad_norm": 0.21240464239704232,
      "kl": 0.2212371826171875,
      "learning_rate": 4.946778184302161e-07,
      "loss": 0.0002,
      "reward": 1.7964286282658577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.796428594738245,
      "rewards/format_reward_func": 1.0,
      "step": 12666
    },
    {
      "completion_length": 255.99554634094238,
      "epoch": 2.1240621987509956,
      "grad_norm": 0.2736472912552943,
      "kl": 0.261810302734375,
      "learning_rate": 4.946753561165462e-07,
      "loss": 0.0003,
      "reward": 1.7482143640518188,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7526786103844643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12668
    },
    {
      "completion_length": 255.18751049041748,
      "epoch": 2.124397501990863,
      "grad_norm": 0.14204252196879363,
      "kl": 0.22735595703125,
      "learning_rate": 4.946728932395431e-07,
      "loss": 0.0002,
      "reward": 1.716071493923664,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7205357402563095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12670
    },
    {
      "completion_length": 244.4464406967163,
      "epoch": 2.1247328052307304,
      "grad_norm": 0.44296381897565446,
      "kl": 0.254974365234375,
      "learning_rate": 4.946704297992121e-07,
      "loss": 0.0003,
      "reward": 1.8035714998841286,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 12672
    },
    {
      "completion_length": 250.7544765472412,
      "epoch": 2.125068108470598,
      "grad_norm": 0.1577363318076351,
      "kl": 0.217864990234375,
      "learning_rate": 4.946679657955591e-07,
      "loss": 0.0002,
      "reward": 1.789285771548748,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857268452644,
      "rewards/format_reward_func": 1.0,
      "step": 12674
    },
    {
      "completion_length": 248.09376049041748,
      "epoch": 2.1254034117104657,
      "grad_norm": 0.14258669660016726,
      "kl": 0.170196533203125,
      "learning_rate": 4.946655012285898e-07,
      "loss": 0.0002,
      "reward": 1.7589286416769028,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7633928880095482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12676
    },
    {
      "completion_length": 243.52233123779297,
      "epoch": 2.1257387149503333,
      "grad_norm": 0.23125273450125225,
      "kl": 0.18536376953125,
      "learning_rate": 4.946630360983098e-07,
      "loss": 0.0002,
      "reward": 1.7379465028643608,
      "reward_std": 0.0473508988507092,
      "rewards/equation_reward_func": 0.7392857559025288,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 12678
    },
    {
      "completion_length": 251.3750114440918,
      "epoch": 2.126074018190201,
      "grad_norm": 0.18106699184120606,
      "kl": 0.211212158203125,
      "learning_rate": 4.946605704047247e-07,
      "loss": 0.0002,
      "reward": 1.753571480512619,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 12680
    },
    {
      "completion_length": 246.27679538726807,
      "epoch": 2.126409321430068,
      "grad_norm": 0.6107476265404755,
      "kl": 0.19964599609375,
      "learning_rate": 4.946581041478404e-07,
      "loss": 0.0002,
      "reward": 1.7750000581145287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 12682
    },
    {
      "completion_length": 250.76340675354004,
      "epoch": 2.1267446246699357,
      "grad_norm": 0.19478372513176684,
      "kl": 0.208770751953125,
      "learning_rate": 4.946556373276622e-07,
      "loss": 0.0002,
      "reward": 1.7535714879631996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714711993933,
      "rewards/format_reward_func": 1.0,
      "step": 12684
    },
    {
      "completion_length": 248.64733123779297,
      "epoch": 2.1270799279098034,
      "grad_norm": 0.0650269589634784,
      "kl": 0.2137451171875,
      "learning_rate": 4.946531699441963e-07,
      "loss": 0.0002,
      "reward": 1.6892857998609543,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.6892857532948256,
      "rewards/format_reward_func": 1.0,
      "step": 12686
    },
    {
      "completion_length": 248.41519165039062,
      "epoch": 2.127415231149671,
      "grad_norm": 0.06310082996410975,
      "kl": 0.1319580078125,
      "learning_rate": 4.946507019974479e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 12688
    },
    {
      "completion_length": 251.1741189956665,
      "epoch": 2.1277505343895387,
      "grad_norm": 0.15660504026504873,
      "kl": 0.162445068359375,
      "learning_rate": 4.946482334874229e-07,
      "loss": 0.0002,
      "reward": 1.8142857626080513,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857365310192,
      "rewards/format_reward_func": 1.0,
      "step": 12690
    },
    {
      "completion_length": 240.34822463989258,
      "epoch": 2.1280858376294063,
      "grad_norm": 0.3681435884554297,
      "kl": 0.135772705078125,
      "learning_rate": 4.94645764414127e-07,
      "loss": 0.0001,
      "reward": 1.8151786178350449,
      "reward_std": 0.049244935624301434,
      "rewards/equation_reward_func": 0.8169643133878708,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 12692
    },
    {
      "completion_length": 234.3750114440918,
      "epoch": 2.1284211408692735,
      "grad_norm": 0.0995885339917896,
      "kl": 0.128265380859375,
      "learning_rate": 4.946432947775657e-07,
      "loss": 0.0001,
      "reward": 1.8035714700818062,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8035714570432901,
      "rewards/format_reward_func": 1.0,
      "step": 12694
    },
    {
      "completion_length": 245.80804538726807,
      "epoch": 2.128756444109141,
      "grad_norm": 0.5983936194365209,
      "kl": 0.15484619140625,
      "learning_rate": 4.946408245777449e-07,
      "loss": 0.0002,
      "reward": 1.7571429312229156,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7660714499652386,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12696
    },
    {
      "completion_length": 240.03572368621826,
      "epoch": 2.1290917473490087,
      "grad_norm": 0.21612487513493533,
      "kl": 0.141876220703125,
      "learning_rate": 4.946383538146704e-07,
      "loss": 0.0001,
      "reward": 1.7785714715719223,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714510828257,
      "rewards/format_reward_func": 1.0,
      "step": 12698
    },
    {
      "completion_length": 243.2544755935669,
      "epoch": 2.1294270505888764,
      "grad_norm": 0.15909532838737647,
      "kl": 0.151611328125,
      "learning_rate": 4.946358824883476e-07,
      "loss": 0.0002,
      "reward": 1.8178571686148643,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8178571723401546,
      "rewards/format_reward_func": 1.0,
      "step": 12700
    },
    {
      "completion_length": 236.58036708831787,
      "epoch": 2.129762353828744,
      "grad_norm": 0.1954038164117403,
      "kl": 0.157379150390625,
      "learning_rate": 4.946334105987822e-07,
      "loss": 0.0002,
      "reward": 1.7214286401867867,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7214286029338837,
      "rewards/format_reward_func": 1.0,
      "step": 12702
    },
    {
      "completion_length": 237.7410831451416,
      "epoch": 2.130097657068611,
      "grad_norm": 0.15764717852820267,
      "kl": 0.182037353515625,
      "learning_rate": 4.9463093814598e-07,
      "loss": 0.0002,
      "reward": 1.7714286223053932,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7714285887777805,
      "rewards/format_reward_func": 1.0,
      "step": 12704
    },
    {
      "completion_length": 241.12501049041748,
      "epoch": 2.130432960308479,
      "grad_norm": 0.1504773466200583,
      "kl": 0.121063232421875,
      "learning_rate": 4.946284651299467e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 12706
    },
    {
      "completion_length": 237.4419765472412,
      "epoch": 2.1307682635483465,
      "grad_norm": 0.11417352849954919,
      "kl": 0.2191009521484375,
      "learning_rate": 4.946259915506879e-07,
      "loss": 0.0002,
      "reward": 1.789285771548748,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857566475868,
      "rewards/format_reward_func": 1.0,
      "step": 12708
    },
    {
      "completion_length": 238.57590579986572,
      "epoch": 2.131103566788214,
      "grad_norm": 0.5617548101285669,
      "kl": 0.231964111328125,
      "learning_rate": 4.946235174082096e-07,
      "loss": 0.0002,
      "reward": 1.7808036282658577,
      "reward_std": 0.03724937466904521,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 12710
    },
    {
      "completion_length": 240.80358219146729,
      "epoch": 2.1314388700280817,
      "grad_norm": 0.23236516414959607,
      "kl": 0.173370361328125,
      "learning_rate": 4.94621042702517e-07,
      "loss": 0.0002,
      "reward": 1.7500000596046448,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000111758709,
      "rewards/format_reward_func": 1.0,
      "step": 12712
    },
    {
      "completion_length": 237.41072463989258,
      "epoch": 2.1317741732679494,
      "grad_norm": 0.24766922153293103,
      "kl": 0.2633056640625,
      "learning_rate": 4.946185674336163e-07,
      "loss": 0.0003,
      "reward": 1.7607143372297287,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 12714
    },
    {
      "completion_length": 248.7410831451416,
      "epoch": 2.1321094765078166,
      "grad_norm": 0.2851471685393492,
      "kl": 0.199249267578125,
      "learning_rate": 4.946160916015128e-07,
      "loss": 0.0002,
      "reward": 1.7678572237491608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.767857164144516,
      "rewards/format_reward_func": 1.0,
      "step": 12716
    },
    {
      "completion_length": 235.63393878936768,
      "epoch": 2.132444779747684,
      "grad_norm": 0.24856425416445746,
      "kl": 0.212646484375,
      "learning_rate": 4.946136152062123e-07,
      "loss": 0.0002,
      "reward": 1.74508935213089,
      "reward_std": 0.03346130205318332,
      "rewards/equation_reward_func": 0.7464286088943481,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 12718
    },
    {
      "completion_length": 246.4062614440918,
      "epoch": 2.132780082987552,
      "grad_norm": 0.1818853615657871,
      "kl": 0.207855224609375,
      "learning_rate": 4.946111382477207e-07,
      "loss": 0.0002,
      "reward": 1.7642857730388641,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857637256384,
      "rewards/format_reward_func": 1.0,
      "step": 12720
    },
    {
      "completion_length": 236.11608028411865,
      "epoch": 2.1331153862274195,
      "grad_norm": 0.2388736773932256,
      "kl": 0.304229736328125,
      "learning_rate": 4.946086607260436e-07,
      "loss": 0.0003,
      "reward": 1.7928572073578835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 12722
    },
    {
      "completion_length": 234.6384038925171,
      "epoch": 2.133450689467287,
      "grad_norm": 0.1481105248262793,
      "kl": 0.19354248046875,
      "learning_rate": 4.946061826411866e-07,
      "loss": 0.0002,
      "reward": 1.8000000566244125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 12724
    },
    {
      "completion_length": 236.2589406967163,
      "epoch": 2.1337859927071543,
      "grad_norm": 0.19108891475425632,
      "kl": 0.15118408203125,
      "learning_rate": 4.946037039931554e-07,
      "loss": 0.0002,
      "reward": 1.787500075995922,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7919643186032772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12726
    },
    {
      "completion_length": 232.9821538925171,
      "epoch": 2.134121295947022,
      "grad_norm": 0.1379382492406183,
      "kl": 0.180023193359375,
      "learning_rate": 4.946012247819559e-07,
      "loss": 0.0002,
      "reward": 1.74642863124609,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 12728
    },
    {
      "completion_length": 238.2142972946167,
      "epoch": 2.1344565991868896,
      "grad_norm": 0.42170481009249156,
      "kl": 0.398681640625,
      "learning_rate": 4.945987450075936e-07,
      "loss": 0.0004,
      "reward": 1.7928571999073029,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7928571812808514,
      "rewards/format_reward_func": 1.0,
      "step": 12730
    },
    {
      "completion_length": 240.73661708831787,
      "epoch": 2.134791902426757,
      "grad_norm": 0.13332983942445129,
      "kl": 0.1611328125,
      "learning_rate": 4.945962646700744e-07,
      "loss": 0.0002,
      "reward": 1.7142857909202576,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857555299997,
      "rewards/format_reward_func": 1.0,
      "step": 12732
    },
    {
      "completion_length": 230.14732933044434,
      "epoch": 2.135127205666625,
      "grad_norm": 0.14906506873937503,
      "kl": 0.3936309814453125,
      "learning_rate": 4.945937837694039e-07,
      "loss": 0.0004,
      "reward": 1.805357202887535,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.8098214641213417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12734
    },
    {
      "completion_length": 242.59822177886963,
      "epoch": 2.1354625089064925,
      "grad_norm": 0.09027486073294985,
      "kl": 0.150177001953125,
      "learning_rate": 4.945913023055877e-07,
      "loss": 0.0001,
      "reward": 1.7214286401867867,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7214286178350449,
      "rewards/format_reward_func": 1.0,
      "step": 12736
    },
    {
      "completion_length": 246.93750953674316,
      "epoch": 2.1357978121463597,
      "grad_norm": 0.5243776630881638,
      "kl": 0.211212158203125,
      "learning_rate": 4.945888202786317e-07,
      "loss": 0.0002,
      "reward": 1.760714329779148,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143335044384,
      "rewards/format_reward_func": 1.0,
      "step": 12738
    },
    {
      "completion_length": 242.2634048461914,
      "epoch": 2.1361331153862273,
      "grad_norm": 0.29612204522098395,
      "kl": 0.139556884765625,
      "learning_rate": 4.945863376885415e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 12740
    },
    {
      "completion_length": 239.7812614440918,
      "epoch": 2.136468418626095,
      "grad_norm": 0.24427987178665603,
      "kl": 0.1544189453125,
      "learning_rate": 4.94583854535323e-07,
      "loss": 0.0002,
      "reward": 1.7696429193019867,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7741071581840515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12742
    },
    {
      "completion_length": 234.2276906967163,
      "epoch": 2.1368037218659626,
      "grad_norm": 0.1073191925162693,
      "kl": 0.1297607421875,
      "learning_rate": 4.945813708189816e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7714285925030708,
      "rewards/format_reward_func": 1.0,
      "step": 12744
    },
    {
      "completion_length": 248.2634048461914,
      "epoch": 2.13713902510583,
      "grad_norm": 0.19538950089251983,
      "kl": 0.14056396484375,
      "learning_rate": 4.945788865395234e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7875000201165676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12746
    },
    {
      "completion_length": 246.08929824829102,
      "epoch": 2.137474328345698,
      "grad_norm": 0.2802740460051694,
      "kl": 0.148651123046875,
      "learning_rate": 4.945764016969538e-07,
      "loss": 0.0001,
      "reward": 1.7196429148316383,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7241071872413158,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12748
    },
    {
      "completion_length": 239.8750123977661,
      "epoch": 2.137809631585565,
      "grad_norm": 0.21579468742525323,
      "kl": 0.1495361328125,
      "learning_rate": 4.945739162912787e-07,
      "loss": 0.0001,
      "reward": 1.800000049173832,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000342726707,
      "rewards/format_reward_func": 1.0,
      "step": 12750
    },
    {
      "completion_length": 241.4375114440918,
      "epoch": 2.1381449348254327,
      "grad_norm": 0.1691322578412735,
      "kl": 0.12725830078125,
      "learning_rate": 4.945714303225037e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428775787354,
      "rewards/format_reward_func": 1.0,
      "step": 12752
    },
    {
      "completion_length": 232.0491180419922,
      "epoch": 2.1384802380653003,
      "grad_norm": 0.24020569233265468,
      "kl": 0.12933349609375,
      "learning_rate": 4.945689437906346e-07,
      "loss": 0.0001,
      "reward": 1.8071429133415222,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428835391998,
      "rewards/format_reward_func": 1.0,
      "step": 12754
    },
    {
      "completion_length": 243.6875123977661,
      "epoch": 2.138815541305168,
      "grad_norm": 0.2156933055364058,
      "kl": 0.146575927734375,
      "learning_rate": 4.945664566956771e-07,
      "loss": 0.0001,
      "reward": 1.8142857551574707,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857477068901,
      "rewards/format_reward_func": 1.0,
      "step": 12756
    },
    {
      "completion_length": 235.9553680419922,
      "epoch": 2.1391508445450356,
      "grad_norm": 0.23128610933191962,
      "kl": 0.15655517578125,
      "learning_rate": 4.94563969037637e-07,
      "loss": 0.0002,
      "reward": 1.710714340209961,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7107143141329288,
      "rewards/format_reward_func": 1.0,
      "step": 12758
    },
    {
      "completion_length": 234.47768878936768,
      "epoch": 2.139486147784903,
      "grad_norm": 0.36352148754101415,
      "kl": 0.148345947265625,
      "learning_rate": 4.945614808165199e-07,
      "loss": 0.0001,
      "reward": 1.732142947614193,
      "reward_std": 0.07576143834739923,
      "rewards/equation_reward_func": 0.7321428842842579,
      "rewards/format_reward_func": 1.0,
      "step": 12760
    },
    {
      "completion_length": 244.48215293884277,
      "epoch": 2.1398214510247704,
      "grad_norm": 0.14300670479017372,
      "kl": 0.15911865234375,
      "learning_rate": 4.945589920323317e-07,
      "loss": 0.0002,
      "reward": 1.7642857730388641,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 12762
    },
    {
      "completion_length": 234.37054634094238,
      "epoch": 2.140156754264638,
      "grad_norm": 0.8349680304170968,
      "kl": 0.28076171875,
      "learning_rate": 4.945565026850779e-07,
      "loss": 0.0003,
      "reward": 1.7642857879400253,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857357859612,
      "rewards/format_reward_func": 1.0,
      "step": 12764
    },
    {
      "completion_length": 241.1875114440918,
      "epoch": 2.1404920575045057,
      "grad_norm": 1.056639847178951,
      "kl": 0.64544677734375,
      "learning_rate": 4.945540127747643e-07,
      "loss": 0.0006,
      "reward": 1.7839286103844643,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7883928772062063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12766
    },
    {
      "completion_length": 227.62501049041748,
      "epoch": 2.1408273607443733,
      "grad_norm": 0.1428928703085797,
      "kl": 0.1385498046875,
      "learning_rate": 4.945515223013969e-07,
      "loss": 0.0001,
      "reward": 1.800000049173832,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000156462193,
      "rewards/format_reward_func": 1.0,
      "step": 12768
    },
    {
      "completion_length": 232.0491180419922,
      "epoch": 2.141162663984241,
      "grad_norm": 0.14921203769840066,
      "kl": 0.152801513671875,
      "learning_rate": 4.94549031264981e-07,
      "loss": 0.0002,
      "reward": 1.8250000551342964,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.825000025331974,
      "rewards/format_reward_func": 1.0,
      "step": 12770
    },
    {
      "completion_length": 233.75000953674316,
      "epoch": 2.141497967224108,
      "grad_norm": 0.2428657901809172,
      "kl": 0.146759033203125,
      "learning_rate": 4.945465396655227e-07,
      "loss": 0.0001,
      "reward": 1.8464286103844643,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8464285954833031,
      "rewards/format_reward_func": 1.0,
      "step": 12772
    },
    {
      "completion_length": 239.32143878936768,
      "epoch": 2.1418332704639758,
      "grad_norm": 0.12902869004400336,
      "kl": 0.138916015625,
      "learning_rate": 4.945440475030276e-07,
      "loss": 0.0001,
      "reward": 1.7500000596046448,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 12774
    },
    {
      "completion_length": 232.8750114440918,
      "epoch": 2.1421685737038434,
      "grad_norm": 0.19235439708019675,
      "kl": 0.1307373046875,
      "learning_rate": 4.945415547775013e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 12776
    },
    {
      "completion_length": 243.0669755935669,
      "epoch": 2.142503876943711,
      "grad_norm": 0.435288208029278,
      "kl": 0.150848388671875,
      "learning_rate": 4.945390614889499e-07,
      "loss": 0.0002,
      "reward": 1.7750000655651093,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7839286029338837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 12778
    },
    {
      "completion_length": 234.4062614440918,
      "epoch": 2.1428391801835787,
      "grad_norm": 0.23064924712066712,
      "kl": 0.206634521484375,
      "learning_rate": 4.945365676373787e-07,
      "loss": 0.0002,
      "reward": 1.7821429446339607,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 12780
    },
    {
      "completion_length": 238.9285831451416,
      "epoch": 2.1431744834234463,
      "grad_norm": 0.1791756501286884,
      "kl": 0.144989013671875,
      "learning_rate": 4.945340732227938e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 1.0,
      "step": 12782
    },
    {
      "completion_length": 244.47769165039062,
      "epoch": 2.1435097866633135,
      "grad_norm": 0.4171668278383133,
      "kl": 0.13970947265625,
      "learning_rate": 4.945315782452008e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571626543999,
      "rewards/format_reward_func": 1.0,
      "step": 12784
    },
    {
      "completion_length": 236.4241189956665,
      "epoch": 2.143845089903181,
      "grad_norm": 0.19374842158996536,
      "kl": 0.1378936767578125,
      "learning_rate": 4.945290827046053e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143227010965,
      "rewards/format_reward_func": 1.0,
      "step": 12786
    },
    {
      "completion_length": 229.32590198516846,
      "epoch": 2.1441803931430488,
      "grad_norm": 0.21744398788212368,
      "kl": 0.120758056640625,
      "learning_rate": 4.945265866010133e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 12788
    },
    {
      "completion_length": 226.24108409881592,
      "epoch": 2.1445156963829164,
      "grad_norm": 0.27278636765832526,
      "kl": 0.154998779296875,
      "learning_rate": 4.945240899344304e-07,
      "loss": 0.0002,
      "reward": 1.7642857655882835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857581377029,
      "rewards/format_reward_func": 1.0,
      "step": 12790
    },
    {
      "completion_length": 234.80358219146729,
      "epoch": 2.144850999622784,
      "grad_norm": 0.3670106294704579,
      "kl": 0.1717529296875,
      "learning_rate": 4.945215927048623e-07,
      "loss": 0.0002,
      "reward": 1.728571504354477,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7285714671015739,
      "rewards/format_reward_func": 1.0,
      "step": 12792
    },
    {
      "completion_length": 238.80804824829102,
      "epoch": 2.145186302862651,
      "grad_norm": 0.34621680075467387,
      "kl": 0.151611328125,
      "learning_rate": 4.94519094912315e-07,
      "loss": 0.0002,
      "reward": 1.7857143580913544,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 12794
    },
    {
      "completion_length": 234.86608028411865,
      "epoch": 2.145521606102519,
      "grad_norm": 0.35599495255633246,
      "kl": 0.12451171875,
      "learning_rate": 4.945165965567939e-07,
      "loss": 0.0001,
      "reward": 1.8035715073347092,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8035714663565159,
      "rewards/format_reward_func": 1.0,
      "step": 12796
    },
    {
      "completion_length": 233.17858123779297,
      "epoch": 2.1458569093423865,
      "grad_norm": 0.3200094677780826,
      "kl": 0.1365966796875,
      "learning_rate": 4.945140976383051e-07,
      "loss": 0.0001,
      "reward": 1.769642911851406,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.774107176810503,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12798
    },
    {
      "completion_length": 233.56251049041748,
      "epoch": 2.146192212582254,
      "grad_norm": 0.4427323544427076,
      "kl": 0.123443603515625,
      "learning_rate": 4.94511598156854e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857491970062,
      "rewards/format_reward_func": 1.0,
      "step": 12800
    },
    {
      "completion_length": 226.57590579986572,
      "epoch": 2.1465275158221218,
      "grad_norm": 0.14726816307483223,
      "kl": 0.128753662109375,
      "learning_rate": 4.945090981124468e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 1.0,
      "step": 12802
    },
    {
      "completion_length": 231.0535831451416,
      "epoch": 2.1468628190619894,
      "grad_norm": 0.29032372274327023,
      "kl": 0.143157958984375,
      "learning_rate": 4.945065975050888e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857544124126,
      "rewards/format_reward_func": 1.0,
      "step": 12804
    },
    {
      "completion_length": 228.04911613464355,
      "epoch": 2.1471981223018566,
      "grad_norm": 0.12590136262960336,
      "kl": 0.145751953125,
      "learning_rate": 4.94504096334786e-07,
      "loss": 0.0001,
      "reward": 1.807142898440361,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8071428835391998,
      "rewards/format_reward_func": 1.0,
      "step": 12806
    },
    {
      "completion_length": 238.69643783569336,
      "epoch": 2.147533425541724,
      "grad_norm": 0.30009913665500965,
      "kl": 0.144683837890625,
      "learning_rate": 4.945015946015442e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428775787354,
      "rewards/format_reward_func": 1.0,
      "step": 12808
    },
    {
      "completion_length": 234.6294755935669,
      "epoch": 2.147868728781592,
      "grad_norm": 0.21758811899769895,
      "kl": 0.137908935546875,
      "learning_rate": 4.944990923053689e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 12810
    },
    {
      "completion_length": 239.12054443359375,
      "epoch": 2.1482040320214595,
      "grad_norm": 0.27609212343688166,
      "kl": 0.1478271484375,
      "learning_rate": 4.944965894462662e-07,
      "loss": 0.0001,
      "reward": 1.7857143580913544,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.785714304074645,
      "rewards/format_reward_func": 1.0,
      "step": 12812
    },
    {
      "completion_length": 240.602689743042,
      "epoch": 2.148539335261327,
      "grad_norm": 0.21616130808787837,
      "kl": 0.164459228515625,
      "learning_rate": 4.944940860242416e-07,
      "loss": 0.0002,
      "reward": 1.7857143431901932,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 12814
    },
    {
      "completion_length": 237.48661994934082,
      "epoch": 2.1488746385011943,
      "grad_norm": 0.24664055919005196,
      "kl": 0.142486572265625,
      "learning_rate": 4.94491582039301e-07,
      "loss": 0.0001,
      "reward": 1.8000000640749931,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000417232513,
      "rewards/format_reward_func": 1.0,
      "step": 12816
    },
    {
      "completion_length": 244.68304634094238,
      "epoch": 2.149209941741062,
      "grad_norm": 0.2179678551592934,
      "kl": 0.149383544921875,
      "learning_rate": 4.944890774914502e-07,
      "loss": 0.0001,
      "reward": 1.7178572192788124,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7178571671247482,
      "rewards/format_reward_func": 1.0,
      "step": 12818
    },
    {
      "completion_length": 238.2366180419922,
      "epoch": 2.1495452449809296,
      "grad_norm": 0.10679201305907572,
      "kl": 0.14337158203125,
      "learning_rate": 4.944865723806948e-07,
      "loss": 0.0001,
      "reward": 1.796428643167019,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 12820
    },
    {
      "completion_length": 233.70983123779297,
      "epoch": 2.149880548220797,
      "grad_norm": 0.3037664974555986,
      "kl": 0.140838623046875,
      "learning_rate": 4.944840667070406e-07,
      "loss": 0.0001,
      "reward": 1.8080357685685158,
      "reward_std": 0.029041885398328304,
      "rewards/equation_reward_func": 0.8098214641213417,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 12822
    },
    {
      "completion_length": 240.33929824829102,
      "epoch": 2.150215851460665,
      "grad_norm": 0.17097031750047306,
      "kl": 0.132476806640625,
      "learning_rate": 4.944815604704936e-07,
      "loss": 0.0001,
      "reward": 1.8071429058909416,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8071428872644901,
      "rewards/format_reward_func": 1.0,
      "step": 12824
    },
    {
      "completion_length": 240.99108123779297,
      "epoch": 2.1505511547005325,
      "grad_norm": 0.2717364472056746,
      "kl": 0.14019775390625,
      "learning_rate": 4.944790536710592e-07,
      "loss": 0.0001,
      "reward": 1.7589286491274834,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7625000290572643,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 12826
    },
    {
      "completion_length": 240.4062614440918,
      "epoch": 2.1508864579403997,
      "grad_norm": 0.06785661992573451,
      "kl": 0.140716552734375,
      "learning_rate": 4.944765463087435e-07,
      "loss": 0.0001,
      "reward": 1.8160714507102966,
      "reward_std": 0.017677669413387775,
      "rewards/equation_reward_func": 0.8205357566475868,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12828
    },
    {
      "completion_length": 250.36608600616455,
      "epoch": 2.1512217611802673,
      "grad_norm": 0.1515955621735846,
      "kl": 0.15386962890625,
      "learning_rate": 4.944740383835521e-07,
      "loss": 0.0002,
      "reward": 1.7357143759727478,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7357143256813288,
      "rewards/format_reward_func": 1.0,
      "step": 12830
    },
    {
      "completion_length": 242.3928680419922,
      "epoch": 2.151557064420135,
      "grad_norm": 0.1461774077036663,
      "kl": 0.129852294921875,
      "learning_rate": 4.944715298954909e-07,
      "loss": 0.0001,
      "reward": 1.8142857551574707,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857477068901,
      "rewards/format_reward_func": 1.0,
      "step": 12832
    },
    {
      "completion_length": 239.88393878936768,
      "epoch": 2.1518923676600026,
      "grad_norm": 0.11773547519069961,
      "kl": 0.15191650390625,
      "learning_rate": 4.944690208445656e-07,
      "loss": 0.0002,
      "reward": 1.8107143267989159,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.810714315623045,
      "rewards/format_reward_func": 1.0,
      "step": 12834
    },
    {
      "completion_length": 242.55358028411865,
      "epoch": 2.15222767089987,
      "grad_norm": 0.13150936149774745,
      "kl": 0.14276123046875,
      "learning_rate": 4.944665112307819e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428827941418,
      "rewards/format_reward_func": 1.0,
      "step": 12836
    },
    {
      "completion_length": 247.94644260406494,
      "epoch": 2.1525629741397374,
      "grad_norm": 0.004567143634278605,
      "kl": 0.137481689453125,
      "learning_rate": 4.944640010541457e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8000000193715096,
      "rewards/format_reward_func": 1.0,
      "step": 12838
    },
    {
      "completion_length": 244.2544755935669,
      "epoch": 2.152898277379605,
      "grad_norm": 0.17917256243490093,
      "kl": 0.1134490966796875,
      "learning_rate": 4.944614903146626e-07,
      "loss": 0.0001,
      "reward": 1.8142857775092125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857328057289,
      "rewards/format_reward_func": 1.0,
      "step": 12840
    },
    {
      "completion_length": 239.4955472946167,
      "epoch": 2.1532335806194727,
      "grad_norm": 0.166648209325707,
      "kl": 0.13653564453125,
      "learning_rate": 4.944589790123387e-07,
      "loss": 0.0001,
      "reward": 1.8107143491506577,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8107143044471741,
      "rewards/format_reward_func": 1.0,
      "step": 12842
    },
    {
      "completion_length": 244.07590579986572,
      "epoch": 2.1535688838593403,
      "grad_norm": 0.2685607135597953,
      "kl": 0.1331787109375,
      "learning_rate": 4.944564671471794e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714626312256,
      "rewards/format_reward_func": 1.0,
      "step": 12844
    },
    {
      "completion_length": 246.5491189956665,
      "epoch": 2.153904187099208,
      "grad_norm": 0.13520245836311928,
      "kl": 0.152587890625,
      "learning_rate": 4.944539547191908e-07,
      "loss": 0.0002,
      "reward": 1.739285796880722,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 1.0,
      "step": 12846
    },
    {
      "completion_length": 251.59822940826416,
      "epoch": 2.1542394903390756,
      "grad_norm": 0.20708427682175876,
      "kl": 0.1339111328125,
      "learning_rate": 4.944514417283786e-07,
      "loss": 0.0001,
      "reward": 1.6928572282195091,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.6928571816533804,
      "rewards/format_reward_func": 1.0,
      "step": 12848
    },
    {
      "completion_length": 245.05804538726807,
      "epoch": 2.1545747935789428,
      "grad_norm": 0.13092118170338693,
      "kl": 0.134552001953125,
      "learning_rate": 4.944489281747485e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 12850
    },
    {
      "completion_length": 238.1785831451416,
      "epoch": 2.1549100968188104,
      "grad_norm": 0.2020475938040691,
      "kl": 0.13726806640625,
      "learning_rate": 4.944464140583063e-07,
      "loss": 0.0001,
      "reward": 1.7616072222590446,
      "reward_std": 0.054295698180794716,
      "rewards/equation_reward_func": 0.7633928917348385,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 12852
    },
    {
      "completion_length": 248.54911613464355,
      "epoch": 2.155245400058678,
      "grad_norm": 0.26429877443982663,
      "kl": 0.129852294921875,
      "learning_rate": 4.944438993790578e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714637488127,
      "rewards/format_reward_func": 1.0,
      "step": 12854
    },
    {
      "completion_length": 250.86608505249023,
      "epoch": 2.1555807032985457,
      "grad_norm": 0.18398285158750388,
      "kl": 0.1419677734375,
      "learning_rate": 4.944413841370088e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 12856
    },
    {
      "completion_length": 244.50447368621826,
      "epoch": 2.1559160065384133,
      "grad_norm": 0.27150227309966846,
      "kl": 0.137847900390625,
      "learning_rate": 4.944388683321652e-07,
      "loss": 0.0001,
      "reward": 1.7500000596046448,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7500000316649675,
      "rewards/format_reward_func": 1.0,
      "step": 12858
    },
    {
      "completion_length": 237.32143878936768,
      "epoch": 2.1562513097782805,
      "grad_norm": 0.15710165366100282,
      "kl": 0.136749267578125,
      "learning_rate": 4.944363519645326e-07,
      "loss": 0.0001,
      "reward": 1.7178572192788124,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7178571745753288,
      "rewards/format_reward_func": 1.0,
      "step": 12860
    },
    {
      "completion_length": 239.290189743042,
      "epoch": 2.156586613018148,
      "grad_norm": 0.15330536446957824,
      "kl": 0.131866455078125,
      "learning_rate": 4.944338350341169e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 12862
    },
    {
      "completion_length": 237.24108409881592,
      "epoch": 2.1569219162580158,
      "grad_norm": 0.058822255525960855,
      "kl": 0.142547607421875,
      "learning_rate": 4.944313175409239e-07,
      "loss": 0.0001,
      "reward": 1.7928571850061417,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571850061417,
      "rewards/format_reward_func": 1.0,
      "step": 12864
    },
    {
      "completion_length": 254.0759048461914,
      "epoch": 2.1572572194978834,
      "grad_norm": 0.20590961982409228,
      "kl": 0.1185302734375,
      "learning_rate": 4.944287994849594e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714656114578,
      "rewards/format_reward_func": 1.0,
      "step": 12866
    },
    {
      "completion_length": 237.5759038925171,
      "epoch": 2.157592522737751,
      "grad_norm": 0.23532305877840948,
      "kl": 0.13934326171875,
      "learning_rate": 4.944262808662292e-07,
      "loss": 0.0001,
      "reward": 1.717857226729393,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7178571838885546,
      "rewards/format_reward_func": 1.0,
      "step": 12868
    },
    {
      "completion_length": 244.3482255935669,
      "epoch": 2.1579278259776187,
      "grad_norm": 0.19126064452006622,
      "kl": 0.137176513671875,
      "learning_rate": 4.94423761684739e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 1.0,
      "step": 12870
    },
    {
      "completion_length": 233.21876049041748,
      "epoch": 2.158263129217486,
      "grad_norm": 0.18566948164981376,
      "kl": 0.13421630859375,
      "learning_rate": 4.944212419404947e-07,
      "loss": 0.0001,
      "reward": 1.814285770058632,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857365310192,
      "rewards/format_reward_func": 1.0,
      "step": 12872
    },
    {
      "completion_length": 236.08036708831787,
      "epoch": 2.1585984324573535,
      "grad_norm": 0.2988504098040542,
      "kl": 0.147857666015625,
      "learning_rate": 4.944187216335021e-07,
      "loss": 0.0001,
      "reward": 1.836607187986374,
      "reward_std": 0.049244935624301434,
      "rewards/equation_reward_func": 0.8383928723633289,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 12874
    },
    {
      "completion_length": 240.16518783569336,
      "epoch": 2.158933735697221,
      "grad_norm": 0.17006143463215537,
      "kl": 0.138519287109375,
      "learning_rate": 4.94416200763767e-07,
      "loss": 0.0001,
      "reward": 1.7696429267525673,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7732143178582191,
      "rewards/format_reward_func": 0.9964285790920258,
      "step": 12876
    },
    {
      "completion_length": 242.79465579986572,
      "epoch": 2.1592690389370888,
      "grad_norm": 0.24022367229575953,
      "kl": 0.15008544921875,
      "learning_rate": 4.944136793312952e-07,
      "loss": 0.0002,
      "reward": 1.7857143506407738,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 12878
    },
    {
      "completion_length": 237.74554824829102,
      "epoch": 2.1596043421769564,
      "grad_norm": 0.12534665865683378,
      "kl": 0.155029296875,
      "learning_rate": 4.944111573360924e-07,
      "loss": 0.0002,
      "reward": 1.7544643729925156,
      "reward_std": 0.03409264795482159,
      "rewards/equation_reward_func": 0.7562500350177288,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 12880
    },
    {
      "completion_length": 238.2053680419922,
      "epoch": 2.159939645416824,
      "grad_norm": 0.2235627000227893,
      "kl": 0.134307861328125,
      "learning_rate": 4.944086347781646e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 12882
    },
    {
      "completion_length": 242.26340198516846,
      "epoch": 2.1602749486566912,
      "grad_norm": 0.21898008921483378,
      "kl": 0.13482666015625,
      "learning_rate": 4.944061116575173e-07,
      "loss": 0.0001,
      "reward": 1.7651786506175995,
      "reward_std": 0.029041884932667017,
      "rewards/equation_reward_func": 0.766964316368103,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 12884
    },
    {
      "completion_length": 242.3035831451416,
      "epoch": 2.160610251896559,
      "grad_norm": 0.24976873447595765,
      "kl": 0.139129638671875,
      "learning_rate": 4.944035879741567e-07,
      "loss": 0.0001,
      "reward": 1.72946435213089,
      "reward_std": 0.049244935624301434,
      "rewards/equation_reward_func": 0.7312500327825546,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 12886
    },
    {
      "completion_length": 246.10268878936768,
      "epoch": 2.1609455551364265,
      "grad_norm": 0.35250302204529516,
      "kl": 0.134246826171875,
      "learning_rate": 4.944010637280884e-07,
      "loss": 0.0001,
      "reward": 1.7580357939004898,
      "reward_std": 0.059346460737288,
      "rewards/equation_reward_func": 0.7598214596509933,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 12888
    },
    {
      "completion_length": 238.5848331451416,
      "epoch": 2.161280858376294,
      "grad_norm": 0.1638936631207704,
      "kl": 0.130126953125,
      "learning_rate": 4.943985389193182e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 12890
    },
    {
      "completion_length": 236.87947273254395,
      "epoch": 2.1616161616161618,
      "grad_norm": 0.117822403773612,
      "kl": 0.1263427734375,
      "learning_rate": 4.943960135478519e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 12892
    },
    {
      "completion_length": 240.99108219146729,
      "epoch": 2.1619514648560294,
      "grad_norm": 0.4960007888798321,
      "kl": 0.125946044921875,
      "learning_rate": 4.943934876136955e-07,
      "loss": 0.0001,
      "reward": 1.800000049173832,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 12894
    },
    {
      "completion_length": 240.76786994934082,
      "epoch": 2.1622867680958966,
      "grad_norm": 0.2388921461736386,
      "kl": 0.147216796875,
      "learning_rate": 4.943909611168546e-07,
      "loss": 0.0001,
      "reward": 1.7214286550879478,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7214285880327225,
      "rewards/format_reward_func": 1.0,
      "step": 12896
    },
    {
      "completion_length": 235.0267972946167,
      "epoch": 2.1626220713357642,
      "grad_norm": 0.1771875945580934,
      "kl": 0.130950927734375,
      "learning_rate": 4.94388434057335e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 12898
    },
    {
      "completion_length": 242.977689743042,
      "epoch": 2.162957374575632,
      "grad_norm": 0.11793385067854463,
      "kl": 0.14813232421875,
      "learning_rate": 4.943859064351426e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143081724644,
      "rewards/format_reward_func": 1.0,
      "step": 12900
    },
    {
      "completion_length": 236.16072273254395,
      "epoch": 2.1632926778154995,
      "grad_norm": 0.1048863379933829,
      "kl": 0.1263427734375,
      "learning_rate": 4.943833782502834e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 12902
    },
    {
      "completion_length": 248.91965293884277,
      "epoch": 2.163627981055367,
      "grad_norm": 0.24007095785937319,
      "kl": 0.14215087890625,
      "learning_rate": 4.94380849502763e-07,
      "loss": 0.0001,
      "reward": 1.7267858013510704,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7312500439584255,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12904
    },
    {
      "completion_length": 243.36161708831787,
      "epoch": 2.1639632842952343,
      "grad_norm": 0.22506223143518816,
      "kl": 0.137115478515625,
      "learning_rate": 4.943783201925873e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571790456772,
      "rewards/format_reward_func": 1.0,
      "step": 12906
    },
    {
      "completion_length": 237.40626049041748,
      "epoch": 2.164298587535102,
      "grad_norm": 0.09530809311521778,
      "kl": 0.143402099609375,
      "learning_rate": 4.943757903197621e-07,
      "loss": 0.0001,
      "reward": 1.835714340209961,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8357142992317677,
      "rewards/format_reward_func": 1.0,
      "step": 12908
    },
    {
      "completion_length": 242.81251049041748,
      "epoch": 2.1646338907749696,
      "grad_norm": 0.2760145257903106,
      "kl": 0.160980224609375,
      "learning_rate": 4.943732598842931e-07,
      "loss": 0.0002,
      "reward": 1.682142935693264,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.6821429096162319,
      "rewards/format_reward_func": 1.0,
      "step": 12910
    },
    {
      "completion_length": 248.42858409881592,
      "epoch": 2.1649691940148372,
      "grad_norm": 0.17348253473498154,
      "kl": 0.135223388671875,
      "learning_rate": 4.943707288861864e-07,
      "loss": 0.0001,
      "reward": 1.7232143580913544,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7276786025613546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12912
    },
    {
      "completion_length": 246.4553680419922,
      "epoch": 2.165304497254705,
      "grad_norm": 0.187496772007708,
      "kl": 0.165252685546875,
      "learning_rate": 4.943681973254476e-07,
      "loss": 0.0002,
      "reward": 1.7321429252624512,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321429122239351,
      "rewards/format_reward_func": 1.0,
      "step": 12914
    },
    {
      "completion_length": 236.97322368621826,
      "epoch": 2.1656398004945725,
      "grad_norm": 0.12459034225641151,
      "kl": 0.131591796875,
      "learning_rate": 4.943656652020825e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7535714507102966,
      "rewards/format_reward_func": 1.0,
      "step": 12916
    },
    {
      "completion_length": 238.65179443359375,
      "epoch": 2.1659751037344397,
      "grad_norm": 0.347977850319333,
      "kl": 0.138275146484375,
      "learning_rate": 4.943631325160971e-07,
      "loss": 0.0001,
      "reward": 1.7892857939004898,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 12918
    },
    {
      "completion_length": 235.53126049041748,
      "epoch": 2.1663104069743073,
      "grad_norm": 0.17829052465307105,
      "kl": 0.133636474609375,
      "learning_rate": 4.943605992674973e-07,
      "loss": 0.0001,
      "reward": 1.7000000849366188,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7000000402331352,
      "rewards/format_reward_func": 1.0,
      "step": 12920
    },
    {
      "completion_length": 242.2455472946167,
      "epoch": 2.166645710214175,
      "grad_norm": 0.20660870590652766,
      "kl": 0.1290283203125,
      "learning_rate": 4.943580654562886e-07,
      "loss": 0.0001,
      "reward": 1.7558036521077156,
      "reward_std": 0.07260471233166754,
      "rewards/equation_reward_func": 0.7571428790688515,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 12922
    },
    {
      "completion_length": 238.7053689956665,
      "epoch": 2.1669810134540426,
      "grad_norm": 0.10815500831875394,
      "kl": 0.177398681640625,
      "learning_rate": 4.943555310824772e-07,
      "loss": 0.0002,
      "reward": 1.7946429029107094,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7991071604192257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12924
    },
    {
      "completion_length": 234.91518878936768,
      "epoch": 2.1673163166939102,
      "grad_norm": 0.2615412690593561,
      "kl": 0.123870849609375,
      "learning_rate": 4.943529961460688e-07,
      "loss": 0.0001,
      "reward": 1.804464340209961,
      "reward_std": 0.03409264795482159,
      "rewards/equation_reward_func": 0.8062500320374966,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 12926
    },
    {
      "completion_length": 241.53125953674316,
      "epoch": 2.1676516199337774,
      "grad_norm": 0.23643354312928433,
      "kl": 0.138916015625,
      "learning_rate": 4.943504606470691e-07,
      "loss": 0.0001,
      "reward": 1.8392857611179352,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8392857238650322,
      "rewards/format_reward_func": 1.0,
      "step": 12928
    },
    {
      "completion_length": 236.5223331451416,
      "epoch": 2.167986923173645,
      "grad_norm": 0.20165447625301341,
      "kl": 0.13446044921875,
      "learning_rate": 4.943479245854841e-07,
      "loss": 0.0001,
      "reward": 1.7839286401867867,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7875000163912773,
      "rewards/format_reward_func": 0.9964285790920258,
      "step": 12930
    },
    {
      "completion_length": 237.37947273254395,
      "epoch": 2.1683222264135127,
      "grad_norm": 0.2801653478125558,
      "kl": 0.139373779296875,
      "learning_rate": 4.943453879613194e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8107143044471741,
      "rewards/format_reward_func": 1.0,
      "step": 12932
    },
    {
      "completion_length": 235.91965293884277,
      "epoch": 2.1686575296533803,
      "grad_norm": 0.24652515854083196,
      "kl": 0.11895751953125,
      "learning_rate": 4.943428507745811e-07,
      "loss": 0.0001,
      "reward": 1.7750000432133675,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000171363354,
      "rewards/format_reward_func": 1.0,
      "step": 12934
    },
    {
      "completion_length": 228.6294765472412,
      "epoch": 2.168992832893248,
      "grad_norm": 0.5322413719457854,
      "kl": 0.122833251953125,
      "learning_rate": 4.94340313025275e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857305705547,
      "rewards/format_reward_func": 1.0,
      "step": 12936
    },
    {
      "completion_length": 233.21429634094238,
      "epoch": 2.1693281361331156,
      "grad_norm": 0.10032357589058459,
      "kl": 0.120361328125,
      "learning_rate": 4.94337774713407e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7357143238186836,
      "rewards/format_reward_func": 1.0,
      "step": 12938
    },
    {
      "completion_length": 236.88394165039062,
      "epoch": 2.169663439372983,
      "grad_norm": 0.14355426238650287,
      "kl": 0.0960540771484375,
      "learning_rate": 4.943352358389827e-07,
      "loss": 0.0001,
      "reward": 1.7642858028411865,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 12940
    },
    {
      "completion_length": 235.64733123779297,
      "epoch": 2.1699987426128504,
      "grad_norm": 0.25435027499296264,
      "kl": 0.12823486328125,
      "learning_rate": 4.943326964020083e-07,
      "loss": 0.0001,
      "reward": 1.8089286237955093,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.8133928813040257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12942
    },
    {
      "completion_length": 244.31250953674316,
      "epoch": 2.170334045852718,
      "grad_norm": 0.2898807685825902,
      "kl": 0.119598388671875,
      "learning_rate": 4.943301564024892e-07,
      "loss": 0.0001,
      "reward": 1.742857202887535,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571581840515,
      "rewards/format_reward_func": 1.0,
      "step": 12944
    },
    {
      "completion_length": 247.62947463989258,
      "epoch": 2.1706693490925857,
      "grad_norm": 0.19631638793532102,
      "kl": 0.12933349609375,
      "learning_rate": 4.943276158404316e-07,
      "loss": 0.0001,
      "reward": 1.7000000700354576,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7000000290572643,
      "rewards/format_reward_func": 1.0,
      "step": 12946
    },
    {
      "completion_length": 243.0892972946167,
      "epoch": 2.1710046523324533,
      "grad_norm": 0.17855137823784265,
      "kl": 0.110931396484375,
      "learning_rate": 4.943250747158413e-07,
      "loss": 0.0001,
      "reward": 1.7392857819795609,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7392857596278191,
      "rewards/format_reward_func": 1.0,
      "step": 12948
    },
    {
      "completion_length": 237.7991180419922,
      "epoch": 2.1713399555723205,
      "grad_norm": 0.3307348578135927,
      "kl": 0.140533447265625,
      "learning_rate": 4.943225330287239e-07,
      "loss": 0.0001,
      "reward": 1.7714286148548126,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 12950
    },
    {
      "completion_length": 235.58036708831787,
      "epoch": 2.171675258812188,
      "grad_norm": 0.4506722857335034,
      "kl": 0.117889404296875,
      "learning_rate": 4.943199907790856e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 12952
    },
    {
      "completion_length": 236.32590293884277,
      "epoch": 2.172010562052056,
      "grad_norm": 0.20793578554397196,
      "kl": 0.1285400390625,
      "learning_rate": 4.943174479669321e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 12954
    },
    {
      "completion_length": 237.71429538726807,
      "epoch": 2.1723458652919234,
      "grad_norm": 0.15868192009752824,
      "kl": 0.105560302734375,
      "learning_rate": 4.943149045922692e-07,
      "loss": 0.0001,
      "reward": 1.7482143342494965,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7526786085218191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12956
    },
    {
      "completion_length": 237.6071538925171,
      "epoch": 2.172681168531791,
      "grad_norm": 0.18955779258601502,
      "kl": 0.109527587890625,
      "learning_rate": 4.943123606551028e-07,
      "loss": 0.0001,
      "reward": 1.8160714954137802,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.8205357380211353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12958
    },
    {
      "completion_length": 248.1562623977661,
      "epoch": 2.1730164717716587,
      "grad_norm": 0.2265311963984262,
      "kl": 0.206207275390625,
      "learning_rate": 4.943098161554388e-07,
      "loss": 0.0002,
      "reward": 1.7696429267525673,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7741071730852127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12960
    },
    {
      "completion_length": 239.08483409881592,
      "epoch": 2.173351775011526,
      "grad_norm": 0.24927354696810647,
      "kl": 0.1038360595703125,
      "learning_rate": 4.94307271093283e-07,
      "loss": 0.0001,
      "reward": 1.7750000730156898,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 12962
    },
    {
      "completion_length": 244.32143688201904,
      "epoch": 2.1736870782513935,
      "grad_norm": 0.12840740398816736,
      "kl": 0.1054229736328125,
      "learning_rate": 4.943047254686413e-07,
      "loss": 0.0001,
      "reward": 1.8035714700818062,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8035714570432901,
      "rewards/format_reward_func": 1.0,
      "step": 12964
    },
    {
      "completion_length": 238.6785831451416,
      "epoch": 2.174022381491261,
      "grad_norm": 0.185315617261064,
      "kl": 0.110504150390625,
      "learning_rate": 4.943021792815194e-07,
      "loss": 0.0001,
      "reward": 1.8214286342263222,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.821428582072258,
      "rewards/format_reward_func": 1.0,
      "step": 12966
    },
    {
      "completion_length": 242.80804824829102,
      "epoch": 2.174357684731129,
      "grad_norm": 0.48087252874133135,
      "kl": 0.13427734375,
      "learning_rate": 4.942996325319234e-07,
      "loss": 0.0001,
      "reward": 1.7625000849366188,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643051922321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12968
    },
    {
      "completion_length": 249.78126049041748,
      "epoch": 2.1746929879709964,
      "grad_norm": 0.15766073915047,
      "kl": 0.11785888671875,
      "learning_rate": 4.942970852198591e-07,
      "loss": 0.0001,
      "reward": 1.6750000715255737,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.6750000286847353,
      "rewards/format_reward_func": 1.0,
      "step": 12970
    },
    {
      "completion_length": 239.28572463989258,
      "epoch": 2.1750282912108636,
      "grad_norm": 0.09665109152968684,
      "kl": 0.13330078125,
      "learning_rate": 4.942945373453323e-07,
      "loss": 0.0001,
      "reward": 1.7589286118745804,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7633928880095482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12972
    },
    {
      "completion_length": 237.196439743042,
      "epoch": 2.1753635944507312,
      "grad_norm": 0.21211440287784858,
      "kl": 0.1326904296875,
      "learning_rate": 4.942919889083488e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000204890966,
      "rewards/format_reward_func": 1.0,
      "step": 12974
    },
    {
      "completion_length": 244.95090675354004,
      "epoch": 2.175698897690599,
      "grad_norm": 0.5472880673504554,
      "kl": 0.135406494140625,
      "learning_rate": 4.942894399089148e-07,
      "loss": 0.0001,
      "reward": 1.814285784959793,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857402563095,
      "rewards/format_reward_func": 1.0,
      "step": 12976
    },
    {
      "completion_length": 232.86608219146729,
      "epoch": 2.1760342009304665,
      "grad_norm": 0.4324622384788192,
      "kl": 0.1165771484375,
      "learning_rate": 4.942868903470357e-07,
      "loss": 0.0001,
      "reward": 1.8214286118745804,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8214285932481289,
      "rewards/format_reward_func": 1.0,
      "step": 12978
    },
    {
      "completion_length": 234.79018878936768,
      "epoch": 2.176369504170334,
      "grad_norm": 0.23861054744834512,
      "kl": 0.129150390625,
      "learning_rate": 4.942843402227178e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 12980
    },
    {
      "completion_length": 244.60715293884277,
      "epoch": 2.176704807410202,
      "grad_norm": 0.17599420950231748,
      "kl": 0.119537353515625,
      "learning_rate": 4.942817895359666e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 12982
    },
    {
      "completion_length": 242.7232265472412,
      "epoch": 2.177040110650069,
      "grad_norm": 0.20729228630689475,
      "kl": 0.131744384765625,
      "learning_rate": 4.942792382867884e-07,
      "loss": 0.0001,
      "reward": 1.7142858058214188,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7142857555299997,
      "rewards/format_reward_func": 1.0,
      "step": 12984
    },
    {
      "completion_length": 234.30804824829102,
      "epoch": 2.1773754138899366,
      "grad_norm": 0.23350022151308217,
      "kl": 0.131134033203125,
      "learning_rate": 4.942766864751886e-07,
      "loss": 0.0001,
      "reward": 1.751785784959793,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7562500350177288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12986
    },
    {
      "completion_length": 243.3303689956665,
      "epoch": 2.1777107171298042,
      "grad_norm": 0.16288105801056224,
      "kl": 0.125274658203125,
      "learning_rate": 4.942741341011733e-07,
      "loss": 0.0001,
      "reward": 1.8035714849829674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714626312256,
      "rewards/format_reward_func": 1.0,
      "step": 12988
    },
    {
      "completion_length": 238.56697177886963,
      "epoch": 2.178046020369672,
      "grad_norm": 0.17009587148192523,
      "kl": 0.134521484375,
      "learning_rate": 4.942715811647484e-07,
      "loss": 0.0001,
      "reward": 1.701785795390606,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7062500342726707,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12990
    },
    {
      "completion_length": 241.102689743042,
      "epoch": 2.1783813236095395,
      "grad_norm": 0.1392263569922293,
      "kl": 0.191131591796875,
      "learning_rate": 4.942690276659198e-07,
      "loss": 0.0002,
      "reward": 1.7964286282658577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 12992
    },
    {
      "completion_length": 241.1116180419922,
      "epoch": 2.1787166268494067,
      "grad_norm": 0.19459083800717636,
      "kl": 0.20111083984375,
      "learning_rate": 4.942664736046933e-07,
      "loss": 0.0002,
      "reward": 1.7321429401636124,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321428768336773,
      "rewards/format_reward_func": 1.0,
      "step": 12994
    },
    {
      "completion_length": 240.47322463989258,
      "epoch": 2.1790519300892743,
      "grad_norm": 0.22337487727499877,
      "kl": 0.15386962890625,
      "learning_rate": 4.942639189810748e-07,
      "loss": 0.0002,
      "reward": 1.7375000715255737,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643253087997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12996
    },
    {
      "completion_length": 247.71876335144043,
      "epoch": 2.179387233329142,
      "grad_norm": 0.39531989522954086,
      "kl": 0.17596435546875,
      "learning_rate": 4.942613637950702e-07,
      "loss": 0.0002,
      "reward": 1.7339286655187607,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.738392885774374,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 12998
    },
    {
      "completion_length": 243.2232255935669,
      "epoch": 2.1797225365690096,
      "grad_norm": 0.10995658678733815,
      "kl": 0.2958984375,
      "learning_rate": 4.942588080466854e-07,
      "loss": 0.0003,
      "reward": 1.7571429312229156,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 13000
    },
    {
      "completion_length": 231.92858123779297,
      "epoch": 2.1800578398088772,
      "grad_norm": 0.2621288294232232,
      "kl": 0.099517822265625,
      "learning_rate": 4.942562517359262e-07,
      "loss": 0.0001,
      "reward": 1.8250000551342964,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8250000216066837,
      "rewards/format_reward_func": 1.0,
      "step": 13002
    },
    {
      "completion_length": 244.7366189956665,
      "epoch": 2.180393143048745,
      "grad_norm": 0.2000719271959465,
      "kl": 0.135772705078125,
      "learning_rate": 4.942536948627986e-07,
      "loss": 0.0001,
      "reward": 1.753571517765522,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 13004
    },
    {
      "completion_length": 236.33036708831787,
      "epoch": 2.180728446288612,
      "grad_norm": 0.19829520653945196,
      "kl": 0.115936279296875,
      "learning_rate": 4.942511374273084e-07,
      "loss": 0.0001,
      "reward": 1.725000075995922,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7250000201165676,
      "rewards/format_reward_func": 1.0,
      "step": 13006
    },
    {
      "completion_length": 242.82143878936768,
      "epoch": 2.1810637495284797,
      "grad_norm": 0.23086917895470546,
      "kl": 0.17926025390625,
      "learning_rate": 4.942485794294616e-07,
      "loss": 0.0002,
      "reward": 1.8178572058677673,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8178571574389935,
      "rewards/format_reward_func": 1.0,
      "step": 13008
    },
    {
      "completion_length": 244.8125123977661,
      "epoch": 2.1813990527683473,
      "grad_norm": 0.12326582796200601,
      "kl": 0.230316162109375,
      "learning_rate": 4.942460208692639e-07,
      "loss": 0.0002,
      "reward": 1.7392857894301414,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7392857447266579,
      "rewards/format_reward_func": 1.0,
      "step": 13010
    },
    {
      "completion_length": 246.26340579986572,
      "epoch": 2.181734356008215,
      "grad_norm": 0.5111766646617298,
      "kl": 0.141357421875,
      "learning_rate": 4.942434617467213e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000149011612,
      "rewards/format_reward_func": 1.0,
      "step": 13012
    },
    {
      "completion_length": 247.73661708831787,
      "epoch": 2.1820696592480826,
      "grad_norm": 0.18897060945754657,
      "kl": 0.142791748046875,
      "learning_rate": 4.942409020618398e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7571428883820772,
      "rewards/format_reward_func": 1.0,
      "step": 13014
    },
    {
      "completion_length": 245.17858600616455,
      "epoch": 2.1824049624879502,
      "grad_norm": 0.192037173413482,
      "kl": 0.22442626953125,
      "learning_rate": 4.942383418146251e-07,
      "loss": 0.0002,
      "reward": 1.7642857804894447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 13016
    },
    {
      "completion_length": 246.90179538726807,
      "epoch": 2.1827402657278174,
      "grad_norm": 0.1740235049742812,
      "kl": 0.486602783203125,
      "learning_rate": 4.942357810050832e-07,
      "loss": 0.0005,
      "reward": 1.743303619325161,
      "reward_std": 0.04987628059461713,
      "rewards/equation_reward_func": 0.749107176437974,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 13018
    },
    {
      "completion_length": 244.92858123779297,
      "epoch": 2.183075568967685,
      "grad_norm": 0.28280715676848855,
      "kl": 0.159271240234375,
      "learning_rate": 4.9423321963322e-07,
      "loss": 0.0002,
      "reward": 1.7669643759727478,
      "reward_std": 0.04671955481171608,
      "rewards/equation_reward_func": 0.7732143066823483,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 13020
    },
    {
      "completion_length": 241.37501049041748,
      "epoch": 2.1834108722075527,
      "grad_norm": 0.19996439624931472,
      "kl": 0.172210693359375,
      "learning_rate": 4.942306576990414e-07,
      "loss": 0.0002,
      "reward": 1.7910714820027351,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7955357395112514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13022
    },
    {
      "completion_length": 249.63840579986572,
      "epoch": 2.1837461754474203,
      "grad_norm": 0.0978921822772246,
      "kl": 0.177581787109375,
      "learning_rate": 4.942280952025531e-07,
      "loss": 0.0002,
      "reward": 1.7857143357396126,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 13024
    },
    {
      "completion_length": 240.29018878936768,
      "epoch": 2.184081478687288,
      "grad_norm": 0.3820928046968436,
      "kl": 0.283355712890625,
      "learning_rate": 4.942255321437614e-07,
      "loss": 0.0003,
      "reward": 1.78035718947649,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7848214656114578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13026
    },
    {
      "completion_length": 234.20983219146729,
      "epoch": 2.1844167819271556,
      "grad_norm": 0.31466234028861156,
      "kl": 0.243011474609375,
      "learning_rate": 4.942229685226719e-07,
      "loss": 0.0002,
      "reward": 1.7642857804894447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 13028
    },
    {
      "completion_length": 241.95536613464355,
      "epoch": 2.184752085167023,
      "grad_norm": 0.14632813185984134,
      "kl": 0.114013671875,
      "learning_rate": 4.942204043392905e-07,
      "loss": 0.0001,
      "reward": 1.7625000774860382,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13030
    },
    {
      "completion_length": 238.94197845458984,
      "epoch": 2.1850873884068904,
      "grad_norm": 0.12801467517810663,
      "kl": 0.314605712890625,
      "learning_rate": 4.942178395936232e-07,
      "loss": 0.0003,
      "reward": 1.717857226729393,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7178571540862322,
      "rewards/format_reward_func": 1.0,
      "step": 13032
    },
    {
      "completion_length": 244.12054634094238,
      "epoch": 2.185422691646758,
      "grad_norm": 0.1482522554310038,
      "kl": 0.12371826171875,
      "learning_rate": 4.942152742856759e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.753571443259716,
      "rewards/format_reward_func": 1.0,
      "step": 13034
    },
    {
      "completion_length": 252.0535831451416,
      "epoch": 2.1857579948866257,
      "grad_norm": 0.328016225135122,
      "kl": 0.243499755859375,
      "learning_rate": 4.942127084154545e-07,
      "loss": 0.0002,
      "reward": 1.7750000804662704,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 13036
    },
    {
      "completion_length": 245.58483409881592,
      "epoch": 2.1860932981264933,
      "grad_norm": 0.2915103242446317,
      "kl": 0.1773681640625,
      "learning_rate": 4.942101419829649e-07,
      "loss": 0.0002,
      "reward": 1.7660715132951736,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7705357447266579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13038
    },
    {
      "completion_length": 242.22768688201904,
      "epoch": 2.1864286013663605,
      "grad_norm": 0.15696621358735263,
      "kl": 0.229949951171875,
      "learning_rate": 4.94207574988213e-07,
      "loss": 0.0002,
      "reward": 1.7035714760422707,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7035714611411095,
      "rewards/format_reward_func": 1.0,
      "step": 13040
    },
    {
      "completion_length": 249.46876430511475,
      "epoch": 2.186763904606228,
      "grad_norm": 0.16961914232860997,
      "kl": 0.256103515625,
      "learning_rate": 4.942050074312047e-07,
      "loss": 0.0003,
      "reward": 1.7553572058677673,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214484751225,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13042
    },
    {
      "completion_length": 246.27233219146729,
      "epoch": 2.187099207846096,
      "grad_norm": 0.08810562177050113,
      "kl": 0.142333984375,
      "learning_rate": 4.94202439311946e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 13044
    },
    {
      "completion_length": 237.0982255935669,
      "epoch": 2.1874345110859634,
      "grad_norm": 0.19939925649510315,
      "kl": 0.15093994140625,
      "learning_rate": 4.941998706304426e-07,
      "loss": 0.0002,
      "reward": 1.7928571924567223,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571812808514,
      "rewards/format_reward_func": 1.0,
      "step": 13046
    },
    {
      "completion_length": 240.7321538925171,
      "epoch": 2.187769814325831,
      "grad_norm": 0.37088907779026276,
      "kl": 0.181976318359375,
      "learning_rate": 4.941973013867007e-07,
      "loss": 0.0002,
      "reward": 1.782142922282219,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428831666708,
      "rewards/format_reward_func": 1.0,
      "step": 13048
    },
    {
      "completion_length": 241.3571548461914,
      "epoch": 2.1881051175656987,
      "grad_norm": 0.34147771560012263,
      "kl": 0.721923828125,
      "learning_rate": 4.94194731580726e-07,
      "loss": 0.0007,
      "reward": 1.766071505844593,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13050
    },
    {
      "completion_length": 249.03572368621826,
      "epoch": 2.188440420805566,
      "grad_norm": 0.19678197516440343,
      "kl": 0.230865478515625,
      "learning_rate": 4.941921612125246e-07,
      "loss": 0.0002,
      "reward": 1.7607143446803093,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.769642885774374,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13052
    },
    {
      "completion_length": 241.30358028411865,
      "epoch": 2.1887757240454335,
      "grad_norm": 0.18948886620622116,
      "kl": 0.227386474609375,
      "learning_rate": 4.941895902821022e-07,
      "loss": 0.0002,
      "reward": 1.8000000566244125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 13054
    },
    {
      "completion_length": 237.2455472946167,
      "epoch": 2.189111027285301,
      "grad_norm": 0.16057536488565435,
      "kl": 0.434478759765625,
      "learning_rate": 4.941870187894648e-07,
      "loss": 0.0004,
      "reward": 1.7500000521540642,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 13056
    },
    {
      "completion_length": 227.5625114440918,
      "epoch": 2.189446330525169,
      "grad_norm": 0.22919726377893454,
      "kl": 0.162689208984375,
      "learning_rate": 4.941844467346183e-07,
      "loss": 0.0002,
      "reward": 1.8107143491506577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8107143007218838,
      "rewards/format_reward_func": 1.0,
      "step": 13058
    },
    {
      "completion_length": 238.60268688201904,
      "epoch": 2.1897816337650364,
      "grad_norm": 0.208520058844818,
      "kl": 0.284759521484375,
      "learning_rate": 4.941818741175689e-07,
      "loss": 0.0003,
      "reward": 1.792857177555561,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571812808514,
      "rewards/format_reward_func": 1.0,
      "step": 13060
    },
    {
      "completion_length": 237.09822463989258,
      "epoch": 2.1901169370049036,
      "grad_norm": 0.13094832960062402,
      "kl": 0.207489013671875,
      "learning_rate": 4.941793009383221e-07,
      "loss": 0.0002,
      "reward": 1.8142857626080513,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8142857365310192,
      "rewards/format_reward_func": 1.0,
      "step": 13062
    },
    {
      "completion_length": 239.67858123779297,
      "epoch": 2.1904522402447713,
      "grad_norm": 0.12603606773818665,
      "kl": 0.319000244140625,
      "learning_rate": 4.94176727196884e-07,
      "loss": 0.0003,
      "reward": 1.7517857775092125,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500350177288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13064
    },
    {
      "completion_length": 230.8125114440918,
      "epoch": 2.190787543484639,
      "grad_norm": 0.15633282016082847,
      "kl": 0.31939697265625,
      "learning_rate": 4.941741528932606e-07,
      "loss": 0.0003,
      "reward": 1.744642935693264,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7491071633994579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13066
    },
    {
      "completion_length": 234.54465293884277,
      "epoch": 2.1911228467245065,
      "grad_norm": 0.43623124322019347,
      "kl": 0.6263427734375,
      "learning_rate": 4.941715780274578e-07,
      "loss": 0.0006,
      "reward": 1.7750000804662704,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.783928606659174,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13068
    },
    {
      "completion_length": 242.11161708831787,
      "epoch": 2.191458149964374,
      "grad_norm": 0.2560604452422528,
      "kl": 0.178070068359375,
      "learning_rate": 4.941690025994814e-07,
      "loss": 0.0002,
      "reward": 1.7357143759727478,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143089175224,
      "rewards/format_reward_func": 1.0,
      "step": 13070
    },
    {
      "completion_length": 229.65626049041748,
      "epoch": 2.191793453204242,
      "grad_norm": 0.11541466072905084,
      "kl": 0.167022705078125,
      "learning_rate": 4.941664266093375e-07,
      "loss": 0.0002,
      "reward": 1.7428572177886963,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 13072
    },
    {
      "completion_length": 238.8125114440918,
      "epoch": 2.192128756444109,
      "grad_norm": 0.1561942106742485,
      "kl": 0.19775390625,
      "learning_rate": 4.941638500570319e-07,
      "loss": 0.0002,
      "reward": 1.6517857909202576,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.6562500298023224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13074
    },
    {
      "completion_length": 237.8884048461914,
      "epoch": 2.1924640596839766,
      "grad_norm": 0.17344246627332055,
      "kl": 0.162841796875,
      "learning_rate": 4.941612729425706e-07,
      "loss": 0.0002,
      "reward": 1.8446429073810577,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8491071686148643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13076
    },
    {
      "completion_length": 241.5491189956665,
      "epoch": 2.1927993629238443,
      "grad_norm": 0.22191151876539467,
      "kl": 0.168365478515625,
      "learning_rate": 4.941586952659595e-07,
      "loss": 0.0002,
      "reward": 1.8071428909897804,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428947150707,
      "rewards/format_reward_func": 1.0,
      "step": 13078
    },
    {
      "completion_length": 245.51787090301514,
      "epoch": 2.193134666163712,
      "grad_norm": 0.41291817489040716,
      "kl": 0.185089111328125,
      "learning_rate": 4.941561170272047e-07,
      "loss": 0.0002,
      "reward": 1.7500000596046448,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 13080
    },
    {
      "completion_length": 234.9866180419922,
      "epoch": 2.1934699694035795,
      "grad_norm": 0.2561760676478586,
      "kl": 0.151092529296875,
      "learning_rate": 4.941535382263119e-07,
      "loss": 0.0002,
      "reward": 1.7642857804894447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 13082
    },
    {
      "completion_length": 228.5625123977661,
      "epoch": 2.1938052726434467,
      "grad_norm": 0.2149898909073194,
      "kl": 0.13726806640625,
      "learning_rate": 4.94150958863287e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7535714507102966,
      "rewards/format_reward_func": 1.0,
      "step": 13084
    },
    {
      "completion_length": 240.17858219146729,
      "epoch": 2.1941405758833143,
      "grad_norm": 0.1678046221919406,
      "kl": 0.171142578125,
      "learning_rate": 4.941483789381362e-07,
      "loss": 0.0002,
      "reward": 1.767857201397419,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 13086
    },
    {
      "completion_length": 234.0491180419922,
      "epoch": 2.194475879123182,
      "grad_norm": 0.11829237495922411,
      "kl": 0.215484619140625,
      "learning_rate": 4.941457984508653e-07,
      "loss": 0.0002,
      "reward": 1.7821429446339607,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 13088
    },
    {
      "completion_length": 226.1785831451416,
      "epoch": 2.1948111823630496,
      "grad_norm": 0.1562739900767488,
      "kl": 0.147369384765625,
      "learning_rate": 4.941432174014803e-07,
      "loss": 0.0001,
      "reward": 1.780357226729393,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13090
    },
    {
      "completion_length": 230.92411613464355,
      "epoch": 2.1951464856029173,
      "grad_norm": 0.16259796276652824,
      "kl": 0.198089599609375,
      "learning_rate": 4.941406357899871e-07,
      "loss": 0.0002,
      "reward": 1.7785714864730835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 1.0,
      "step": 13092
    },
    {
      "completion_length": 232.1875123977661,
      "epoch": 2.195481788842785,
      "grad_norm": 0.20673151685475438,
      "kl": 0.217559814453125,
      "learning_rate": 4.941380536163915e-07,
      "loss": 0.0002,
      "reward": 1.8089286237955093,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8133928813040257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13094
    },
    {
      "completion_length": 236.8571538925171,
      "epoch": 2.195817092082652,
      "grad_norm": 0.1114181529377033,
      "kl": 0.215423583984375,
      "learning_rate": 4.941354708806996e-07,
      "loss": 0.0002,
      "reward": 1.7107143476605415,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7107143104076385,
      "rewards/format_reward_func": 1.0,
      "step": 13096
    },
    {
      "completion_length": 236.03126430511475,
      "epoch": 2.1961523953225197,
      "grad_norm": 0.2740025408477465,
      "kl": 0.39141845703125,
      "learning_rate": 4.941328875829175e-07,
      "loss": 0.0004,
      "reward": 1.7589286267757416,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928917348385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13098
    },
    {
      "completion_length": 228.6875114440918,
      "epoch": 2.1964876985623873,
      "grad_norm": 0.0945623183408348,
      "kl": 0.166290283203125,
      "learning_rate": 4.941303037230508e-07,
      "loss": 0.0002,
      "reward": 1.8250000402331352,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8250000290572643,
      "rewards/format_reward_func": 1.0,
      "step": 13100
    },
    {
      "completion_length": 232.227689743042,
      "epoch": 2.196823001802255,
      "grad_norm": 0.22985184246306434,
      "kl": 0.164337158203125,
      "learning_rate": 4.941277193011057e-07,
      "loss": 0.0002,
      "reward": 1.7500000670552254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 13102
    },
    {
      "completion_length": 234.11608219146729,
      "epoch": 2.1971583050421226,
      "grad_norm": 0.20757059453359875,
      "kl": 0.194183349609375,
      "learning_rate": 4.94125134317088e-07,
      "loss": 0.0002,
      "reward": 1.7357143238186836,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143312692642,
      "rewards/format_reward_func": 1.0,
      "step": 13104
    },
    {
      "completion_length": 246.06697463989258,
      "epoch": 2.19749360828199,
      "grad_norm": 0.23602206759001917,
      "kl": 0.2412109375,
      "learning_rate": 4.941225487710038e-07,
      "loss": 0.0002,
      "reward": 1.7232143729925156,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7276786006987095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13106
    },
    {
      "completion_length": 237.20983123779297,
      "epoch": 2.1978289115218574,
      "grad_norm": 0.19656052106365124,
      "kl": 0.30426025390625,
      "learning_rate": 4.94119962662859e-07,
      "loss": 0.0003,
      "reward": 1.7767857760190964,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7812500242143869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13108
    },
    {
      "completion_length": 233.30804824829102,
      "epoch": 2.198164214761725,
      "grad_norm": 0.1888572394239757,
      "kl": 0.25299072265625,
      "learning_rate": 4.941173759926595e-07,
      "loss": 0.0003,
      "reward": 1.7214286625385284,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.721428606659174,
      "rewards/format_reward_func": 1.0,
      "step": 13110
    },
    {
      "completion_length": 234.5134038925171,
      "epoch": 2.1984995180015927,
      "grad_norm": 0.14359310245026907,
      "kl": 0.162506103515625,
      "learning_rate": 4.941147887604113e-07,
      "loss": 0.0002,
      "reward": 1.7808036133646965,
      "reward_std": 0.02714784862473607,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 13112
    },
    {
      "completion_length": 230.72768878936768,
      "epoch": 2.1988348212414603,
      "grad_norm": 0.1815342261677029,
      "kl": 0.1749267578125,
      "learning_rate": 4.941122009661202e-07,
      "loss": 0.0002,
      "reward": 1.7571429163217545,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 13114
    },
    {
      "completion_length": 234.0134038925171,
      "epoch": 2.199170124481328,
      "grad_norm": 0.22855059183061707,
      "kl": 0.1219482421875,
      "learning_rate": 4.941096126097926e-07,
      "loss": 0.0001,
      "reward": 1.7250000834465027,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7250000275671482,
      "rewards/format_reward_func": 1.0,
      "step": 13116
    },
    {
      "completion_length": 235.95536708831787,
      "epoch": 2.199505427721195,
      "grad_norm": 0.3166301610655401,
      "kl": 0.2043609619140625,
      "learning_rate": 4.94107023691434e-07,
      "loss": 0.0002,
      "reward": 1.757142923772335,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428921073675,
      "rewards/format_reward_func": 1.0,
      "step": 13118
    },
    {
      "completion_length": 237.70536613464355,
      "epoch": 2.199840730961063,
      "grad_norm": 0.17579007961883955,
      "kl": 0.1287078857421875,
      "learning_rate": 4.941044342110504e-07,
      "loss": 0.0001,
      "reward": 1.7196429371833801,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.724107176065445,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13120
    },
    {
      "completion_length": 240.477689743042,
      "epoch": 2.2001760342009304,
      "grad_norm": 0.25137758864888454,
      "kl": 0.1131591796875,
      "learning_rate": 4.94101844168648e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 13122
    },
    {
      "completion_length": 236.290189743042,
      "epoch": 2.200511337440798,
      "grad_norm": 0.19700262041976796,
      "kl": 0.11376953125,
      "learning_rate": 4.940992535642327e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7964286170899868,
      "rewards/format_reward_func": 1.0,
      "step": 13124
    },
    {
      "completion_length": 231.4375123977661,
      "epoch": 2.2008466406806657,
      "grad_norm": 0.07527575658926482,
      "kl": 0.156158447265625,
      "learning_rate": 4.940966623978103e-07,
      "loss": 0.0002,
      "reward": 1.8214286044239998,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8214286044239998,
      "rewards/format_reward_func": 1.0,
      "step": 13126
    },
    {
      "completion_length": 242.0625114440918,
      "epoch": 2.201181943920533,
      "grad_norm": 0.21908432403033265,
      "kl": 0.110595703125,
      "learning_rate": 4.94094070669387e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 13128
    },
    {
      "completion_length": 244.4509048461914,
      "epoch": 2.2015172471604005,
      "grad_norm": 0.19426442618048773,
      "kl": 0.130706787109375,
      "learning_rate": 4.940914783789685e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.775000024586916,
      "rewards/format_reward_func": 1.0,
      "step": 13130
    },
    {
      "completion_length": 230.9821538925171,
      "epoch": 2.201852550400268,
      "grad_norm": 0.13175834022378757,
      "kl": 0.1300048828125,
      "learning_rate": 4.940888855265611e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428943425417,
      "rewards/format_reward_func": 1.0,
      "step": 13132
    },
    {
      "completion_length": 225.47768783569336,
      "epoch": 2.202187853640136,
      "grad_norm": 0.33402959027393114,
      "kl": 0.1414337158203125,
      "learning_rate": 4.940862921121705e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 13134
    },
    {
      "completion_length": 236.03125953674316,
      "epoch": 2.2025231568800034,
      "grad_norm": 0.1890372794346963,
      "kl": 0.11383056640625,
      "learning_rate": 4.940836981358027e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714454948902,
      "rewards/format_reward_func": 1.0,
      "step": 13136
    },
    {
      "completion_length": 230.97322463989258,
      "epoch": 2.202858460119871,
      "grad_norm": 0.1619307898339516,
      "kl": 0.1507568359375,
      "learning_rate": 4.940811035974638e-07,
      "loss": 0.0002,
      "reward": 1.7642857879400253,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 13138
    },
    {
      "completion_length": 236.21429634094238,
      "epoch": 2.2031937633597383,
      "grad_norm": 0.24074576556891386,
      "kl": 0.128509521484375,
      "learning_rate": 4.940785084971597e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000186264515,
      "rewards/format_reward_func": 1.0,
      "step": 13140
    },
    {
      "completion_length": 223.4732265472412,
      "epoch": 2.203529066599606,
      "grad_norm": 0.11759453324945487,
      "kl": 0.268890380859375,
      "learning_rate": 4.940759128348963e-07,
      "loss": 0.0003,
      "reward": 1.7928571850061417,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.8017857410013676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13142
    },
    {
      "completion_length": 236.6428689956665,
      "epoch": 2.2038643698394735,
      "grad_norm": 0.17489192703639114,
      "kl": 0.14190673828125,
      "learning_rate": 4.940733166106797e-07,
      "loss": 0.0001,
      "reward": 1.7428572177886963,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571619093418,
      "rewards/format_reward_func": 1.0,
      "step": 13144
    },
    {
      "completion_length": 222.54465198516846,
      "epoch": 2.204199673079341,
      "grad_norm": 0.2115976973391719,
      "kl": 0.13714599609375,
      "learning_rate": 4.940707198245158e-07,
      "loss": 0.0001,
      "reward": 1.7571429386734962,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 13146
    },
    {
      "completion_length": 220.4330472946167,
      "epoch": 2.204534976319209,
      "grad_norm": 0.10467149392557401,
      "kl": 0.148040771484375,
      "learning_rate": 4.940681224764107e-07,
      "loss": 0.0001,
      "reward": 1.7321429252624512,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7321428880095482,
      "rewards/format_reward_func": 1.0,
      "step": 13148
    },
    {
      "completion_length": 224.23661613464355,
      "epoch": 2.2048702795590764,
      "grad_norm": 0.218368171315283,
      "kl": 0.23980712890625,
      "learning_rate": 4.940655245663702e-07,
      "loss": 0.0002,
      "reward": 1.7500000670552254,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 13150
    },
    {
      "completion_length": 230.19197463989258,
      "epoch": 2.2052055827989436,
      "grad_norm": 0.1515065738307242,
      "kl": 0.230987548828125,
      "learning_rate": 4.940629260944004e-07,
      "loss": 0.0002,
      "reward": 1.785714328289032,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 13152
    },
    {
      "completion_length": 223.79018688201904,
      "epoch": 2.2055408860388113,
      "grad_norm": 0.35081248015729055,
      "kl": 0.125213623046875,
      "learning_rate": 4.940603270605072e-07,
      "loss": 0.0001,
      "reward": 1.7946429252624512,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7991071678698063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13154
    },
    {
      "completion_length": 223.76340293884277,
      "epoch": 2.205876189278679,
      "grad_norm": 0.4487984507252899,
      "kl": 0.14996337890625,
      "learning_rate": 4.940577274646967e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7428571786731482,
      "rewards/format_reward_func": 1.0,
      "step": 13156
    },
    {
      "completion_length": 229.00893878936768,
      "epoch": 2.2062114925185465,
      "grad_norm": 0.12167153735920415,
      "kl": 0.259735107421875,
      "learning_rate": 4.940551273069748e-07,
      "loss": 0.0003,
      "reward": 1.8053571805357933,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8098214510828257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13158
    },
    {
      "completion_length": 229.25447463989258,
      "epoch": 2.206546795758414,
      "grad_norm": 0.1730710604136419,
      "kl": 0.174346923828125,
      "learning_rate": 4.940525265873475e-07,
      "loss": 0.0002,
      "reward": 1.778571493923664,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 13160
    },
    {
      "completion_length": 225.1875114440918,
      "epoch": 2.206882098998282,
      "grad_norm": 0.007503315078522894,
      "kl": 0.306854248046875,
      "learning_rate": 4.940499253058208e-07,
      "loss": 0.0003,
      "reward": 1.757142923772335,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 13162
    },
    {
      "completion_length": 228.38393878936768,
      "epoch": 2.207217402238149,
      "grad_norm": 0.16247735776919064,
      "kl": 0.292724609375,
      "learning_rate": 4.940473234624008e-07,
      "loss": 0.0003,
      "reward": 1.742857187986374,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7517857439815998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13164
    },
    {
      "completion_length": 223.4241180419922,
      "epoch": 2.2075527054780166,
      "grad_norm": 0.4386678302614182,
      "kl": 0.526611328125,
      "learning_rate": 4.940447210570932e-07,
      "loss": 0.0005,
      "reward": 1.7714286297559738,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 13166
    },
    {
      "completion_length": 229.11607933044434,
      "epoch": 2.2078880087178843,
      "grad_norm": 0.1466795277008058,
      "kl": 0.63433837890625,
      "learning_rate": 4.940421180899042e-07,
      "loss": 0.0006,
      "reward": 1.74642863124609,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464286126196384,
      "rewards/format_reward_func": 1.0,
      "step": 13168
    },
    {
      "completion_length": 223.15626049041748,
      "epoch": 2.208223311957752,
      "grad_norm": 0.1795933383348911,
      "kl": 0.19903564453125,
      "learning_rate": 4.940395145608398e-07,
      "loss": 0.0002,
      "reward": 1.7750000655651093,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 13170
    },
    {
      "completion_length": 224.09822368621826,
      "epoch": 2.2085586151976195,
      "grad_norm": 0.15875580392748082,
      "kl": 0.1873779296875,
      "learning_rate": 4.940369104699059e-07,
      "loss": 0.0002,
      "reward": 1.7785715013742447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 13172
    },
    {
      "completion_length": 237.92411708831787,
      "epoch": 2.2088939184374867,
      "grad_norm": 0.21953287765735693,
      "kl": 0.3541259765625,
      "learning_rate": 4.940343058171086e-07,
      "loss": 0.0004,
      "reward": 1.7696429044008255,
      "reward_std": 0.08333758264780045,
      "rewards/equation_reward_func": 0.7741071805357933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13174
    },
    {
      "completion_length": 226.10268878936768,
      "epoch": 2.2092292216773544,
      "grad_norm": 0.23376125371304202,
      "kl": 0.185394287109375,
      "learning_rate": 4.940317006024539e-07,
      "loss": 0.0002,
      "reward": 1.8107143267989159,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.8196428716182709,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13176
    },
    {
      "completion_length": 241.2991189956665,
      "epoch": 2.209564524917222,
      "grad_norm": 0.14580594406211703,
      "kl": 0.4073486328125,
      "learning_rate": 4.940290948259477e-07,
      "loss": 0.0004,
      "reward": 1.7267857939004898,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7312500197440386,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13178
    },
    {
      "completion_length": 227.40625858306885,
      "epoch": 2.2098998281570896,
      "grad_norm": 0.21299387186202437,
      "kl": 0.16229248046875,
      "learning_rate": 4.94026488487596e-07,
      "loss": 0.0002,
      "reward": 1.7767857760190964,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7812500260770321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13180
    },
    {
      "completion_length": 224.30804634094238,
      "epoch": 2.2102351313969573,
      "grad_norm": 0.27077338842068727,
      "kl": 0.13043212890625,
      "learning_rate": 4.940238815874049e-07,
      "loss": 0.0001,
      "reward": 1.8017857819795609,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.806250024586916,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13182
    },
    {
      "completion_length": 230.0491180419922,
      "epoch": 2.210570434636825,
      "grad_norm": 0.17648422454315202,
      "kl": 0.1396484375,
      "learning_rate": 4.940212741253803e-07,
      "loss": 0.0001,
      "reward": 1.7660715132951736,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7794643230736256,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13184
    },
    {
      "completion_length": 230.9285831451416,
      "epoch": 2.210905737876692,
      "grad_norm": 0.19626567358030633,
      "kl": 0.319732666015625,
      "learning_rate": 4.940186661015283e-07,
      "loss": 0.0003,
      "reward": 1.7857143357396126,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143394649029,
      "rewards/format_reward_func": 1.0,
      "step": 13186
    },
    {
      "completion_length": 233.7767972946167,
      "epoch": 2.2112410411165597,
      "grad_norm": 0.3021643043558887,
      "kl": 0.31884765625,
      "learning_rate": 4.940160575158549e-07,
      "loss": 0.0003,
      "reward": 1.76071435213089,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7696429062634706,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13188
    },
    {
      "completion_length": 228.15179538726807,
      "epoch": 2.2115763443564274,
      "grad_norm": 0.3841834809821566,
      "kl": 0.54290771484375,
      "learning_rate": 4.94013448368366e-07,
      "loss": 0.0005,
      "reward": 1.714285783469677,
      "reward_std": 0.12121830508112907,
      "rewards/equation_reward_func": 0.7321428917348385,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 13190
    },
    {
      "completion_length": 223.93750953674316,
      "epoch": 2.211911647596295,
      "grad_norm": 0.32209311190475476,
      "kl": 0.2110443115234375,
      "learning_rate": 4.940108386590676e-07,
      "loss": 0.0002,
      "reward": 1.714285783469677,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7232143133878708,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13192
    },
    {
      "completion_length": 235.66072463989258,
      "epoch": 2.2122469508361626,
      "grad_norm": 0.15596004633128052,
      "kl": 0.768280029296875,
      "learning_rate": 4.940082283879658e-07,
      "loss": 0.0008,
      "reward": 1.7357143461704254,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7446428947150707,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13194
    },
    {
      "completion_length": 226.04018783569336,
      "epoch": 2.21258225407603,
      "grad_norm": 0.21343340214385512,
      "kl": 0.2079620361328125,
      "learning_rate": 4.940056175550666e-07,
      "loss": 0.0002,
      "reward": 1.7875000461935997,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.791964303702116,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13196
    },
    {
      "completion_length": 229.76340293884277,
      "epoch": 2.2129175573158975,
      "grad_norm": 0.26267101558592826,
      "kl": 0.22930908203125,
      "learning_rate": 4.940030061603761e-07,
      "loss": 0.0002,
      "reward": 1.8053571730852127,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8098214641213417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13198
    },
    {
      "completion_length": 229.1384038925171,
      "epoch": 2.213252860555765,
      "grad_norm": 0.23785842627214238,
      "kl": 0.151458740234375,
      "learning_rate": 4.940003942039002e-07,
      "loss": 0.0002,
      "reward": 1.7125000581145287,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.7258928902447224,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13200
    },
    {
      "completion_length": 241.13394260406494,
      "epoch": 2.2135881637956327,
      "grad_norm": 0.2702312817671659,
      "kl": 0.29144287109375,
      "learning_rate": 4.939977816856447e-07,
      "loss": 0.0003,
      "reward": 1.7500000670552254,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7500000447034836,
      "rewards/format_reward_func": 1.0,
      "step": 13202
    },
    {
      "completion_length": 240.51340293884277,
      "epoch": 2.2139234670355004,
      "grad_norm": 0.14928824964527235,
      "kl": 0.140411376953125,
      "learning_rate": 4.939951686056161e-07,
      "loss": 0.0001,
      "reward": 1.7375000715255737,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7419643215835094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13204
    },
    {
      "completion_length": 234.3125114440918,
      "epoch": 2.214258770275368,
      "grad_norm": 0.15811116279839596,
      "kl": 0.374359130859375,
      "learning_rate": 4.939925549638201e-07,
      "loss": 0.0004,
      "reward": 1.8535714745521545,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8535714522004128,
      "rewards/format_reward_func": 1.0,
      "step": 13206
    },
    {
      "completion_length": 249.30804824829102,
      "epoch": 2.214594073515235,
      "grad_norm": 0.3955689485987965,
      "kl": 0.528076171875,
      "learning_rate": 4.939899407602627e-07,
      "loss": 0.0005,
      "reward": 1.778571456670761,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7875000406056643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13208
    },
    {
      "completion_length": 241.12501335144043,
      "epoch": 2.214929376755103,
      "grad_norm": 0.4672937520059191,
      "kl": 0.34112548828125,
      "learning_rate": 4.939873259949499e-07,
      "loss": 0.0003,
      "reward": 1.8035714700818062,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.8125000149011612,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13210
    },
    {
      "completion_length": 249.0446538925171,
      "epoch": 2.2152646799949705,
      "grad_norm": 0.09535743739495414,
      "kl": 0.265960693359375,
      "learning_rate": 4.939847106678881e-07,
      "loss": 0.0003,
      "reward": 1.7321429252624512,
      "reward_std": 0.05555839091539383,
      "rewards/equation_reward_func": 0.741071468219161,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13212
    },
    {
      "completion_length": 250.58037090301514,
      "epoch": 2.215599983234838,
      "grad_norm": 0.11639680227151614,
      "kl": 0.145843505859375,
      "learning_rate": 4.939820947790828e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714656114578,
      "rewards/format_reward_func": 1.0,
      "step": 13214
    },
    {
      "completion_length": 236.36608219146729,
      "epoch": 2.2159352864747057,
      "grad_norm": 0.22684060354639632,
      "kl": 0.2822265625,
      "learning_rate": 4.939794783285403e-07,
      "loss": 0.0003,
      "reward": 1.8410714864730835,
      "reward_std": 0.0833375845104456,
      "rewards/equation_reward_func": 0.8544643111526966,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13216
    },
    {
      "completion_length": 241.20536518096924,
      "epoch": 2.216270589714573,
      "grad_norm": 0.06232753458071066,
      "kl": 0.929351806640625,
      "learning_rate": 4.939768613162666e-07,
      "loss": 0.0009,
      "reward": 1.7696429044008255,
      "reward_std": 0.0328299580141902,
      "rewards/equation_reward_func": 0.7830357365310192,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13218
    },
    {
      "completion_length": 225.6250114440918,
      "epoch": 2.2166058929544405,
      "grad_norm": 0.17169788855031926,
      "kl": 0.16473388671875,
      "learning_rate": 4.939742437422677e-07,
      "loss": 0.0002,
      "reward": 1.8500000312924385,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.858928594738245,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13220
    },
    {
      "completion_length": 231.65179634094238,
      "epoch": 2.216941196194308,
      "grad_norm": 0.22926052211592493,
      "kl": 0.64813232421875,
      "learning_rate": 4.939716256065496e-07,
      "loss": 0.0006,
      "reward": 1.7928571999073029,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.8017857559025288,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13222
    },
    {
      "completion_length": 231.18304824829102,
      "epoch": 2.217276499434176,
      "grad_norm": 0.22491708701033208,
      "kl": 0.4256591796875,
      "learning_rate": 4.939690069091185e-07,
      "loss": 0.0004,
      "reward": 1.696428656578064,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.7053571753203869,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13224
    },
    {
      "completion_length": 225.24554443359375,
      "epoch": 2.2176118026740435,
      "grad_norm": 0.1770503668273261,
      "kl": 0.747161865234375,
      "learning_rate": 4.939663876499801e-07,
      "loss": 0.0007,
      "reward": 1.8142857775092125,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857328057289,
      "rewards/format_reward_func": 1.0,
      "step": 13226
    },
    {
      "completion_length": 240.5044755935669,
      "epoch": 2.217947105913911,
      "grad_norm": 0.1134760793769743,
      "kl": 1.253082275390625,
      "learning_rate": 4.939637678291408e-07,
      "loss": 0.0013,
      "reward": 1.7678572088479996,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.776785746216774,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13228
    },
    {
      "completion_length": 223.47322463989258,
      "epoch": 2.2182824091537783,
      "grad_norm": 0.32150448400287074,
      "kl": 0.858489990234375,
      "learning_rate": 4.939611474466063e-07,
      "loss": 0.0009,
      "reward": 1.744642935693264,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7491071671247482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13230
    },
    {
      "completion_length": 223.62947463989258,
      "epoch": 2.218617712393646,
      "grad_norm": 0.2605905038870512,
      "kl": 0.36175537109375,
      "learning_rate": 4.939585265023828e-07,
      "loss": 0.0004,
      "reward": 1.7660714909434319,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7705357521772385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13232
    },
    {
      "completion_length": 239.5937614440918,
      "epoch": 2.2189530156335135,
      "grad_norm": 0.2594317754827288,
      "kl": 0.962310791015625,
      "learning_rate": 4.939559049964764e-07,
      "loss": 0.001,
      "reward": 1.7285715118050575,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7375000398606062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13234
    },
    {
      "completion_length": 238.602689743042,
      "epoch": 2.219288318873381,
      "grad_norm": 0.2423681610989229,
      "kl": 0.459014892578125,
      "learning_rate": 4.93953282928893e-07,
      "loss": 0.0005,
      "reward": 1.7553572058677673,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214633762836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13236
    },
    {
      "completion_length": 246.2857255935669,
      "epoch": 2.219623622113249,
      "grad_norm": 0.18164710354197572,
      "kl": 0.57537841796875,
      "learning_rate": 4.939506602996388e-07,
      "loss": 0.0006,
      "reward": 1.7642857655882835,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7642857618629932,
      "rewards/format_reward_func": 1.0,
      "step": 13238
    },
    {
      "completion_length": 231.62054347991943,
      "epoch": 2.219958925353116,
      "grad_norm": 0.1970273094202194,
      "kl": 0.5147247314453125,
      "learning_rate": 4.939480371087196e-07,
      "loss": 0.0005,
      "reward": 1.8089286163449287,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.813392885029316,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13240
    },
    {
      "completion_length": 246.5625114440918,
      "epoch": 2.2202942285929836,
      "grad_norm": 0.2030433424392116,
      "kl": 0.534576416015625,
      "learning_rate": 4.939454133561415e-07,
      "loss": 0.0005,
      "reward": 1.7196429148316383,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.733035746961832,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13242
    },
    {
      "completion_length": 242.4151906967163,
      "epoch": 2.2206295318328513,
      "grad_norm": 0.1793058229309209,
      "kl": 0.400634765625,
      "learning_rate": 4.939427890419108e-07,
      "loss": 0.0004,
      "reward": 1.7482143715023994,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526785954833031,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13244
    },
    {
      "completion_length": 246.6116189956665,
      "epoch": 2.220964835072719,
      "grad_norm": 0.18217815077685284,
      "kl": 0.204193115234375,
      "learning_rate": 4.939401641660332e-07,
      "loss": 0.0002,
      "reward": 1.7535715103149414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714656114578,
      "rewards/format_reward_func": 1.0,
      "step": 13246
    },
    {
      "completion_length": 236.48661613464355,
      "epoch": 2.2213001383125865,
      "grad_norm": 0.703389472035067,
      "kl": 0.359527587890625,
      "learning_rate": 4.93937538728515e-07,
      "loss": 0.0004,
      "reward": 1.750000074505806,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7589285895228386,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13248
    },
    {
      "completion_length": 230.27233123779297,
      "epoch": 2.221635441552454,
      "grad_norm": 0.15593416735907303,
      "kl": 0.16259765625,
      "learning_rate": 4.939349127293621e-07,
      "loss": 0.0002,
      "reward": 1.7821429073810577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 13250
    },
    {
      "completion_length": 247.27679920196533,
      "epoch": 2.2219707447923214,
      "grad_norm": 0.31791197842722513,
      "kl": 0.48486328125,
      "learning_rate": 4.939322861685806e-07,
      "loss": 0.0005,
      "reward": 1.7321429252624512,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7410714700818062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13252
    },
    {
      "completion_length": 224.00447273254395,
      "epoch": 2.222306048032189,
      "grad_norm": 0.12457813021055096,
      "kl": 0.187591552734375,
      "learning_rate": 4.939296590461765e-07,
      "loss": 0.0002,
      "reward": 1.79464291036129,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7991071529686451,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13254
    },
    {
      "completion_length": 233.6607265472412,
      "epoch": 2.2226413512720566,
      "grad_norm": 0.16283221198189626,
      "kl": 0.1414794921875,
      "learning_rate": 4.939270313621559e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857506871223,
      "rewards/format_reward_func": 1.0,
      "step": 13256
    },
    {
      "completion_length": 237.87500667572021,
      "epoch": 2.2229766545119243,
      "grad_norm": 0.26208369447346597,
      "kl": 0.2811279296875,
      "learning_rate": 4.939244031165248e-07,
      "loss": 0.0003,
      "reward": 1.7857143580913544,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 13258
    },
    {
      "completion_length": 237.2009048461914,
      "epoch": 2.223311957751792,
      "grad_norm": 0.14783752676243633,
      "kl": 0.203338623046875,
      "learning_rate": 4.939217743092894e-07,
      "loss": 0.0002,
      "reward": 1.7250000685453415,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7250000350177288,
      "rewards/format_reward_func": 1.0,
      "step": 13260
    },
    {
      "completion_length": 233.46876335144043,
      "epoch": 2.223647260991659,
      "grad_norm": 0.5945979191598133,
      "kl": 0.155548095703125,
      "learning_rate": 4.939191449404555e-07,
      "loss": 0.0002,
      "reward": 1.8178571984171867,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8178571686148643,
      "rewards/format_reward_func": 1.0,
      "step": 13262
    },
    {
      "completion_length": 235.95090579986572,
      "epoch": 2.2239825642315267,
      "grad_norm": 0.08776234428861629,
      "kl": 0.2462158203125,
      "learning_rate": 4.939165150100294e-07,
      "loss": 0.0002,
      "reward": 1.707142949104309,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7071428950875998,
      "rewards/format_reward_func": 1.0,
      "step": 13264
    },
    {
      "completion_length": 233.6250114440918,
      "epoch": 2.2243178674713944,
      "grad_norm": 0.12217827536623904,
      "kl": 0.211669921875,
      "learning_rate": 4.93913884518017e-07,
      "loss": 0.0002,
      "reward": 1.7910714820027351,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7955357544124126,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13266
    },
    {
      "completion_length": 229.37947463989258,
      "epoch": 2.224653170711262,
      "grad_norm": 0.19298089285952225,
      "kl": 0.270233154296875,
      "learning_rate": 4.939112534644245e-07,
      "loss": 0.0003,
      "reward": 1.7517857924103737,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7562500201165676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13268
    },
    {
      "completion_length": 229.87054538726807,
      "epoch": 2.2249884739511296,
      "grad_norm": 0.14434383594107386,
      "kl": 0.153961181640625,
      "learning_rate": 4.939086218492577e-07,
      "loss": 0.0002,
      "reward": 1.7107143551111221,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7107143308967352,
      "rewards/format_reward_func": 1.0,
      "step": 13270
    },
    {
      "completion_length": 230.58036708831787,
      "epoch": 2.2253237771909973,
      "grad_norm": 0.16133111860810595,
      "kl": 0.159027099609375,
      "learning_rate": 4.939059896725228e-07,
      "loss": 0.0002,
      "reward": 1.8000000640749931,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 13272
    },
    {
      "completion_length": 229.78572368621826,
      "epoch": 2.2256590804308645,
      "grad_norm": 0.2167370871722688,
      "kl": 0.201263427734375,
      "learning_rate": 4.939033569342259e-07,
      "loss": 0.0002,
      "reward": 1.7714286223053932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714286185801029,
      "rewards/format_reward_func": 1.0,
      "step": 13274
    },
    {
      "completion_length": 229.45090293884277,
      "epoch": 2.225994383670732,
      "grad_norm": 0.2795147417582892,
      "kl": 0.13800048828125,
      "learning_rate": 4.939007236343732e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 1.0,
      "step": 13276
    },
    {
      "completion_length": 229.4419755935669,
      "epoch": 2.2263296869105997,
      "grad_norm": 0.26344600167128623,
      "kl": 0.23748779296875,
      "learning_rate": 4.938980897729704e-07,
      "loss": 0.0002,
      "reward": 1.775000050663948,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 13278
    },
    {
      "completion_length": 234.10715293884277,
      "epoch": 2.2266649901504674,
      "grad_norm": 0.09384786079257194,
      "kl": 0.217926025390625,
      "learning_rate": 4.938954553500238e-07,
      "loss": 0.0002,
      "reward": 1.7678571864962578,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 13280
    },
    {
      "completion_length": 221.8928680419922,
      "epoch": 2.227000293390335,
      "grad_norm": 0.009029269373749394,
      "kl": 0.191070556640625,
      "learning_rate": 4.938928203655396e-07,
      "loss": 0.0002,
      "reward": 1.846428595483303,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8464285880327225,
      "rewards/format_reward_func": 1.0,
      "step": 13282
    },
    {
      "completion_length": 229.92858219146729,
      "epoch": 2.2273355966302026,
      "grad_norm": 0.10794024296635173,
      "kl": 0.130828857421875,
      "learning_rate": 4.938901848195236e-07,
      "loss": 0.0001,
      "reward": 1.7767857611179352,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7803571745753288,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 13284
    },
    {
      "completion_length": 232.4821538925171,
      "epoch": 2.22767089987007,
      "grad_norm": 0.10866707186198227,
      "kl": 0.20367431640625,
      "learning_rate": 4.938875487119819e-07,
      "loss": 0.0002,
      "reward": 1.7642857804894447,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7642857581377029,
      "rewards/format_reward_func": 1.0,
      "step": 13286
    },
    {
      "completion_length": 235.95983219146729,
      "epoch": 2.2280062031099375,
      "grad_norm": 0.06240347822895245,
      "kl": 0.181243896484375,
      "learning_rate": 4.938849120429207e-07,
      "loss": 0.0002,
      "reward": 1.7642857655882835,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7642857488244772,
      "rewards/format_reward_func": 1.0,
      "step": 13288
    },
    {
      "completion_length": 228.5937623977661,
      "epoch": 2.228341506349805,
      "grad_norm": 0.1655080256398611,
      "kl": 0.1346435546875,
      "learning_rate": 4.93882274812346e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 13290
    },
    {
      "completion_length": 234.00000858306885,
      "epoch": 2.2286768095896727,
      "grad_norm": 0.1567514946119727,
      "kl": 0.19720458984375,
      "learning_rate": 4.938796370202639e-07,
      "loss": 0.0002,
      "reward": 1.821428619325161,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8214285783469677,
      "rewards/format_reward_func": 1.0,
      "step": 13292
    },
    {
      "completion_length": 227.3928680419922,
      "epoch": 2.2290121128295404,
      "grad_norm": 0.1766307361388533,
      "kl": 0.14697265625,
      "learning_rate": 4.938769986666804e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7500000465661287,
      "rewards/format_reward_func": 1.0,
      "step": 13294
    },
    {
      "completion_length": 238.21429920196533,
      "epoch": 2.229347416069408,
      "grad_norm": 0.09715960629355828,
      "kl": 0.177581787109375,
      "learning_rate": 4.938743597516017e-07,
      "loss": 0.0002,
      "reward": 1.7142857983708382,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7142857536673546,
      "rewards/format_reward_func": 1.0,
      "step": 13296
    },
    {
      "completion_length": 232.15179634094238,
      "epoch": 2.229682719309275,
      "grad_norm": 0.4376923634979283,
      "kl": 0.1544189453125,
      "learning_rate": 4.938717202750338e-07,
      "loss": 0.0002,
      "reward": 1.7964286357164383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7964286096394062,
      "rewards/format_reward_func": 1.0,
      "step": 13298
    },
    {
      "completion_length": 239.44644165039062,
      "epoch": 2.230018022549143,
      "grad_norm": 0.14513764982478072,
      "kl": 0.25518798828125,
      "learning_rate": 4.938690802369827e-07,
      "loss": 0.0003,
      "reward": 1.750000074505806,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 13300
    },
    {
      "completion_length": 235.77679634094238,
      "epoch": 2.2303533257890105,
      "grad_norm": 0.22726923725353268,
      "kl": 0.1492919921875,
      "learning_rate": 4.938664396374545e-07,
      "loss": 0.0001,
      "reward": 1.8000000640749931,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 13302
    },
    {
      "completion_length": 231.8794755935669,
      "epoch": 2.230688629028878,
      "grad_norm": 0.24320216391633032,
      "kl": 0.149627685546875,
      "learning_rate": 4.938637984764555e-07,
      "loss": 0.0001,
      "reward": 1.8071429058909416,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428723633289,
      "rewards/format_reward_func": 1.0,
      "step": 13304
    },
    {
      "completion_length": 245.3571538925171,
      "epoch": 2.2310239322687457,
      "grad_norm": 0.22138009336756811,
      "kl": 0.219757080078125,
      "learning_rate": 4.938611567539915e-07,
      "loss": 0.0002,
      "reward": 1.814285770058632,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857328057289,
      "rewards/format_reward_func": 1.0,
      "step": 13306
    },
    {
      "completion_length": 243.50001049041748,
      "epoch": 2.231359235508613,
      "grad_norm": 0.37270995546541225,
      "kl": 0.167022705078125,
      "learning_rate": 4.938585144700688e-07,
      "loss": 0.0002,
      "reward": 1.7767857909202576,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7812500223517418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13308
    },
    {
      "completion_length": 237.3125123977661,
      "epoch": 2.2316945387484806,
      "grad_norm": 0.3370206438808287,
      "kl": 0.197845458984375,
      "learning_rate": 4.938558716246933e-07,
      "loss": 0.0002,
      "reward": 1.8125000447034836,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8169643133878708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13310
    },
    {
      "completion_length": 238.8080472946167,
      "epoch": 2.232029841988348,
      "grad_norm": 0.10725523459583197,
      "kl": 0.18255615234375,
      "learning_rate": 4.938532282178713e-07,
      "loss": 0.0002,
      "reward": 1.7428572103381157,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571693599224,
      "rewards/format_reward_func": 1.0,
      "step": 13312
    },
    {
      "completion_length": 249.25447368621826,
      "epoch": 2.232365145228216,
      "grad_norm": 0.05706609548541911,
      "kl": 0.271392822265625,
      "learning_rate": 4.938505842496086e-07,
      "loss": 0.0003,
      "reward": 1.769642911851406,
      "reward_std": 0.012626906856894493,
      "rewards/equation_reward_func": 0.7741071805357933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13314
    },
    {
      "completion_length": 241.30358409881592,
      "epoch": 2.2327004484680835,
      "grad_norm": 0.297844963742036,
      "kl": 0.2213134765625,
      "learning_rate": 4.938479397199115e-07,
      "loss": 0.0002,
      "reward": 1.7750000804662704,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7750000394880772,
      "rewards/format_reward_func": 1.0,
      "step": 13316
    },
    {
      "completion_length": 242.61608505249023,
      "epoch": 2.233035751707951,
      "grad_norm": 0.06049902245495544,
      "kl": 0.239166259765625,
      "learning_rate": 4.93845294628786e-07,
      "loss": 0.0002,
      "reward": 1.7660714909434319,
      "reward_std": 0.017677669413387775,
      "rewards/equation_reward_func": 0.7705357484519482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13318
    },
    {
      "completion_length": 253.62500858306885,
      "epoch": 2.2333710549478183,
      "grad_norm": 0.20782510215756428,
      "kl": 0.47796630859375,
      "learning_rate": 4.938426489762382e-07,
      "loss": 0.0005,
      "reward": 1.8107143715023994,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 13320
    },
    {
      "completion_length": 241.4062623977661,
      "epoch": 2.233706358187686,
      "grad_norm": 0.15660245756764526,
      "kl": 0.2474365234375,
      "learning_rate": 4.938400027622744e-07,
      "loss": 0.0002,
      "reward": 1.800000049173832,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8000000342726707,
      "rewards/format_reward_func": 1.0,
      "step": 13322
    },
    {
      "completion_length": 239.20536994934082,
      "epoch": 2.2340416614275536,
      "grad_norm": 0.1377272999131664,
      "kl": 0.541748046875,
      "learning_rate": 4.938373559869003e-07,
      "loss": 0.0005,
      "reward": 1.7446429282426834,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13324
    },
    {
      "completion_length": 234.62054538726807,
      "epoch": 2.234376964667421,
      "grad_norm": 0.4585849996854235,
      "kl": 0.53997802734375,
      "learning_rate": 4.938347086501223e-07,
      "loss": 0.0005,
      "reward": 1.755357213318348,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214522004128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13326
    },
    {
      "completion_length": 233.2634048461914,
      "epoch": 2.234712267907289,
      "grad_norm": 0.15034751021902695,
      "kl": 0.90093994140625,
      "learning_rate": 4.938320607519464e-07,
      "loss": 0.0009,
      "reward": 1.7410714998841286,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357369035482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13328
    },
    {
      "completion_length": 249.1116180419922,
      "epoch": 2.235047571147156,
      "grad_norm": 0.18235629112200313,
      "kl": 0.187347412109375,
      "learning_rate": 4.938294122923785e-07,
      "loss": 0.0002,
      "reward": 1.7803572118282318,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7848214618861675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13330
    },
    {
      "completion_length": 238.00893878936768,
      "epoch": 2.2353828743870237,
      "grad_norm": 0.2001295789142171,
      "kl": 1.100799560546875,
      "learning_rate": 4.938267632714252e-07,
      "loss": 0.0011,
      "reward": 1.7821429148316383,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.7910714522004128,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13332
    },
    {
      "completion_length": 231.4910831451416,
      "epoch": 2.2357181776268913,
      "grad_norm": 0.7877813828478731,
      "kl": 0.62371826171875,
      "learning_rate": 4.938241136890921e-07,
      "loss": 0.0006,
      "reward": 1.753571517765522,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7625000178813934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13334
    },
    {
      "completion_length": 235.11161708831787,
      "epoch": 2.236053480866759,
      "grad_norm": 0.15697719157481013,
      "kl": 0.374969482421875,
      "learning_rate": 4.938214635453854e-07,
      "loss": 0.0004,
      "reward": 1.7464286535978317,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 13336
    },
    {
      "completion_length": 239.9241180419922,
      "epoch": 2.2363887841066266,
      "grad_norm": 0.45109904732269696,
      "kl": 0.39056396484375,
      "learning_rate": 4.938188128403114e-07,
      "loss": 0.0004,
      "reward": 1.7017858028411865,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7062500193715096,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13338
    },
    {
      "completion_length": 239.34376049041748,
      "epoch": 2.236724087346494,
      "grad_norm": 0.21322803207128538,
      "kl": 1.0438232421875,
      "learning_rate": 4.938161615738762e-07,
      "loss": 0.001,
      "reward": 1.7178571820259094,
      "reward_std": 0.04545686487108469,
      "rewards/equation_reward_func": 0.7267857324331999,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13340
    },
    {
      "completion_length": 240.7142972946167,
      "epoch": 2.2370593905863614,
      "grad_norm": 0.1442597372037303,
      "kl": 0.367919921875,
      "learning_rate": 4.938135097460856e-07,
      "loss": 0.0004,
      "reward": 1.7482143491506577,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526786103844643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13342
    },
    {
      "completion_length": 239.2053689956665,
      "epoch": 2.237394693826229,
      "grad_norm": 0.26108253632311346,
      "kl": 0.6842041015625,
      "learning_rate": 4.93810857356946e-07,
      "loss": 0.0007,
      "reward": 1.719642959535122,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7241071835160255,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13344
    },
    {
      "completion_length": 240.7366180419922,
      "epoch": 2.2377299970660967,
      "grad_norm": 0.19328804186926843,
      "kl": 0.4044189453125,
      "learning_rate": 4.938082044064634e-07,
      "loss": 0.0004,
      "reward": 1.7392857745289803,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.7482143193483353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13346
    },
    {
      "completion_length": 236.4285831451416,
      "epoch": 2.2380653003059643,
      "grad_norm": 0.10448958503500268,
      "kl": 0.347503662109375,
      "learning_rate": 4.938055508946439e-07,
      "loss": 0.0003,
      "reward": 1.7053572088479996,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7098214663565159,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13348
    },
    {
      "completion_length": 226.47322463989258,
      "epoch": 2.238400603545832,
      "grad_norm": 0.3026737005415429,
      "kl": 0.236053466796875,
      "learning_rate": 4.938028968214937e-07,
      "loss": 0.0002,
      "reward": 1.7410715147852898,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7455357499420643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13350
    },
    {
      "completion_length": 227.45090293884277,
      "epoch": 2.238735906785699,
      "grad_norm": 0.21333624860225478,
      "kl": 0.204345703125,
      "learning_rate": 4.938002421870187e-07,
      "loss": 0.0002,
      "reward": 1.741071492433548,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7455357499420643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13352
    },
    {
      "completion_length": 242.4241189956665,
      "epoch": 2.2390712100255667,
      "grad_norm": 0.11296037504989383,
      "kl": 0.187530517578125,
      "learning_rate": 4.937975869912252e-07,
      "loss": 0.0002,
      "reward": 1.7718750685453415,
      "reward_std": 0.035986683797091246,
      "rewards/equation_reward_func": 0.7776785846799612,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 13354
    },
    {
      "completion_length": 238.2500114440918,
      "epoch": 2.2394065132654344,
      "grad_norm": 0.4362049106828727,
      "kl": 0.346435546875,
      "learning_rate": 4.937949312341193e-07,
      "loss": 0.0003,
      "reward": 1.7183036357164383,
      "reward_std": 0.07513009523972869,
      "rewards/equation_reward_func": 0.733035746961832,
      "rewards/format_reward_func": 0.9852678664028645,
      "step": 13356
    },
    {
      "completion_length": 229.33929538726807,
      "epoch": 2.239741816505302,
      "grad_norm": 0.18681869868795595,
      "kl": 0.319091796875,
      "learning_rate": 4.937922749157071e-07,
      "loss": 0.0003,
      "reward": 1.750446505844593,
      "reward_std": 0.05997780757024884,
      "rewards/equation_reward_func": 0.7562500238418579,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 13358
    },
    {
      "completion_length": 230.15179634094238,
      "epoch": 2.2400771197451697,
      "grad_norm": 0.565797210429418,
      "kl": 0.379913330078125,
      "learning_rate": 4.937896180359946e-07,
      "loss": 0.0004,
      "reward": 1.7678572162985802,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7678571604192257,
      "rewards/format_reward_func": 1.0,
      "step": 13360
    },
    {
      "completion_length": 229.35715293884277,
      "epoch": 2.2404124229850373,
      "grad_norm": 0.25123464149445224,
      "kl": 0.290374755859375,
      "learning_rate": 4.937869605949881e-07,
      "loss": 0.0003,
      "reward": 1.7910714745521545,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7955357357859612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13362
    },
    {
      "completion_length": 233.9732255935669,
      "epoch": 2.2407477262249045,
      "grad_norm": 0.936629094464273,
      "kl": 0.34619140625,
      "learning_rate": 4.937843025926936e-07,
      "loss": 0.0003,
      "reward": 1.7446429282426834,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7491071578115225,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13364
    },
    {
      "completion_length": 237.1160831451416,
      "epoch": 2.241083029464772,
      "grad_norm": 0.1248168838109331,
      "kl": 0.30145263671875,
      "learning_rate": 4.937816440291172e-07,
      "loss": 0.0003,
      "reward": 1.7750000730156898,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 13366
    },
    {
      "completion_length": 224.27233123779297,
      "epoch": 2.2414183327046397,
      "grad_norm": 0.07464853245578626,
      "kl": 0.222442626953125,
      "learning_rate": 4.937789849042651e-07,
      "loss": 0.0002,
      "reward": 1.7803572043776512,
      "reward_std": 0.017677669413387775,
      "rewards/equation_reward_func": 0.7848214581608772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13368
    },
    {
      "completion_length": 218.9910831451416,
      "epoch": 2.2417536359445074,
      "grad_norm": 0.3339117384872975,
      "kl": 0.615142822265625,
      "learning_rate": 4.937763252181434e-07,
      "loss": 0.0006,
      "reward": 1.7892857789993286,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7982143089175224,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13370
    },
    {
      "completion_length": 221.57143878936768,
      "epoch": 2.242088939184375,
      "grad_norm": 0.23947230221231972,
      "kl": 0.119140625,
      "learning_rate": 4.937736649707582e-07,
      "loss": 0.0001,
      "reward": 1.853571467101574,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.8625000193715096,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13372
    },
    {
      "completion_length": 231.50447273254395,
      "epoch": 2.242424242424242,
      "grad_norm": 0.19491513244225595,
      "kl": 0.304962158203125,
      "learning_rate": 4.937710041621156e-07,
      "loss": 0.0003,
      "reward": 1.7964286133646965,
      "reward_std": 0.07576143927872181,
      "rewards/equation_reward_func": 0.805357176810503,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13374
    },
    {
      "completion_length": 217.4017972946167,
      "epoch": 2.24275954566411,
      "grad_norm": 0.2010389179975474,
      "kl": 0.448394775390625,
      "learning_rate": 4.937683427922218e-07,
      "loss": 0.0004,
      "reward": 1.775000050663948,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000152736902,
      "rewards/format_reward_func": 1.0,
      "step": 13376
    },
    {
      "completion_length": 228.98661613464355,
      "epoch": 2.2430948489039775,
      "grad_norm": 0.34878759281384486,
      "kl": 0.397796630859375,
      "learning_rate": 4.93765680861083e-07,
      "loss": 0.0004,
      "reward": 1.7767857760190964,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.781250037252903,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13378
    },
    {
      "completion_length": 231.84822463989258,
      "epoch": 2.243430152143845,
      "grad_norm": 0.40787454075194673,
      "kl": 0.804473876953125,
      "learning_rate": 4.93763018368705e-07,
      "loss": 0.0008,
      "reward": 1.7982143089175224,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.8116071708500385,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13380
    },
    {
      "completion_length": 228.08929443359375,
      "epoch": 2.2437654553837127,
      "grad_norm": 0.3151023477866274,
      "kl": 0.313934326171875,
      "learning_rate": 4.937603553150944e-07,
      "loss": 0.0003,
      "reward": 1.7803571969270706,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214600235224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13382
    },
    {
      "completion_length": 237.78572368621826,
      "epoch": 2.2441007586235804,
      "grad_norm": 0.13640015825158425,
      "kl": 0.5728759765625,
      "learning_rate": 4.937576917002569e-07,
      "loss": 0.0006,
      "reward": 1.662500061094761,
      "reward_std": 0.06313453428447247,
      "rewards/equation_reward_func": 0.6758929062634706,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13384
    },
    {
      "completion_length": 236.40179443359375,
      "epoch": 2.2444360618634476,
      "grad_norm": 0.17800346490208352,
      "kl": 0.17962646484375,
      "learning_rate": 4.937550275241989e-07,
      "loss": 0.0002,
      "reward": 1.7941965013742447,
      "reward_std": 0.05871511623263359,
      "rewards/equation_reward_func": 0.8017857484519482,
      "rewards/format_reward_func": 0.9924107193946838,
      "step": 13386
    },
    {
      "completion_length": 232.49108409881592,
      "epoch": 2.244771365103315,
      "grad_norm": 0.6126242940664247,
      "kl": 0.170166015625,
      "learning_rate": 4.937523627869264e-07,
      "loss": 0.0002,
      "reward": 1.7683036029338837,
      "reward_std": 0.054927044780924916,
      "rewards/equation_reward_func": 0.7830357439815998,
      "rewards/format_reward_func": 0.9852678664028645,
      "step": 13388
    },
    {
      "completion_length": 246.34822463989258,
      "epoch": 2.245106668343183,
      "grad_norm": 0.32475932657472756,
      "kl": 0.211639404296875,
      "learning_rate": 4.937496974884457e-07,
      "loss": 0.0002,
      "reward": 1.725000075995922,
      "reward_std": 0.11616754159331322,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 13390
    },
    {
      "completion_length": 239.3794765472412,
      "epoch": 2.2454419715830505,
      "grad_norm": 0.5876372394781416,
      "kl": 0.22247314453125,
      "learning_rate": 4.937470316287627e-07,
      "loss": 0.0002,
      "reward": 1.7875000461935997,
      "reward_std": 0.06818529404699802,
      "rewards/equation_reward_func": 0.8008928894996643,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13392
    },
    {
      "completion_length": 236.8080472946167,
      "epoch": 2.245777274822918,
      "grad_norm": 0.36167189638405006,
      "kl": 0.2952880859375,
      "learning_rate": 4.937443652078836e-07,
      "loss": 0.0003,
      "reward": 1.709375075995922,
      "reward_std": 0.04735090141184628,
      "rewards/equation_reward_func": 0.7196428962051868,
      "rewards/format_reward_func": 0.9897321499884129,
      "step": 13394
    },
    {
      "completion_length": 232.4241180419922,
      "epoch": 2.2461125780627853,
      "grad_norm": 0.2621123567593843,
      "kl": 0.179229736328125,
      "learning_rate": 4.937416982258146e-07,
      "loss": 0.0002,
      "reward": 1.7910714745521545,
      "reward_std": 0.08333758357912302,
      "rewards/equation_reward_func": 0.8044643141329288,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13396
    },
    {
      "completion_length": 227.4196538925171,
      "epoch": 2.246447881302653,
      "grad_norm": 0.15406421507840143,
      "kl": 0.264862060546875,
      "learning_rate": 4.93739030682562e-07,
      "loss": 0.0003,
      "reward": 1.8053571805357933,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.8187500312924385,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13398
    },
    {
      "completion_length": 221.07590293884277,
      "epoch": 2.2467831845425206,
      "grad_norm": 0.19532516413637616,
      "kl": 0.7318115234375,
      "learning_rate": 4.937363625781317e-07,
      "loss": 0.0007,
      "reward": 1.7982143238186836,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.802678607404232,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13400
    },
    {
      "completion_length": 219.62054634094238,
      "epoch": 2.247118487782388,
      "grad_norm": 0.1643623346466658,
      "kl": 0.149444580078125,
      "learning_rate": 4.937336939125299e-07,
      "loss": 0.0001,
      "reward": 1.7571429088711739,
      "reward_std": 0.030304577201604843,
      "rewards/equation_reward_func": 0.7660714592784643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13402
    },
    {
      "completion_length": 224.09822368621826,
      "epoch": 2.247453791022256,
      "grad_norm": 0.1959789257161625,
      "kl": 0.298797607421875,
      "learning_rate": 4.937310246857628e-07,
      "loss": 0.0003,
      "reward": 1.7857143431901932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 13404
    },
    {
      "completion_length": 221.84375953674316,
      "epoch": 2.2477890942621235,
      "grad_norm": 0.2076420808661432,
      "kl": 0.147674560546875,
      "learning_rate": 4.937283548978365e-07,
      "loss": 0.0001,
      "reward": 1.755357213318348,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7598214671015739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13406
    },
    {
      "completion_length": 223.008939743042,
      "epoch": 2.2481243975019907,
      "grad_norm": 0.14167261600573977,
      "kl": 0.41143798828125,
      "learning_rate": 4.937256845487572e-07,
      "loss": 0.0004,
      "reward": 1.7839286178350449,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7875000350177288,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 13408
    },
    {
      "completion_length": 221.56250858306885,
      "epoch": 2.2484597007418583,
      "grad_norm": 0.2508994403319969,
      "kl": 0.654296875,
      "learning_rate": 4.93723013638531e-07,
      "loss": 0.0007,
      "reward": 1.725000075995922,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.7339285966008902,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13410
    },
    {
      "completion_length": 230.81251049041748,
      "epoch": 2.248795003981726,
      "grad_norm": 0.14106200207665437,
      "kl": 0.626007080078125,
      "learning_rate": 4.93720342167164e-07,
      "loss": 0.0006,
      "reward": 1.778571479022503,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 13412
    },
    {
      "completion_length": 218.97322463989258,
      "epoch": 2.2491303072215936,
      "grad_norm": 0.36174404106793095,
      "kl": 0.28924560546875,
      "learning_rate": 4.937176701346623e-07,
      "loss": 0.0003,
      "reward": 1.778571493923664,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 13414
    },
    {
      "completion_length": 226.5491189956665,
      "epoch": 2.249465610461461,
      "grad_norm": 0.23314871303389753,
      "kl": 0.13134765625,
      "learning_rate": 4.937149975410324e-07,
      "loss": 0.0001,
      "reward": 1.7053572461009026,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7098214589059353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13416
    },
    {
      "completion_length": 213.50000953674316,
      "epoch": 2.249800913701329,
      "grad_norm": 0.27470988123417955,
      "kl": 0.274169921875,
      "learning_rate": 4.937123243862801e-07,
      "loss": 0.0003,
      "reward": 1.796428620815277,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 13418
    },
    {
      "completion_length": 219.67858028411865,
      "epoch": 2.250136216941196,
      "grad_norm": 0.3176363415668159,
      "kl": 0.127593994140625,
      "learning_rate": 4.937096506704116e-07,
      "loss": 0.0001,
      "reward": 1.8089286163449287,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.8133928887546062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13420
    },
    {
      "completion_length": 221.87501049041748,
      "epoch": 2.2504715201810637,
      "grad_norm": 0.224929363983825,
      "kl": 0.194793701171875,
      "learning_rate": 4.937069763934333e-07,
      "loss": 0.0002,
      "reward": 1.759375050663948,
      "reward_std": 0.027147849323228,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 13422
    },
    {
      "completion_length": 225.65179538726807,
      "epoch": 2.2508068234209313,
      "grad_norm": 0.018372357162259132,
      "kl": 0.30120849609375,
      "learning_rate": 4.937043015553511e-07,
      "loss": 0.0003,
      "reward": 1.7892857566475868,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 13424
    },
    {
      "completion_length": 226.90626049041748,
      "epoch": 2.251142126660799,
      "grad_norm": 0.23131515399617453,
      "kl": 0.152099609375,
      "learning_rate": 4.937016261561712e-07,
      "loss": 0.0002,
      "reward": 1.8017857670783997,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8062500264495611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13426
    },
    {
      "completion_length": 233.4910831451416,
      "epoch": 2.2514774299006666,
      "grad_norm": 0.2435184314392744,
      "kl": 0.39764404296875,
      "learning_rate": 4.936989501958997e-07,
      "loss": 0.0004,
      "reward": 1.7392857670783997,
      "reward_std": 0.0656599160283804,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 13428
    },
    {
      "completion_length": 229.040189743042,
      "epoch": 2.251812733140534,
      "grad_norm": 0.14798224341983937,
      "kl": 0.17626953125,
      "learning_rate": 4.93696273674543e-07,
      "loss": 0.0002,
      "reward": 1.817857213318348,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.817857164889574,
      "rewards/format_reward_func": 1.0,
      "step": 13430
    },
    {
      "completion_length": 234.95536708831787,
      "epoch": 2.2521480363804014,
      "grad_norm": 0.2096833069883818,
      "kl": 0.19940185546875,
      "learning_rate": 4.93693596592107e-07,
      "loss": 0.0002,
      "reward": 1.725000061094761,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7250000182539225,
      "rewards/format_reward_func": 1.0,
      "step": 13432
    },
    {
      "completion_length": 219.80358028411865,
      "epoch": 2.252483339620269,
      "grad_norm": 0.2303255478167584,
      "kl": 0.118499755859375,
      "learning_rate": 4.936909189485981e-07,
      "loss": 0.0001,
      "reward": 1.785714365541935,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 13434
    },
    {
      "completion_length": 228.52679443359375,
      "epoch": 2.2528186428601367,
      "grad_norm": 0.15056984969771808,
      "kl": 0.13458251953125,
      "learning_rate": 4.936882407440225e-07,
      "loss": 0.0001,
      "reward": 1.8035714849829674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 13436
    },
    {
      "completion_length": 234.61608219146729,
      "epoch": 2.2531539461000043,
      "grad_norm": 0.23384952531418834,
      "kl": 0.36761474609375,
      "learning_rate": 4.936855619783859e-07,
      "loss": 0.0004,
      "reward": 1.7642857730388641,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 13438
    },
    {
      "completion_length": 242.87054824829102,
      "epoch": 2.253489249339872,
      "grad_norm": 0.2586868516066045,
      "kl": 0.386260986328125,
      "learning_rate": 4.93682882651695e-07,
      "loss": 0.0004,
      "reward": 1.7625000551342964,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13440
    },
    {
      "completion_length": 235.42411708831787,
      "epoch": 2.253824552579739,
      "grad_norm": 0.15906009822520661,
      "kl": 0.144134521484375,
      "learning_rate": 4.936802027639557e-07,
      "loss": 0.0001,
      "reward": 1.789285771548748,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 13442
    },
    {
      "completion_length": 228.6562614440918,
      "epoch": 2.2541598558196068,
      "grad_norm": 0.18259206308634707,
      "kl": 0.108551025390625,
      "learning_rate": 4.936775223151742e-07,
      "loss": 0.0001,
      "reward": 1.7321429401636124,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.741071455180645,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13444
    },
    {
      "completion_length": 251.96430015563965,
      "epoch": 2.2544951590594744,
      "grad_norm": 0.0743592992559579,
      "kl": 0.2841796875,
      "learning_rate": 4.936748413053567e-07,
      "loss": 0.0003,
      "reward": 1.7517857775092125,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7562500312924385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13446
    },
    {
      "completion_length": 230.42411708831787,
      "epoch": 2.254830462299342,
      "grad_norm": 0.1739434056610246,
      "kl": 0.193389892578125,
      "learning_rate": 4.936721597345093e-07,
      "loss": 0.0002,
      "reward": 1.8107143342494965,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143193483353,
      "rewards/format_reward_func": 1.0,
      "step": 13448
    },
    {
      "completion_length": 242.21876049041748,
      "epoch": 2.2551657655392097,
      "grad_norm": 0.25470983698517347,
      "kl": 0.884552001953125,
      "learning_rate": 4.936694776026384e-07,
      "loss": 0.0009,
      "reward": 1.6821429207921028,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.6821429040282965,
      "rewards/format_reward_func": 1.0,
      "step": 13450
    },
    {
      "completion_length": 236.9955472946167,
      "epoch": 2.2555010687790773,
      "grad_norm": 0.19159469148946703,
      "kl": 0.196136474609375,
      "learning_rate": 4.9366679490975e-07,
      "loss": 0.0002,
      "reward": 1.7839286476373672,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7883928865194321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13452
    },
    {
      "completion_length": 243.32143878936768,
      "epoch": 2.2558363720189445,
      "grad_norm": 0.2567005414121196,
      "kl": 0.144256591796875,
      "learning_rate": 4.936641116558502e-07,
      "loss": 0.0001,
      "reward": 1.7821429297327995,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 13454
    },
    {
      "completion_length": 237.25893783569336,
      "epoch": 2.256171675258812,
      "grad_norm": 0.30205311916900046,
      "kl": 0.12261962890625,
      "learning_rate": 4.936614278409452e-07,
      "loss": 0.0001,
      "reward": 1.7625000774860382,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7669643051922321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13456
    },
    {
      "completion_length": 240.8928680419922,
      "epoch": 2.2565069784986798,
      "grad_norm": 0.1936254461303534,
      "kl": 0.12506103515625,
      "learning_rate": 4.936587434650414e-07,
      "loss": 0.0001,
      "reward": 1.819642886519432,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8241071626543999,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13458
    },
    {
      "completion_length": 233.42858219146729,
      "epoch": 2.2568422817385474,
      "grad_norm": 0.06795896946491352,
      "kl": 0.112396240234375,
      "learning_rate": 4.936560585281447e-07,
      "loss": 0.0001,
      "reward": 1.7732143253087997,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7776786014437675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13460
    },
    {
      "completion_length": 229.0044755935669,
      "epoch": 2.257177584978415,
      "grad_norm": 0.24394781620102302,
      "kl": 0.173187255859375,
      "learning_rate": 4.936533730302615e-07,
      "loss": 0.0002,
      "reward": 1.7535714879631996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 13462
    },
    {
      "completion_length": 226.38393783569336,
      "epoch": 2.257512888218282,
      "grad_norm": 0.26312285892319304,
      "kl": 0.642669677734375,
      "learning_rate": 4.936506869713979e-07,
      "loss": 0.0006,
      "reward": 1.7696429193019867,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.774107176810503,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13464
    },
    {
      "completion_length": 228.5000114440918,
      "epoch": 2.25784819145815,
      "grad_norm": 0.327693517526902,
      "kl": 0.1544189453125,
      "learning_rate": 4.9364800035156e-07,
      "loss": 0.0002,
      "reward": 1.7607143744826317,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7607142981141806,
      "rewards/format_reward_func": 1.0,
      "step": 13466
    },
    {
      "completion_length": 228.26340198516846,
      "epoch": 2.2581834946980175,
      "grad_norm": 0.2079338402791647,
      "kl": 0.130828857421875,
      "learning_rate": 4.936453131707542e-07,
      "loss": 0.0001,
      "reward": 1.7214286625385284,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7214286141097546,
      "rewards/format_reward_func": 1.0,
      "step": 13468
    },
    {
      "completion_length": 235.0491180419922,
      "epoch": 2.258518797937885,
      "grad_norm": 0.09090744850758828,
      "kl": 0.2340087890625,
      "learning_rate": 4.936426254289865e-07,
      "loss": 0.0002,
      "reward": 1.7696429416537285,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7741071712225676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13470
    },
    {
      "completion_length": 223.758939743042,
      "epoch": 2.2588541011777528,
      "grad_norm": 0.24446218583199866,
      "kl": 0.10858154296875,
      "learning_rate": 4.936399371262631e-07,
      "loss": 0.0001,
      "reward": 1.7571429163217545,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 13472
    },
    {
      "completion_length": 231.61608028411865,
      "epoch": 2.2591894044176204,
      "grad_norm": 0.37153122030690916,
      "kl": 0.2354736328125,
      "learning_rate": 4.936372482625902e-07,
      "loss": 0.0002,
      "reward": 1.7696429044008255,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7741071693599224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13474
    },
    {
      "completion_length": 224.7991180419922,
      "epoch": 2.2595247076574876,
      "grad_norm": 0.27591459797161505,
      "kl": 0.130615234375,
      "learning_rate": 4.936345588379742e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 13476
    },
    {
      "completion_length": 218.008939743042,
      "epoch": 2.259860010897355,
      "grad_norm": 0.19964699963299606,
      "kl": 0.11248779296875,
      "learning_rate": 4.936318688524209e-07,
      "loss": 0.0001,
      "reward": 1.7785715088248253,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 1.0,
      "step": 13478
    },
    {
      "completion_length": 222.89286613464355,
      "epoch": 2.260195314137223,
      "grad_norm": 0.19133067031787318,
      "kl": 0.102752685546875,
      "learning_rate": 4.936291783059367e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7392857372760773,
      "rewards/format_reward_func": 1.0,
      "step": 13480
    },
    {
      "completion_length": 225.29465293884277,
      "epoch": 2.2605306173770905,
      "grad_norm": 0.11899468868728356,
      "kl": 0.1959228515625,
      "learning_rate": 4.93626487198528e-07,
      "loss": 0.0002,
      "reward": 1.803571492433548,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714458674192,
      "rewards/format_reward_func": 1.0,
      "step": 13482
    },
    {
      "completion_length": 219.39733123779297,
      "epoch": 2.260865920616958,
      "grad_norm": 0.21464143001614744,
      "kl": 0.13031005859375,
      "learning_rate": 4.936237955302006e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7714285925030708,
      "rewards/format_reward_func": 1.0,
      "step": 13484
    },
    {
      "completion_length": 207.0759048461914,
      "epoch": 2.2612012238568253,
      "grad_norm": 0.25222121509954876,
      "kl": 0.114593505859375,
      "learning_rate": 4.936211033009611e-07,
      "loss": 0.0001,
      "reward": 1.7464286088943481,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464286033064127,
      "rewards/format_reward_func": 1.0,
      "step": 13486
    },
    {
      "completion_length": 217.8616180419922,
      "epoch": 2.261536527096693,
      "grad_norm": 0.15743116447756114,
      "kl": 0.11029052734375,
      "learning_rate": 4.936184105108153e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 13488
    },
    {
      "completion_length": 211.71429538726807,
      "epoch": 2.2618718303365606,
      "grad_norm": 0.33981219507890775,
      "kl": 0.143707275390625,
      "learning_rate": 4.936157171597697e-07,
      "loss": 0.0001,
      "reward": 1.8142857775092125,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8142857253551483,
      "rewards/format_reward_func": 1.0,
      "step": 13490
    },
    {
      "completion_length": 222.19197273254395,
      "epoch": 2.262207133576428,
      "grad_norm": 0.20660793689804108,
      "kl": 0.121429443359375,
      "learning_rate": 4.936130232478303e-07,
      "loss": 0.0001,
      "reward": 1.7678571864962578,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571920841932,
      "rewards/format_reward_func": 1.0,
      "step": 13492
    },
    {
      "completion_length": 216.99108028411865,
      "epoch": 2.262542436816296,
      "grad_norm": 0.18929196634838816,
      "kl": 0.161285400390625,
      "learning_rate": 4.936103287750035e-07,
      "loss": 0.0002,
      "reward": 1.7321429401636124,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321428842842579,
      "rewards/format_reward_func": 1.0,
      "step": 13494
    },
    {
      "completion_length": 211.60715103149414,
      "epoch": 2.2628777400561635,
      "grad_norm": 0.24113602668906872,
      "kl": 0.107635498046875,
      "learning_rate": 4.936076337412954e-07,
      "loss": 0.0001,
      "reward": 1.7803572043776512,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7848214395344257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13496
    },
    {
      "completion_length": 212.38393783569336,
      "epoch": 2.2632130432960307,
      "grad_norm": 0.13588768404975396,
      "kl": 0.2230224609375,
      "learning_rate": 4.936049381467121e-07,
      "loss": 0.0002,
      "reward": 1.807142898440361,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8071428909897804,
      "rewards/format_reward_func": 1.0,
      "step": 13498
    },
    {
      "completion_length": 215.57590198516846,
      "epoch": 2.2635483465358983,
      "grad_norm": 0.17110466396901336,
      "kl": 0.12445068359375,
      "learning_rate": 4.9360224199126e-07,
      "loss": 0.0001,
      "reward": 1.832142896950245,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8321428708732128,
      "rewards/format_reward_func": 1.0,
      "step": 13500
    },
    {
      "completion_length": 217.27233028411865,
      "epoch": 2.263883649775766,
      "grad_norm": 0.2187412522132426,
      "kl": 0.113922119140625,
      "learning_rate": 4.935995452749452e-07,
      "loss": 0.0001,
      "reward": 1.7178572043776512,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7258928902447224,
      "rewards/format_reward_func": 0.9919642955064774,
      "step": 13502
    },
    {
      "completion_length": 215.1205472946167,
      "epoch": 2.2642189530156336,
      "grad_norm": 0.15344155196626352,
      "kl": 0.102447509765625,
      "learning_rate": 4.935968479977738e-07,
      "loss": 0.0001,
      "reward": 1.8035714849829674,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8035714477300644,
      "rewards/format_reward_func": 1.0,
      "step": 13504
    },
    {
      "completion_length": 214.22768688201904,
      "epoch": 2.264554256255501,
      "grad_norm": 0.06149945562538831,
      "kl": 0.119384765625,
      "learning_rate": 4.935941501597523e-07,
      "loss": 0.0001,
      "reward": 1.7357143834233284,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.735714316368103,
      "rewards/format_reward_func": 1.0,
      "step": 13506
    },
    {
      "completion_length": 211.74554538726807,
      "epoch": 2.2648895594953684,
      "grad_norm": 0.06281976986005453,
      "kl": 0.101837158203125,
      "learning_rate": 4.935914517608867e-07,
      "loss": 0.0001,
      "reward": 1.8107143193483353,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.810714315623045,
      "rewards/format_reward_func": 1.0,
      "step": 13508
    },
    {
      "completion_length": 216.7410831451416,
      "epoch": 2.265224862735236,
      "grad_norm": 0.18027577339039147,
      "kl": 0.097869873046875,
      "learning_rate": 4.935887528011833e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 13510
    },
    {
      "completion_length": 220.33929538726807,
      "epoch": 2.2655601659751037,
      "grad_norm": 0.2485328935508096,
      "kl": 0.113037109375,
      "learning_rate": 4.935860532806482e-07,
      "loss": 0.0001,
      "reward": 1.8196429163217545,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8241071663796902,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13512
    },
    {
      "completion_length": 210.77679634094238,
      "epoch": 2.2658954692149713,
      "grad_norm": 0.1617038734571405,
      "kl": 0.128143310546875,
      "learning_rate": 4.935833531992877e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571604192257,
      "rewards/format_reward_func": 1.0,
      "step": 13514
    },
    {
      "completion_length": 210.24554538726807,
      "epoch": 2.266230772454839,
      "grad_norm": 0.1882055746221909,
      "kl": 0.122772216796875,
      "learning_rate": 4.93580652557108e-07,
      "loss": 0.0001,
      "reward": 1.7178572118282318,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7178571783006191,
      "rewards/format_reward_func": 1.0,
      "step": 13516
    },
    {
      "completion_length": 203.39732837677002,
      "epoch": 2.2665660756947066,
      "grad_norm": 0.05495894567000175,
      "kl": 0.1014556884765625,
      "learning_rate": 4.935779513541154e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 13518
    },
    {
      "completion_length": 212.12947273254395,
      "epoch": 2.2669013789345738,
      "grad_norm": 0.1476861958550593,
      "kl": 0.106475830078125,
      "learning_rate": 4.93575249590316e-07,
      "loss": 0.0001,
      "reward": 1.728571504354477,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714484751225,
      "rewards/format_reward_func": 1.0,
      "step": 13520
    },
    {
      "completion_length": 212.41072273254395,
      "epoch": 2.2672366821744414,
      "grad_norm": 0.32315519728920805,
      "kl": 0.111785888671875,
      "learning_rate": 4.935725472657161e-07,
      "loss": 0.0001,
      "reward": 1.8178571984171867,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8178571611642838,
      "rewards/format_reward_func": 1.0,
      "step": 13522
    },
    {
      "completion_length": 215.58036708831787,
      "epoch": 2.267571985414309,
      "grad_norm": 0.19584402312119173,
      "kl": 0.115447998046875,
      "learning_rate": 4.935698443803218e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 13524
    },
    {
      "completion_length": 207.04911613464355,
      "epoch": 2.2679072886541767,
      "grad_norm": 0.14119223662442124,
      "kl": 0.130584716796875,
      "learning_rate": 4.935671409341394e-07,
      "loss": 0.0001,
      "reward": 1.7428572177886963,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571730852127,
      "rewards/format_reward_func": 1.0,
      "step": 13526
    },
    {
      "completion_length": 209.13840007781982,
      "epoch": 2.2682425918940443,
      "grad_norm": 0.18142706946143367,
      "kl": 0.10601806640625,
      "learning_rate": 4.935644369271753e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857491970062,
      "rewards/format_reward_func": 1.0,
      "step": 13528
    },
    {
      "completion_length": 217.67858123779297,
      "epoch": 2.2685778951339115,
      "grad_norm": 0.3059452585174862,
      "kl": 0.129547119140625,
      "learning_rate": 4.935617323594355e-07,
      "loss": 0.0001,
      "reward": 1.74642863124609,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.746428593993187,
      "rewards/format_reward_func": 1.0,
      "step": 13530
    },
    {
      "completion_length": 206.62947463989258,
      "epoch": 2.268913198373779,
      "grad_norm": 0.2880253369534212,
      "kl": 0.128936767578125,
      "learning_rate": 4.935590272309261e-07,
      "loss": 0.0001,
      "reward": 1.7821428999304771,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7821428962051868,
      "rewards/format_reward_func": 1.0,
      "step": 13532
    },
    {
      "completion_length": 208.76786708831787,
      "epoch": 2.2692485016136468,
      "grad_norm": 0.2830816012342474,
      "kl": 0.1079864501953125,
      "learning_rate": 4.935563215416537e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714544355869,
      "rewards/format_reward_func": 1.0,
      "step": 13534
    },
    {
      "completion_length": 216.34376049041748,
      "epoch": 2.2695838048535144,
      "grad_norm": 0.1421908094574526,
      "kl": 0.144805908203125,
      "learning_rate": 4.935536152916243e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714380443096,
      "rewards/format_reward_func": 1.0,
      "step": 13536
    },
    {
      "completion_length": 205.62500762939453,
      "epoch": 2.269919108093382,
      "grad_norm": 0.08285467355914013,
      "kl": 0.24676513671875,
      "learning_rate": 4.935509084808441e-07,
      "loss": 0.0002,
      "reward": 1.7821429073810577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428999304771,
      "rewards/format_reward_func": 1.0,
      "step": 13538
    },
    {
      "completion_length": 211.83483123779297,
      "epoch": 2.2702544113332497,
      "grad_norm": 0.13214472741476582,
      "kl": 0.20220947265625,
      "learning_rate": 4.935482011093195e-07,
      "loss": 0.0002,
      "reward": 1.7500000819563866,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 13540
    },
    {
      "completion_length": 207.53572463989258,
      "epoch": 2.270589714573117,
      "grad_norm": 0.08028323710669062,
      "kl": 0.10394287109375,
      "learning_rate": 4.935454931770567e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143245637417,
      "rewards/format_reward_func": 1.0,
      "step": 13542
    },
    {
      "completion_length": 206.92858028411865,
      "epoch": 2.2709250178129845,
      "grad_norm": 0.18685389674187627,
      "kl": 0.113983154296875,
      "learning_rate": 4.935427846840617e-07,
      "loss": 0.0001,
      "reward": 1.7678572237491608,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 13544
    },
    {
      "completion_length": 202.05358028411865,
      "epoch": 2.271260321052852,
      "grad_norm": 0.3338371078997882,
      "kl": 0.1292724609375,
      "learning_rate": 4.935400756303411e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 13546
    },
    {
      "completion_length": 211.33929538726807,
      "epoch": 2.2715956242927198,
      "grad_norm": 0.18042370161512275,
      "kl": 0.106903076171875,
      "learning_rate": 4.935373660159008e-07,
      "loss": 0.0001,
      "reward": 1.7785714715719223,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 13548
    },
    {
      "completion_length": 212.53572463989258,
      "epoch": 2.2719309275325874,
      "grad_norm": 0.15448487921955562,
      "kl": 0.162017822265625,
      "learning_rate": 4.935346558407472e-07,
      "loss": 0.0002,
      "reward": 1.782142922282219,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 13550
    },
    {
      "completion_length": 207.77233028411865,
      "epoch": 2.2722662307724546,
      "grad_norm": 0.1929360171540079,
      "kl": 0.11138916015625,
      "learning_rate": 4.935319451048866e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714507102966,
      "rewards/format_reward_func": 1.0,
      "step": 13552
    },
    {
      "completion_length": 207.915189743042,
      "epoch": 2.2726015340123222,
      "grad_norm": 0.22371807794920298,
      "kl": 0.11895751953125,
      "learning_rate": 4.935292338083251e-07,
      "loss": 0.0001,
      "reward": 1.850000038743019,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8500000275671482,
      "rewards/format_reward_func": 1.0,
      "step": 13554
    },
    {
      "completion_length": 215.48215293884277,
      "epoch": 2.27293683725219,
      "grad_norm": 0.2786729445248081,
      "kl": 0.112701416015625,
      "learning_rate": 4.93526521951069e-07,
      "loss": 0.0001,
      "reward": 1.7750000432133675,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000469386578,
      "rewards/format_reward_func": 1.0,
      "step": 13556
    },
    {
      "completion_length": 217.72768592834473,
      "epoch": 2.2732721404920575,
      "grad_norm": 0.1350406475402067,
      "kl": 0.15032958984375,
      "learning_rate": 4.935238095331246e-07,
      "loss": 0.0002,
      "reward": 1.7714286223053932,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 13558
    },
    {
      "completion_length": 208.49107933044434,
      "epoch": 2.273607443731925,
      "grad_norm": 0.3304221891155211,
      "kl": 0.2548828125,
      "learning_rate": 4.935210965544981e-07,
      "loss": 0.0003,
      "reward": 1.7892857789993286,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 13560
    },
    {
      "completion_length": 215.14733028411865,
      "epoch": 2.2739427469717928,
      "grad_norm": 0.2805466392633232,
      "kl": 0.131866455078125,
      "learning_rate": 4.935183830151958e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 13562
    },
    {
      "completion_length": 203.00000858306885,
      "epoch": 2.2742780502116604,
      "grad_norm": 0.2473474152110955,
      "kl": 0.250518798828125,
      "learning_rate": 4.935156689152238e-07,
      "loss": 0.0002,
      "reward": 1.8392857313156128,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8392857350409031,
      "rewards/format_reward_func": 1.0,
      "step": 13564
    },
    {
      "completion_length": 219.06697273254395,
      "epoch": 2.2746133534515276,
      "grad_norm": 0.5644412380449078,
      "kl": 0.20465087890625,
      "learning_rate": 4.935129542545885e-07,
      "loss": 0.0002,
      "reward": 1.809821479022503,
      "reward_std": 0.04671955434605479,
      "rewards/equation_reward_func": 0.8160714544355869,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 13566
    },
    {
      "completion_length": 213.90179538726807,
      "epoch": 2.2749486566913952,
      "grad_norm": 0.11572995366572475,
      "kl": 0.11407470703125,
      "learning_rate": 4.93510239033296e-07,
      "loss": 0.0001,
      "reward": 1.7428572252392769,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571693599224,
      "rewards/format_reward_func": 1.0,
      "step": 13568
    },
    {
      "completion_length": 216.8973331451416,
      "epoch": 2.275283959931263,
      "grad_norm": 0.1595908587723372,
      "kl": 0.116973876953125,
      "learning_rate": 4.935075232513528e-07,
      "loss": 0.0001,
      "reward": 1.8285714760422707,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.8375000096857548,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13570
    },
    {
      "completion_length": 217.70983123779297,
      "epoch": 2.2756192631711305,
      "grad_norm": 0.49599002369885,
      "kl": 0.1279296875,
      "learning_rate": 4.935048069087648e-07,
      "loss": 0.0001,
      "reward": 1.814285770058632,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857365310192,
      "rewards/format_reward_func": 1.0,
      "step": 13572
    },
    {
      "completion_length": 224.6071548461914,
      "epoch": 2.275954566410998,
      "grad_norm": 0.21004672028786117,
      "kl": 0.272674560546875,
      "learning_rate": 4.935020900055386e-07,
      "loss": 0.0003,
      "reward": 1.7625000551342964,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13574
    },
    {
      "completion_length": 224.84822463989258,
      "epoch": 2.2762898696508653,
      "grad_norm": 0.29430683920252254,
      "kl": 0.14007568359375,
      "learning_rate": 4.934993725416803e-07,
      "loss": 0.0001,
      "reward": 1.7607143223285675,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7696428783237934,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13576
    },
    {
      "completion_length": 224.74108409881592,
      "epoch": 2.276625172890733,
      "grad_norm": 0.1017204526352656,
      "kl": 0.1552734375,
      "learning_rate": 4.93496654517196e-07,
      "loss": 0.0002,
      "reward": 1.7535714954137802,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7625000216066837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13578
    },
    {
      "completion_length": 225.508939743042,
      "epoch": 2.2769604761306006,
      "grad_norm": 0.09881116987726597,
      "kl": 0.22412109375,
      "learning_rate": 4.934939359320924e-07,
      "loss": 0.0002,
      "reward": 1.8071429207921028,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.8160714544355869,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13580
    },
    {
      "completion_length": 218.7098331451416,
      "epoch": 2.2772957793704682,
      "grad_norm": 0.1220410963795954,
      "kl": 0.153106689453125,
      "learning_rate": 4.934912167863752e-07,
      "loss": 0.0002,
      "reward": 1.7642857804894447,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 13582
    },
    {
      "completion_length": 231.77679538726807,
      "epoch": 2.277631082610336,
      "grad_norm": 0.39462603179441164,
      "kl": 0.32989501953125,
      "learning_rate": 4.934884970800511e-07,
      "loss": 0.0003,
      "reward": 1.7303572073578835,
      "reward_std": 0.05808377172797918,
      "rewards/equation_reward_func": 0.7437500320374966,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13584
    },
    {
      "completion_length": 236.2991180419922,
      "epoch": 2.2779663858502035,
      "grad_norm": 0.14974869337815552,
      "kl": 0.371826171875,
      "learning_rate": 4.934857768131261e-07,
      "loss": 0.0004,
      "reward": 1.7214286550879478,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7303571682423353,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13586
    },
    {
      "completion_length": 225.37054538726807,
      "epoch": 2.2783016890900707,
      "grad_norm": 0.11756198998018,
      "kl": 0.312469482421875,
      "learning_rate": 4.934830559856067e-07,
      "loss": 0.0003,
      "reward": 1.735714390873909,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7357143219560385,
      "rewards/format_reward_func": 1.0,
      "step": 13588
    },
    {
      "completion_length": 231.43751049041748,
      "epoch": 2.2786369923299383,
      "grad_norm": 0.22023421549487873,
      "kl": 0.3299560546875,
      "learning_rate": 4.934803345974989e-07,
      "loss": 0.0003,
      "reward": 1.8107143267989159,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.8196428865194321,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13590
    },
    {
      "completion_length": 238.9687614440918,
      "epoch": 2.278972295569806,
      "grad_norm": 0.14824793990543636,
      "kl": 0.301483154296875,
      "learning_rate": 4.934776126488091e-07,
      "loss": 0.0003,
      "reward": 1.7500000521540642,
      "reward_std": 0.03030457627028227,
      "rewards/equation_reward_func": 0.7589285969734192,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13592
    },
    {
      "completion_length": 244.1294755935669,
      "epoch": 2.2793075988096736,
      "grad_norm": 15.409615537389927,
      "kl": 2.38897705078125,
      "learning_rate": 4.934748901395436e-07,
      "loss": 0.0024,
      "reward": 1.7214286550879478,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7303571775555611,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13594
    },
    {
      "completion_length": 227.65179634094238,
      "epoch": 2.2796429020495412,
      "grad_norm": 0.24654334349817073,
      "kl": 0.198028564453125,
      "learning_rate": 4.934721670697087e-07,
      "loss": 0.0002,
      "reward": 1.8125000223517418,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.816964328289032,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13596
    },
    {
      "completion_length": 235.83929920196533,
      "epoch": 2.2799782052894084,
      "grad_norm": 0.30922367505226184,
      "kl": 0.369232177734375,
      "learning_rate": 4.934694434393105e-07,
      "loss": 0.0004,
      "reward": 1.719642959535122,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7241071797907352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13598
    },
    {
      "completion_length": 237.27679443359375,
      "epoch": 2.280313508529276,
      "grad_norm": 0.15791678668274603,
      "kl": 0.239410400390625,
      "learning_rate": 4.934667192483553e-07,
      "loss": 0.0002,
      "reward": 1.757142923772335,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571429014205933,
      "rewards/format_reward_func": 1.0,
      "step": 13600
    },
    {
      "completion_length": 230.6384038925171,
      "epoch": 2.2806488117691437,
      "grad_norm": 0.2840192280306608,
      "kl": 0.96710205078125,
      "learning_rate": 4.934639944968496e-07,
      "loss": 0.001,
      "reward": 1.7285715341567993,
      "reward_std": 0.0707106776535511,
      "rewards/equation_reward_func": 0.7375000305473804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13602
    },
    {
      "completion_length": 225.9732265472412,
      "epoch": 2.2809841150090113,
      "grad_norm": 0.19712667125561653,
      "kl": 0.264129638671875,
      "learning_rate": 4.934612691847994e-07,
      "loss": 0.0003,
      "reward": 1.7142857983708382,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7142857573926449,
      "rewards/format_reward_func": 1.0,
      "step": 13604
    },
    {
      "completion_length": 228.01340293884277,
      "epoch": 2.281319418248879,
      "grad_norm": 0.514089876213886,
      "kl": 0.3343505859375,
      "learning_rate": 4.934585433122112e-07,
      "loss": 0.0003,
      "reward": 1.7428572177886963,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 13606
    },
    {
      "completion_length": 233.58483219146729,
      "epoch": 2.2816547214887466,
      "grad_norm": 0.2498944000336924,
      "kl": 0.374603271484375,
      "learning_rate": 4.934558168790912e-07,
      "loss": 0.0004,
      "reward": 1.7218750789761543,
      "reward_std": 0.04987628129310906,
      "rewards/equation_reward_func": 0.7276785895228386,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 13608
    },
    {
      "completion_length": 234.68304920196533,
      "epoch": 2.281990024728614,
      "grad_norm": 0.4071963245632583,
      "kl": 0.64990234375,
      "learning_rate": 4.934530898854456e-07,
      "loss": 0.0006,
      "reward": 1.7714286297559738,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7803571596741676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13610
    },
    {
      "completion_length": 239.5089406967163,
      "epoch": 2.2823253279684814,
      "grad_norm": 0.1957224771809233,
      "kl": 0.480499267578125,
      "learning_rate": 4.934503623312806e-07,
      "loss": 0.0005,
      "reward": 1.7428572103381157,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7517857477068901,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13612
    },
    {
      "completion_length": 225.28126049041748,
      "epoch": 2.282660631208349,
      "grad_norm": 0.05916231324178276,
      "kl": 0.1220703125,
      "learning_rate": 4.934476342166026e-07,
      "loss": 0.0001,
      "reward": 1.8053571581840515,
      "reward_std": 0.012626906856894493,
      "rewards/equation_reward_func": 0.8098214715719223,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13614
    },
    {
      "completion_length": 228.18304634094238,
      "epoch": 2.2829959344482167,
      "grad_norm": 0.07388420505371393,
      "kl": 0.17498779296875,
      "learning_rate": 4.93444905541418e-07,
      "loss": 0.0002,
      "reward": 1.7553571984171867,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7598214671015739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13616
    },
    {
      "completion_length": 243.58036613464355,
      "epoch": 2.2833312376880843,
      "grad_norm": 0.20886338374905272,
      "kl": 0.190582275390625,
      "learning_rate": 4.93442176305733e-07,
      "loss": 0.0002,
      "reward": 1.7196429520845413,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.7241071723401546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13618
    },
    {
      "completion_length": 244.8705472946167,
      "epoch": 2.2836665409279515,
      "grad_norm": 0.16759168247428288,
      "kl": 0.393280029296875,
      "learning_rate": 4.934394465095537e-07,
      "loss": 0.0004,
      "reward": 1.7375000715255737,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7508928924798965,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13620
    },
    {
      "completion_length": 236.58482933044434,
      "epoch": 2.284001844167819,
      "grad_norm": 0.23261784632714255,
      "kl": 0.28619384765625,
      "learning_rate": 4.934367161528866e-07,
      "loss": 0.0003,
      "reward": 1.789285771548748,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7982142977416515,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13622
    },
    {
      "completion_length": 242.65625953674316,
      "epoch": 2.284337147407687,
      "grad_norm": 0.07630423716878422,
      "kl": 0.18927001953125,
      "learning_rate": 4.93433985235738e-07,
      "loss": 0.0002,
      "reward": 1.8071429133415222,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8071428723633289,
      "rewards/format_reward_func": 1.0,
      "step": 13624
    },
    {
      "completion_length": 249.1651906967163,
      "epoch": 2.2846724506475544,
      "grad_norm": 0.13332822719857498,
      "kl": 0.256378173828125,
      "learning_rate": 4.934312537581141e-07,
      "loss": 0.0003,
      "reward": 1.7375000789761543,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7419643178582191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13626
    },
    {
      "completion_length": 246.47769165039062,
      "epoch": 2.285007753887422,
      "grad_norm": 0.2380732736718792,
      "kl": 0.46343994140625,
      "learning_rate": 4.93428521720021e-07,
      "loss": 0.0005,
      "reward": 1.8089286237955093,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8133928813040257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13628
    },
    {
      "completion_length": 257.6339387893677,
      "epoch": 2.2853430571272897,
      "grad_norm": 0.14169497889649182,
      "kl": 0.411865234375,
      "learning_rate": 4.934257891214653e-07,
      "loss": 0.0004,
      "reward": 1.7625000700354576,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.775892898440361,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13630
    },
    {
      "completion_length": 248.64733505249023,
      "epoch": 2.285678360367157,
      "grad_norm": 0.13220287571089961,
      "kl": 0.20489501953125,
      "learning_rate": 4.934230559624533e-07,
      "loss": 0.0002,
      "reward": 1.7321429029107094,
      "reward_std": 0.025253813713788986,
      "rewards/equation_reward_func": 0.7410714533179998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13632
    },
    {
      "completion_length": 255.30358409881592,
      "epoch": 2.2860136636070245,
      "grad_norm": 0.18408713303571822,
      "kl": 0.283905029296875,
      "learning_rate": 4.93420322242991e-07,
      "loss": 0.0003,
      "reward": 1.7517857775092125,
      "reward_std": 0.08838834706693888,
      "rewards/equation_reward_func": 0.774107176810503,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 13634
    },
    {
      "completion_length": 252.05804824829102,
      "epoch": 2.286348966846892,
      "grad_norm": 0.164943595184681,
      "kl": 0.294219970703125,
      "learning_rate": 4.934175879630849e-07,
      "loss": 0.0003,
      "reward": 1.7214286774396896,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.730357188731432,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13636
    },
    {
      "completion_length": 251.68751049041748,
      "epoch": 2.28668427008676,
      "grad_norm": 0.08371264433430739,
      "kl": 0.379974365234375,
      "learning_rate": 4.934148531227413e-07,
      "loss": 0.0004,
      "reward": 1.7196429297327995,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.724107176065445,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13638
    },
    {
      "completion_length": 238.352689743042,
      "epoch": 2.2870195733266274,
      "grad_norm": 0.22735480378235184,
      "kl": 0.209197998046875,
      "learning_rate": 4.934121177219664e-07,
      "loss": 0.0002,
      "reward": 1.8339286148548126,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8383928835391998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13640
    },
    {
      "completion_length": 238.80358505249023,
      "epoch": 2.2873548765664946,
      "grad_norm": 0.19841604872169602,
      "kl": 0.206878662109375,
      "learning_rate": 4.934093817607666e-07,
      "loss": 0.0002,
      "reward": 1.7000000849366188,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.7089286148548126,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13642
    },
    {
      "completion_length": 244.9687614440918,
      "epoch": 2.2876901798063622,
      "grad_norm": 0.18402523268310608,
      "kl": 0.249176025390625,
      "learning_rate": 4.934066452391482e-07,
      "loss": 0.0002,
      "reward": 1.7446429282426834,
      "reward_std": 0.017677669413387775,
      "rewards/equation_reward_func": 0.7491071783006191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13644
    },
    {
      "completion_length": 248.65179634094238,
      "epoch": 2.28802548304623,
      "grad_norm": 0.2082230382842461,
      "kl": 0.2586669921875,
      "learning_rate": 4.934039081571174e-07,
      "loss": 0.0003,
      "reward": 1.7500000521540642,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 13646
    },
    {
      "completion_length": 241.821439743042,
      "epoch": 2.2883607862860975,
      "grad_norm": 0.17611803208102136,
      "kl": 0.144500732421875,
      "learning_rate": 4.934011705146805e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964286096394062,
      "rewards/format_reward_func": 1.0,
      "step": 13648
    },
    {
      "completion_length": 227.36161708831787,
      "epoch": 2.288696089525965,
      "grad_norm": 0.14735190264686204,
      "kl": 0.245391845703125,
      "learning_rate": 4.93398432311844e-07,
      "loss": 0.0002,
      "reward": 1.7785714864730835,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7875000275671482,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13650
    },
    {
      "completion_length": 237.90179824829102,
      "epoch": 2.289031392765833,
      "grad_norm": 0.25570229818737517,
      "kl": 0.161041259765625,
      "learning_rate": 4.93395693548614e-07,
      "loss": 0.0002,
      "reward": 1.7625000551342964,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643200933933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13652
    },
    {
      "completion_length": 238.3928689956665,
      "epoch": 2.2893666960057,
      "grad_norm": 0.1867942311968773,
      "kl": 0.15655517578125,
      "learning_rate": 4.933929542249968e-07,
      "loss": 0.0002,
      "reward": 1.7428572103381157,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7517857439815998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13654
    },
    {
      "completion_length": 224.59376049041748,
      "epoch": 2.2897019992455676,
      "grad_norm": 0.21953006511119944,
      "kl": 0.207977294921875,
      "learning_rate": 4.933902143409988e-07,
      "loss": 0.0002,
      "reward": 1.7705357745289803,
      "reward_std": 0.06187184248119593,
      "rewards/equation_reward_func": 0.7767857573926449,
      "rewards/format_reward_func": 0.9937500059604645,
      "step": 13656
    },
    {
      "completion_length": 247.86161994934082,
      "epoch": 2.2900373024854352,
      "grad_norm": 0.3480181787400742,
      "kl": 0.210174560546875,
      "learning_rate": 4.933874738966264e-07,
      "loss": 0.0002,
      "reward": 1.7553572058677673,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214596509933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13658
    },
    {
      "completion_length": 233.80358123779297,
      "epoch": 2.290372605725303,
      "grad_norm": 0.10373686583389505,
      "kl": 0.1361083984375,
      "learning_rate": 4.933847328918857e-07,
      "loss": 0.0001,
      "reward": 1.7732143551111221,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7776786014437675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13660
    },
    {
      "completion_length": 243.1384048461914,
      "epoch": 2.2907079089651705,
      "grad_norm": 0.10525744507139827,
      "kl": 0.15374755859375,
      "learning_rate": 4.933819913267831e-07,
      "loss": 0.0002,
      "reward": 1.7750000730156898,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 13662
    },
    {
      "completion_length": 246.4732265472412,
      "epoch": 2.2910432122050377,
      "grad_norm": 0.1968361967529215,
      "kl": 0.165618896484375,
      "learning_rate": 4.933792492013249e-07,
      "loss": 0.0002,
      "reward": 1.7785714715719223,
      "reward_std": 0.07071067672222853,
      "rewards/equation_reward_func": 0.7875000163912773,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13664
    },
    {
      "completion_length": 235.36161708831787,
      "epoch": 2.2913785154449053,
      "grad_norm": 0.08932347581233609,
      "kl": 0.140716552734375,
      "learning_rate": 4.933765065155175e-07,
      "loss": 0.0001,
      "reward": 1.7839286103844643,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7883928865194321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13666
    },
    {
      "completion_length": 229.04018878936768,
      "epoch": 2.291713818684773,
      "grad_norm": 0.0748245735877684,
      "kl": 0.135772705078125,
      "learning_rate": 4.933737632693671e-07,
      "loss": 0.0001,
      "reward": 1.8392857313156128,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.839285746216774,
      "rewards/format_reward_func": 1.0,
      "step": 13668
    },
    {
      "completion_length": 231.80358123779297,
      "epoch": 2.2920491219246406,
      "grad_norm": 0.09018754149817187,
      "kl": 0.127655029296875,
      "learning_rate": 4.9337101946288e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 13670
    },
    {
      "completion_length": 236.7142972946167,
      "epoch": 2.2923844251645082,
      "grad_norm": 0.400827240902266,
      "kl": 0.14300537109375,
      "learning_rate": 4.933682750960627e-07,
      "loss": 0.0001,
      "reward": 1.7750000432133675,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000264495611,
      "rewards/format_reward_func": 1.0,
      "step": 13672
    },
    {
      "completion_length": 231.1741180419922,
      "epoch": 2.292719728404376,
      "grad_norm": 0.3008411517499098,
      "kl": 0.14361572265625,
      "learning_rate": 4.933655301689214e-07,
      "loss": 0.0001,
      "reward": 1.7446429133415222,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7580357510596514,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13674
    },
    {
      "completion_length": 239.4687623977661,
      "epoch": 2.293055031644243,
      "grad_norm": 0.24453466502782206,
      "kl": 0.164886474609375,
      "learning_rate": 4.933627846814623e-07,
      "loss": 0.0002,
      "reward": 1.669642947614193,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.674107177183032,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13676
    },
    {
      "completion_length": 240.4553689956665,
      "epoch": 2.2933903348841107,
      "grad_norm": 0.08003393432284185,
      "kl": 0.157135009765625,
      "learning_rate": 4.933600386336919e-07,
      "loss": 0.0002,
      "reward": 1.7500000670552254,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 13678
    },
    {
      "completion_length": 244.01340293884277,
      "epoch": 2.2937256381239783,
      "grad_norm": 0.22095536044182848,
      "kl": 0.130096435546875,
      "learning_rate": 4.933572920256165e-07,
      "loss": 0.0001,
      "reward": 1.7571429163217545,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7660714536905289,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13680
    },
    {
      "completion_length": 248.5803680419922,
      "epoch": 2.294060941363846,
      "grad_norm": 0.1975708507074759,
      "kl": 0.145477294921875,
      "learning_rate": 4.933545448572422e-07,
      "loss": 0.0001,
      "reward": 1.7325893640518188,
      "reward_std": 0.04482551896944642,
      "rewards/equation_reward_func": 0.7473214603960514,
      "rewards/format_reward_func": 0.9852678664028645,
      "step": 13682
    },
    {
      "completion_length": 235.977689743042,
      "epoch": 2.2943962446037136,
      "grad_norm": 0.0816102094427138,
      "kl": 0.14111328125,
      "learning_rate": 4.933517971285758e-07,
      "loss": 0.0001,
      "reward": 1.8071428909897804,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8071428760886192,
      "rewards/format_reward_func": 1.0,
      "step": 13684
    },
    {
      "completion_length": 227.9375114440918,
      "epoch": 2.294731547843581,
      "grad_norm": 0.3697824132658964,
      "kl": 0.158905029296875,
      "learning_rate": 4.933490488396232e-07,
      "loss": 0.0002,
      "reward": 1.8000000640749931,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8000000305473804,
      "rewards/format_reward_func": 1.0,
      "step": 13686
    },
    {
      "completion_length": 222.3794708251953,
      "epoch": 2.2950668510834484,
      "grad_norm": 0.3124872607707278,
      "kl": 0.131500244140625,
      "learning_rate": 4.933462999903908e-07,
      "loss": 0.0001,
      "reward": 1.7982143461704254,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.8026786036789417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13688
    },
    {
      "completion_length": 216.25447177886963,
      "epoch": 2.295402154323316,
      "grad_norm": 0.4198675461813113,
      "kl": 0.137725830078125,
      "learning_rate": 4.93343550580885e-07,
      "loss": 0.0001,
      "reward": 1.8089286386966705,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.8133928924798965,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13690
    },
    {
      "completion_length": 219.85715293884277,
      "epoch": 2.2957374575631837,
      "grad_norm": 0.09580375324629241,
      "kl": 0.1383056640625,
      "learning_rate": 4.933408006111121e-07,
      "loss": 0.0001,
      "reward": 1.7500000596046448,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 13692
    },
    {
      "completion_length": 222.63393878936768,
      "epoch": 2.2960727608030513,
      "grad_norm": 0.10388410768653196,
      "kl": 0.134613037109375,
      "learning_rate": 4.933380500810785e-07,
      "loss": 0.0001,
      "reward": 1.7553572058677673,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.759821455925703,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13694
    },
    {
      "completion_length": 211.35715293884277,
      "epoch": 2.296408064042919,
      "grad_norm": 0.13906267429864502,
      "kl": 0.145172119140625,
      "learning_rate": 4.933352989907906e-07,
      "loss": 0.0001,
      "reward": 1.7428572177886963,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7428571619093418,
      "rewards/format_reward_func": 1.0,
      "step": 13696
    },
    {
      "completion_length": 219.80358219146729,
      "epoch": 2.2967433672827866,
      "grad_norm": 0.13736673153418766,
      "kl": 0.1318359375,
      "learning_rate": 4.933325473402545e-07,
      "loss": 0.0001,
      "reward": 1.7660714909434319,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.769642885774374,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 13698
    },
    {
      "completion_length": 223.89733028411865,
      "epoch": 2.297078670522654,
      "grad_norm": 0.20006064325335238,
      "kl": 0.117767333984375,
      "learning_rate": 4.933297951294767e-07,
      "loss": 0.0001,
      "reward": 1.7392857745289803,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7392857596278191,
      "rewards/format_reward_func": 1.0,
      "step": 13700
    },
    {
      "completion_length": 218.90179634094238,
      "epoch": 2.2974139737625214,
      "grad_norm": 0.06124106879307074,
      "kl": 0.128936767578125,
      "learning_rate": 4.933270423584634e-07,
      "loss": 0.0001,
      "reward": 1.7821429371833801,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 13702
    },
    {
      "completion_length": 216.00000858306885,
      "epoch": 2.297749277002389,
      "grad_norm": 0.12457756050918946,
      "kl": 0.126312255859375,
      "learning_rate": 4.933242890272211e-07,
      "loss": 0.0001,
      "reward": 1.7982143238186836,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.8026785999536514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13704
    },
    {
      "completion_length": 215.63840103149414,
      "epoch": 2.2980845802422567,
      "grad_norm": 0.19463971761404908,
      "kl": 0.1267242431640625,
      "learning_rate": 4.933215351357561e-07,
      "loss": 0.0001,
      "reward": 1.7107143625617027,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7107143271714449,
      "rewards/format_reward_func": 1.0,
      "step": 13706
    },
    {
      "completion_length": 222.50001049041748,
      "epoch": 2.2984198834821243,
      "grad_norm": 0.1579762789545742,
      "kl": 0.125274658203125,
      "learning_rate": 4.933187806840747e-07,
      "loss": 0.0001,
      "reward": 1.821428619325161,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8214286044239998,
      "rewards/format_reward_func": 1.0,
      "step": 13708
    },
    {
      "completion_length": 230.0044755935669,
      "epoch": 2.2987551867219915,
      "grad_norm": 0.2105210773896155,
      "kl": 0.15521240234375,
      "learning_rate": 4.933160256721831e-07,
      "loss": 0.0002,
      "reward": 1.7339286357164383,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7383928932249546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13710
    },
    {
      "completion_length": 213.31251049041748,
      "epoch": 2.299090489961859,
      "grad_norm": 0.27241427140351204,
      "kl": 0.122161865234375,
      "learning_rate": 4.933132701000879e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 13712
    },
    {
      "completion_length": 219.40626049041748,
      "epoch": 2.299425793201727,
      "grad_norm": 0.259448491171959,
      "kl": 0.1417236328125,
      "learning_rate": 4.933105139677954e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.775000037625432,
      "rewards/format_reward_func": 1.0,
      "step": 13714
    },
    {
      "completion_length": 219.7366180419922,
      "epoch": 2.2997610964415944,
      "grad_norm": 0.25233047889996035,
      "kl": 0.137237548828125,
      "learning_rate": 4.933077572753118e-07,
      "loss": 0.0001,
      "reward": 1.7107143849134445,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7107143346220255,
      "rewards/format_reward_func": 1.0,
      "step": 13716
    },
    {
      "completion_length": 221.102689743042,
      "epoch": 2.300096399681462,
      "grad_norm": 0.3017274371156937,
      "kl": 0.140655517578125,
      "learning_rate": 4.933050000226437e-07,
      "loss": 0.0001,
      "reward": 1.7250000685453415,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7250000294297934,
      "rewards/format_reward_func": 1.0,
      "step": 13718
    },
    {
      "completion_length": 209.57143783569336,
      "epoch": 2.3004317029213297,
      "grad_norm": 0.11165261758553362,
      "kl": 0.136688232421875,
      "learning_rate": 4.93302242209797e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 13720
    },
    {
      "completion_length": 216.10269165039062,
      "epoch": 2.300767006161197,
      "grad_norm": 0.29489920734681696,
      "kl": 0.1304931640625,
      "learning_rate": 4.932994838367786e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 13722
    },
    {
      "completion_length": 220.83929538726807,
      "epoch": 2.3011023094010645,
      "grad_norm": 0.0707173188716113,
      "kl": 0.13385009765625,
      "learning_rate": 4.932967249035944e-07,
      "loss": 0.0001,
      "reward": 1.739285796880722,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7392857559025288,
      "rewards/format_reward_func": 1.0,
      "step": 13724
    },
    {
      "completion_length": 221.10268878936768,
      "epoch": 2.301437612640932,
      "grad_norm": 0.3062096065804645,
      "kl": 0.14453125,
      "learning_rate": 4.93293965410251e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 13726
    },
    {
      "completion_length": 226.51786613464355,
      "epoch": 2.3017729158808,
      "grad_norm": 0.23534533660186088,
      "kl": 0.1221923828125,
      "learning_rate": 4.932912053567546e-07,
      "loss": 0.0001,
      "reward": 1.807142898440361,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8071428909897804,
      "rewards/format_reward_func": 1.0,
      "step": 13728
    },
    {
      "completion_length": 213.85715293884277,
      "epoch": 2.3021082191206674,
      "grad_norm": 0.0034166384522816224,
      "kl": 0.1214599609375,
      "learning_rate": 4.932884447431118e-07,
      "loss": 0.0001,
      "reward": 1.8107143267989159,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8107143193483353,
      "rewards/format_reward_func": 1.0,
      "step": 13730
    },
    {
      "completion_length": 220.40179538726807,
      "epoch": 2.3024435223605346,
      "grad_norm": 0.15621227615878694,
      "kl": 0.13421630859375,
      "learning_rate": 4.932856835693287e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 13732
    },
    {
      "completion_length": 218.62501049041748,
      "epoch": 2.3027788256004023,
      "grad_norm": 0.15452761746025206,
      "kl": 0.120635986328125,
      "learning_rate": 4.932829218354117e-07,
      "loss": 0.0001,
      "reward": 1.8125000223517418,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8169643133878708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13734
    },
    {
      "completion_length": 225.28572368621826,
      "epoch": 2.30311412884027,
      "grad_norm": 0.2775790745195352,
      "kl": 0.13287353515625,
      "learning_rate": 4.932801595413673e-07,
      "loss": 0.0001,
      "reward": 1.775000050663948,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 13736
    },
    {
      "completion_length": 225.00893688201904,
      "epoch": 2.3034494320801375,
      "grad_norm": 0.10617741628463849,
      "kl": 0.125,
      "learning_rate": 4.932773966872017e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7428571842610836,
      "rewards/format_reward_func": 1.0,
      "step": 13738
    },
    {
      "completion_length": 216.41965198516846,
      "epoch": 2.303784735320005,
      "grad_norm": 0.13860951964324789,
      "kl": 0.115509033203125,
      "learning_rate": 4.932746332729214e-07,
      "loss": 0.0001,
      "reward": 1.7803571820259094,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.784821443259716,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13740
    },
    {
      "completion_length": 229.27679634094238,
      "epoch": 2.304120038559873,
      "grad_norm": 0.4581590275723739,
      "kl": 0.1278839111328125,
      "learning_rate": 4.932718692985327e-07,
      "loss": 0.0001,
      "reward": 1.7196429446339607,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7241071686148643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13742
    },
    {
      "completion_length": 224.69197368621826,
      "epoch": 2.30445534179974,
      "grad_norm": 0.2416406933707828,
      "kl": 0.1219482421875,
      "learning_rate": 4.93269104764042e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7607143260538578,
      "rewards/format_reward_func": 1.0,
      "step": 13744
    },
    {
      "completion_length": 225.04911708831787,
      "epoch": 2.3047906450396076,
      "grad_norm": 0.31154448894666115,
      "kl": 0.120697021484375,
      "learning_rate": 4.932663396694554e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714693367481,
      "rewards/format_reward_func": 1.0,
      "step": 13746
    },
    {
      "completion_length": 222.9419755935669,
      "epoch": 2.3051259482794753,
      "grad_norm": 0.14378458200593694,
      "kl": 0.121368408203125,
      "learning_rate": 4.932635740147797e-07,
      "loss": 0.0001,
      "reward": 1.8107143342494965,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 13748
    },
    {
      "completion_length": 226.48661613464355,
      "epoch": 2.305461251519343,
      "grad_norm": 0.2690045565333607,
      "kl": 0.12030029296875,
      "learning_rate": 4.93260807800021e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714507102966,
      "rewards/format_reward_func": 1.0,
      "step": 13750
    },
    {
      "completion_length": 228.49107933044434,
      "epoch": 2.3057965547592105,
      "grad_norm": 0.3038352147540726,
      "kl": 0.12451171875,
      "learning_rate": 4.932580410251857e-07,
      "loss": 0.0001,
      "reward": 1.7392857894301414,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 1.0,
      "step": 13752
    },
    {
      "completion_length": 216.6071538925171,
      "epoch": 2.3061318579990777,
      "grad_norm": 0.19908074429539108,
      "kl": 0.13238525390625,
      "learning_rate": 4.932552736902803e-07,
      "loss": 0.0001,
      "reward": 1.7535714879631996,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714525729418,
      "rewards/format_reward_func": 1.0,
      "step": 13754
    },
    {
      "completion_length": 220.8794755935669,
      "epoch": 2.3064671612389454,
      "grad_norm": 0.2938801381346632,
      "kl": 0.1182708740234375,
      "learning_rate": 4.932525057953109e-07,
      "loss": 0.0001,
      "reward": 1.814285770058632,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8142857439815998,
      "rewards/format_reward_func": 1.0,
      "step": 13756
    },
    {
      "completion_length": 224.69643783569336,
      "epoch": 2.306802464478813,
      "grad_norm": 0.08695741183974194,
      "kl": 0.1104888916015625,
      "learning_rate": 4.932497373402842e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 13758
    },
    {
      "completion_length": 223.71429538726807,
      "epoch": 2.3071377677186806,
      "grad_norm": 0.280091163735147,
      "kl": 0.1106109619140625,
      "learning_rate": 4.932469683252062e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 13760
    },
    {
      "completion_length": 226.7723331451416,
      "epoch": 2.3074730709585483,
      "grad_norm": 0.36408205627825097,
      "kl": 0.1078948974609375,
      "learning_rate": 4.932441987500837e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 13762
    },
    {
      "completion_length": 223.4241180419922,
      "epoch": 2.307808374198416,
      "grad_norm": 0.18035158937625873,
      "kl": 0.13739013671875,
      "learning_rate": 4.932414286149228e-07,
      "loss": 0.0001,
      "reward": 1.7482143566012383,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7526785992085934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13764
    },
    {
      "completion_length": 220.99554634094238,
      "epoch": 2.308143677438283,
      "grad_norm": 0.12655154468340585,
      "kl": 0.12335205078125,
      "learning_rate": 4.9323865791973e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 13766
    },
    {
      "completion_length": 221.41965293884277,
      "epoch": 2.3084789806781507,
      "grad_norm": 0.155988583871809,
      "kl": 0.122344970703125,
      "learning_rate": 4.932358866645116e-07,
      "loss": 0.0001,
      "reward": 1.8428571820259094,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8428571578115225,
      "rewards/format_reward_func": 1.0,
      "step": 13768
    },
    {
      "completion_length": 230.5044765472412,
      "epoch": 2.3088142839180184,
      "grad_norm": 0.21690628723905842,
      "kl": 0.1177978515625,
      "learning_rate": 4.932331148492739e-07,
      "loss": 0.0001,
      "reward": 1.8178571909666061,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8178571686148643,
      "rewards/format_reward_func": 1.0,
      "step": 13770
    },
    {
      "completion_length": 230.7500114440918,
      "epoch": 2.309149587157886,
      "grad_norm": 0.22875686319240196,
      "kl": 0.10528564453125,
      "learning_rate": 4.932303424740234e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714285887777805,
      "rewards/format_reward_func": 1.0,
      "step": 13772
    },
    {
      "completion_length": 240.36608219146729,
      "epoch": 2.3094848903977536,
      "grad_norm": 0.13109625511493334,
      "kl": 0.117034912109375,
      "learning_rate": 4.932275695387666e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7714285925030708,
      "rewards/format_reward_func": 1.0,
      "step": 13774
    },
    {
      "completion_length": 236.40179920196533,
      "epoch": 2.309820193637621,
      "grad_norm": 0.15888328027894855,
      "kl": 0.106597900390625,
      "learning_rate": 4.932247960435096e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 13776
    },
    {
      "completion_length": 234.8125114440918,
      "epoch": 2.3101554968774884,
      "grad_norm": 0.15421914534549414,
      "kl": 0.1151123046875,
      "learning_rate": 4.932220219882591e-07,
      "loss": 0.0001,
      "reward": 1.825000062584877,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8250000178813934,
      "rewards/format_reward_func": 1.0,
      "step": 13778
    },
    {
      "completion_length": 236.29911613464355,
      "epoch": 2.310490800117356,
      "grad_norm": 0.09890678034383357,
      "kl": 0.12774658203125,
      "learning_rate": 4.932192473730213e-07,
      "loss": 0.0001,
      "reward": 1.7910714447498322,
      "reward_std": 0.012626906856894493,
      "rewards/equation_reward_func": 0.7955357562750578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13780
    },
    {
      "completion_length": 238.1875123977661,
      "epoch": 2.3108261033572237,
      "grad_norm": 0.16175457520796585,
      "kl": 0.127197265625,
      "learning_rate": 4.932164721978024e-07,
      "loss": 0.0001,
      "reward": 1.7017858028411865,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.706250024959445,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13782
    },
    {
      "completion_length": 231.01340198516846,
      "epoch": 2.3111614065970914,
      "grad_norm": 0.19291834599943525,
      "kl": 0.1243896484375,
      "learning_rate": 4.932136964626093e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000432133675,
      "rewards/format_reward_func": 1.0,
      "step": 13784
    },
    {
      "completion_length": 247.75894165039062,
      "epoch": 2.311496709836959,
      "grad_norm": 0.1155973995569849,
      "kl": 0.126373291015625,
      "learning_rate": 4.932109201674478e-07,
      "loss": 0.0001,
      "reward": 1.7821428924798965,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 13786
    },
    {
      "completion_length": 242.05358219146729,
      "epoch": 2.311832013076826,
      "grad_norm": 0.2314433409991325,
      "kl": 0.156005859375,
      "learning_rate": 4.932081433123248e-07,
      "loss": 0.0002,
      "reward": 1.7875000834465027,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7919642943888903,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13788
    },
    {
      "completion_length": 236.2946548461914,
      "epoch": 2.312167316316694,
      "grad_norm": 0.0022804056792775137,
      "kl": 0.1068115234375,
      "learning_rate": 4.932053658972464e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 13790
    },
    {
      "completion_length": 229.5982255935669,
      "epoch": 2.3125026195565614,
      "grad_norm": 0.22285338579905276,
      "kl": 0.125946044921875,
      "learning_rate": 4.932025879222191e-07,
      "loss": 0.0001,
      "reward": 1.7392858117818832,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7392857372760773,
      "rewards/format_reward_func": 1.0,
      "step": 13792
    },
    {
      "completion_length": 221.48661708831787,
      "epoch": 2.312837922796429,
      "grad_norm": 0.32382712126000757,
      "kl": 0.119415283203125,
      "learning_rate": 4.931998093872492e-07,
      "loss": 0.0001,
      "reward": 1.8285714760422707,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8285714574158192,
      "rewards/format_reward_func": 1.0,
      "step": 13794
    },
    {
      "completion_length": 228.21875858306885,
      "epoch": 2.3131732260362967,
      "grad_norm": 0.19368082670345033,
      "kl": 0.11956787109375,
      "learning_rate": 4.931970302923433e-07,
      "loss": 0.0001,
      "reward": 1.7214286401867867,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7214285992085934,
      "rewards/format_reward_func": 1.0,
      "step": 13796
    },
    {
      "completion_length": 226.20090293884277,
      "epoch": 2.313508529276164,
      "grad_norm": 0.29231224698111935,
      "kl": 0.12823486328125,
      "learning_rate": 4.931942506375076e-07,
      "loss": 0.0001,
      "reward": 1.7750000730156898,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 13798
    },
    {
      "completion_length": 230.42411708831787,
      "epoch": 2.3138438325160315,
      "grad_norm": 0.18361270601123025,
      "kl": 0.19329833984375,
      "learning_rate": 4.931914704227486e-07,
      "loss": 0.0002,
      "reward": 1.7714286521077156,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 13800
    },
    {
      "completion_length": 228.7053680419922,
      "epoch": 2.314179135755899,
      "grad_norm": 0.20926399834142734,
      "kl": 0.152862548828125,
      "learning_rate": 4.931886896480726e-07,
      "loss": 0.0002,
      "reward": 1.7321429401636124,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428954601288,
      "rewards/format_reward_func": 1.0,
      "step": 13802
    },
    {
      "completion_length": 225.32143783569336,
      "epoch": 2.314514438995767,
      "grad_norm": 0.24781507311502457,
      "kl": 0.138458251953125,
      "learning_rate": 4.931859083134861e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 13804
    },
    {
      "completion_length": 229.91518783569336,
      "epoch": 2.3148497422356344,
      "grad_norm": 0.13919086617690282,
      "kl": 0.14959716796875,
      "learning_rate": 4.931831264189954e-07,
      "loss": 0.0001,
      "reward": 1.7785714715719223,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714566707611,
      "rewards/format_reward_func": 1.0,
      "step": 13806
    },
    {
      "completion_length": 223.39286994934082,
      "epoch": 2.315185045475502,
      "grad_norm": 0.2862038229346431,
      "kl": 0.141693115234375,
      "learning_rate": 4.931803439646071e-07,
      "loss": 0.0001,
      "reward": 1.7321429252624512,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321429029107094,
      "rewards/format_reward_func": 1.0,
      "step": 13808
    },
    {
      "completion_length": 229.4330472946167,
      "epoch": 2.3155203487153693,
      "grad_norm": 0.536633798934367,
      "kl": 0.44781494140625,
      "learning_rate": 4.931775609503274e-07,
      "loss": 0.0004,
      "reward": 1.7642857804894447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 13810
    },
    {
      "completion_length": 224.14286708831787,
      "epoch": 2.315855651955237,
      "grad_norm": 0.17537454963307,
      "kl": 0.14599609375,
      "learning_rate": 4.931747773761628e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 13812
    },
    {
      "completion_length": 230.0982255935669,
      "epoch": 2.3161909551951045,
      "grad_norm": 0.21195475663310653,
      "kl": 0.161041259765625,
      "learning_rate": 4.931719932421198e-07,
      "loss": 0.0002,
      "reward": 1.7678572088479996,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571827709675,
      "rewards/format_reward_func": 1.0,
      "step": 13814
    },
    {
      "completion_length": 234.0491189956665,
      "epoch": 2.316526258434972,
      "grad_norm": 0.3517252391572919,
      "kl": 0.397308349609375,
      "learning_rate": 4.931692085482045e-07,
      "loss": 0.0004,
      "reward": 1.7522322237491608,
      "reward_std": 0.04735089954920113,
      "rewards/equation_reward_func": 0.753571480512619,
      "rewards/format_reward_func": 0.9986607171595097,
      "step": 13816
    },
    {
      "completion_length": 218.95982933044434,
      "epoch": 2.31686156167484,
      "grad_norm": 0.21637620176464864,
      "kl": 0.149993896484375,
      "learning_rate": 4.931664232944237e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 13818
    },
    {
      "completion_length": 233.7946538925171,
      "epoch": 2.317196864914707,
      "grad_norm": 0.1493142580467565,
      "kl": 0.184417724609375,
      "learning_rate": 4.931636374807836e-07,
      "loss": 0.0002,
      "reward": 1.757142923772335,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428958326578,
      "rewards/format_reward_func": 1.0,
      "step": 13820
    },
    {
      "completion_length": 238.07590293884277,
      "epoch": 2.3175321681545746,
      "grad_norm": 0.23369268131692472,
      "kl": 0.210357666015625,
      "learning_rate": 4.931608511072906e-07,
      "loss": 0.0002,
      "reward": 1.7464286461472511,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 13822
    },
    {
      "completion_length": 222.6517972946167,
      "epoch": 2.3178674713944423,
      "grad_norm": 0.18778130385122507,
      "kl": 0.2147216796875,
      "learning_rate": 4.931580641739513e-07,
      "loss": 0.0002,
      "reward": 1.8214286342263222,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8214285932481289,
      "rewards/format_reward_func": 1.0,
      "step": 13824
    },
    {
      "completion_length": 230.2321548461914,
      "epoch": 2.31820277463431,
      "grad_norm": 0.09521929881811564,
      "kl": 0.161590576171875,
      "learning_rate": 4.931552766807718e-07,
      "loss": 0.0002,
      "reward": 1.7892857417464256,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 13826
    },
    {
      "completion_length": 234.70536708831787,
      "epoch": 2.3185380778741775,
      "grad_norm": 0.31405589164753217,
      "kl": 0.1912841796875,
      "learning_rate": 4.931524886277587e-07,
      "loss": 0.0002,
      "reward": 1.7267858237028122,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7312500197440386,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13828
    },
    {
      "completion_length": 242.9107255935669,
      "epoch": 2.318873381114045,
      "grad_norm": 0.14866603002861797,
      "kl": 0.142608642578125,
      "learning_rate": 4.931497000149185e-07,
      "loss": 0.0001,
      "reward": 1.7401786372065544,
      "reward_std": 0.03409264795482159,
      "rewards/equation_reward_func": 0.7419643048197031,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 13830
    },
    {
      "completion_length": 234.8973331451416,
      "epoch": 2.319208684353913,
      "grad_norm": 0.25162094245415106,
      "kl": 0.150634765625,
      "learning_rate": 4.931469108422576e-07,
      "loss": 0.0002,
      "reward": 1.7482143566012383,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7526785992085934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13832
    },
    {
      "completion_length": 246.16519355773926,
      "epoch": 2.31954398759378,
      "grad_norm": 0.11166127332516824,
      "kl": 0.158416748046875,
      "learning_rate": 4.931441211097822e-07,
      "loss": 0.0002,
      "reward": 1.7937500327825546,
      "reward_std": 0.029041884932667017,
      "rewards/equation_reward_func": 0.7955357301980257,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 13834
    },
    {
      "completion_length": 246.6875114440918,
      "epoch": 2.3198792908336476,
      "grad_norm": 0.21306896478258996,
      "kl": 0.1680908203125,
      "learning_rate": 4.93141330817499e-07,
      "loss": 0.0002,
      "reward": 1.725000075995922,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000312924385,
      "rewards/format_reward_func": 1.0,
      "step": 13836
    },
    {
      "completion_length": 245.06251335144043,
      "epoch": 2.3202145940735153,
      "grad_norm": 0.12055812100393454,
      "kl": 0.15869140625,
      "learning_rate": 4.931385399654143e-07,
      "loss": 0.0002,
      "reward": 1.751785784959793,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7562500238418579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13838
    },
    {
      "completion_length": 242.83929634094238,
      "epoch": 2.320549897313383,
      "grad_norm": 0.17749519435879707,
      "kl": 0.128753662109375,
      "learning_rate": 4.931357485535345e-07,
      "loss": 0.0001,
      "reward": 1.803571455180645,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8035714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 13840
    },
    {
      "completion_length": 247.52233409881592,
      "epoch": 2.3208852005532505,
      "grad_norm": 0.05211308765186071,
      "kl": 0.16229248046875,
      "learning_rate": 4.93132956581866e-07,
      "loss": 0.0002,
      "reward": 1.775000050663948,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 13842
    },
    {
      "completion_length": 242.63840198516846,
      "epoch": 2.3212205037931177,
      "grad_norm": 0.22663570203284325,
      "kl": 0.146820068359375,
      "learning_rate": 4.931301640504154e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 13844
    },
    {
      "completion_length": 253.0312614440918,
      "epoch": 2.3215558070329854,
      "grad_norm": 0.16115678842341616,
      "kl": 0.141632080078125,
      "learning_rate": 4.93127370959189e-07,
      "loss": 0.0001,
      "reward": 1.7714286595582962,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714285925030708,
      "rewards/format_reward_func": 1.0,
      "step": 13846
    },
    {
      "completion_length": 257.8259000778198,
      "epoch": 2.321891110272853,
      "grad_norm": 0.23600484644051928,
      "kl": 0.141021728515625,
      "learning_rate": 4.931245773081931e-07,
      "loss": 0.0001,
      "reward": 1.7285715192556381,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.728571455925703,
      "rewards/format_reward_func": 1.0,
      "step": 13848
    },
    {
      "completion_length": 240.5000123977661,
      "epoch": 2.3222264135127206,
      "grad_norm": 0.14853695490593902,
      "kl": 0.16864013671875,
      "learning_rate": 4.931217830974344e-07,
      "loss": 0.0002,
      "reward": 1.7785714715719223,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 13850
    },
    {
      "completion_length": 253.27679634094238,
      "epoch": 2.3225617167525883,
      "grad_norm": 0.442639076297675,
      "kl": 0.14019775390625,
      "learning_rate": 4.931189883269193e-07,
      "loss": 0.0001,
      "reward": 1.7428572177886963,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7607143074274063,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 13852
    },
    {
      "completion_length": 256.1741199493408,
      "epoch": 2.322897019992456,
      "grad_norm": 0.2172474753408119,
      "kl": 0.162078857421875,
      "learning_rate": 4.931161929966541e-07,
      "loss": 0.0002,
      "reward": 1.821428619325161,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.821428595110774,
      "rewards/format_reward_func": 1.0,
      "step": 13854
    },
    {
      "completion_length": 272.4464416503906,
      "epoch": 2.323232323232323,
      "grad_norm": 0.25720741051210894,
      "kl": 0.141754150390625,
      "learning_rate": 4.931133971066451e-07,
      "loss": 0.0001,
      "reward": 1.7410714849829674,
      "reward_std": 0.09343910869210958,
      "rewards/equation_reward_func": 0.7544643096625805,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13856
    },
    {
      "completion_length": 252.602689743042,
      "epoch": 2.3235676264721907,
      "grad_norm": 0.19960302259651616,
      "kl": 0.136322021484375,
      "learning_rate": 4.931106006568991e-07,
      "loss": 0.0001,
      "reward": 1.7839286252856255,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7883928939700127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13858
    },
    {
      "completion_length": 257.57143783569336,
      "epoch": 2.3239029297120584,
      "grad_norm": 0.22315189200039845,
      "kl": 0.1318359375,
      "learning_rate": 4.931078036474224e-07,
      "loss": 0.0001,
      "reward": 1.7982143461704254,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8026785999536514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13860
    },
    {
      "completion_length": 268.7410840988159,
      "epoch": 2.324238232951926,
      "grad_norm": 0.6520892118623528,
      "kl": 0.4154052734375,
      "learning_rate": 4.931050060782212e-07,
      "loss": 0.0004,
      "reward": 1.7732143476605415,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.777678593993187,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13862
    },
    {
      "completion_length": 263.2634038925171,
      "epoch": 2.3245735361917936,
      "grad_norm": 0.29998155060416803,
      "kl": 0.1378173828125,
      "learning_rate": 4.931022079493024e-07,
      "loss": 0.0001,
      "reward": 1.7285714969038963,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7285714522004128,
      "rewards/format_reward_func": 1.0,
      "step": 13864
    },
    {
      "completion_length": 277.0759038925171,
      "epoch": 2.324908839431661,
      "grad_norm": 0.09117265999784231,
      "kl": 0.144500732421875,
      "learning_rate": 4.930994092606719e-07,
      "loss": 0.0001,
      "reward": 1.7946429029107094,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7991071604192257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13866
    },
    {
      "completion_length": 279.62947368621826,
      "epoch": 2.3252441426715285,
      "grad_norm": 0.17952208593591043,
      "kl": 0.141357421875,
      "learning_rate": 4.930966100123366e-07,
      "loss": 0.0001,
      "reward": 1.8535714820027351,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8535714484751225,
      "rewards/format_reward_func": 1.0,
      "step": 13868
    },
    {
      "completion_length": 270.15179538726807,
      "epoch": 2.325579445911396,
      "grad_norm": 0.1446632522156985,
      "kl": 0.142425537109375,
      "learning_rate": 4.930938102043027e-07,
      "loss": 0.0001,
      "reward": 1.8446428999304771,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.8491071723401546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13870
    },
    {
      "completion_length": 287.2901906967163,
      "epoch": 2.3259147491512637,
      "grad_norm": 0.09508477728521918,
      "kl": 0.1434326171875,
      "learning_rate": 4.930910098365768e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7767857275903225,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13872
    },
    {
      "completion_length": 287.87947845458984,
      "epoch": 2.3262500523911314,
      "grad_norm": 0.15208370071022806,
      "kl": 0.157379150390625,
      "learning_rate": 4.930882089091652e-07,
      "loss": 0.0002,
      "reward": 1.796428643167019,
      "reward_std": 0.04545686487108469,
      "rewards/equation_reward_func": 0.8053571730852127,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13874
    },
    {
      "completion_length": 267.7678689956665,
      "epoch": 2.326585355630999,
      "grad_norm": 0.18081008050190195,
      "kl": 0.143707275390625,
      "learning_rate": 4.930854074220744e-07,
      "loss": 0.0001,
      "reward": 1.8000000342726707,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 13876
    },
    {
      "completion_length": 276.84376335144043,
      "epoch": 2.326920658870866,
      "grad_norm": 0.048354361422872846,
      "kl": 0.125518798828125,
      "learning_rate": 4.930826053753109e-07,
      "loss": 0.0001,
      "reward": 1.7732143253087997,
      "reward_std": 0.007576144300401211,
      "rewards/equation_reward_func": 0.7776786051690578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13878
    },
    {
      "completion_length": 278.4732303619385,
      "epoch": 2.327255962110734,
      "grad_norm": 0.1011312406782036,
      "kl": 0.16424560546875,
      "learning_rate": 4.930798027688811e-07,
      "loss": 0.0002,
      "reward": 1.7928571924567223,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 13880
    },
    {
      "completion_length": 274.27233123779297,
      "epoch": 2.3275912653506015,
      "grad_norm": 0.11005274619203892,
      "kl": 0.13623046875,
      "learning_rate": 4.930769996027915e-07,
      "loss": 0.0001,
      "reward": 1.7732143476605415,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.777678593993187,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13882
    },
    {
      "completion_length": 269.85268783569336,
      "epoch": 2.327926568590469,
      "grad_norm": 0.08785502020528425,
      "kl": 0.13714599609375,
      "learning_rate": 4.930741958770485e-07,
      "loss": 0.0001,
      "reward": 1.769642911851406,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7741071842610836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13884
    },
    {
      "completion_length": 273.22322940826416,
      "epoch": 2.3282618718303367,
      "grad_norm": 0.04808767039662426,
      "kl": 0.149322509765625,
      "learning_rate": 4.930713915916585e-07,
      "loss": 0.0001,
      "reward": 1.7767857760190964,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7812500223517418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13886
    },
    {
      "completion_length": 267.8437614440918,
      "epoch": 2.328597175070204,
      "grad_norm": 0.0033872799839121735,
      "kl": 0.1512451171875,
      "learning_rate": 4.930685867466281e-07,
      "loss": 0.0002,
      "reward": 1.8214285969734192,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.8214286006987095,
      "rewards/format_reward_func": 1.0,
      "step": 13888
    },
    {
      "completion_length": 277.64287185668945,
      "epoch": 2.3289324783100716,
      "grad_norm": 0.17916554169069204,
      "kl": 0.153594970703125,
      "learning_rate": 4.930657813419638e-07,
      "loss": 0.0002,
      "reward": 1.7303571850061417,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7437500283122063,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13890
    },
    {
      "completion_length": 281.90626525878906,
      "epoch": 2.329267781549939,
      "grad_norm": 0.3585679228241435,
      "kl": 0.16705322265625,
      "learning_rate": 4.930629753776717e-07,
      "loss": 0.0002,
      "reward": 1.7464286386966705,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7553571686148643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13892
    },
    {
      "completion_length": 265.11162185668945,
      "epoch": 2.329603084789807,
      "grad_norm": 0.18914141170325072,
      "kl": 0.13250732421875,
      "learning_rate": 4.930601688537586e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.753571443259716,
      "rewards/format_reward_func": 1.0,
      "step": 13894
    },
    {
      "completion_length": 262.7500104904175,
      "epoch": 2.3299383880296745,
      "grad_norm": 0.03789824323268437,
      "kl": 0.121246337890625,
      "learning_rate": 4.930573617702311e-07,
      "loss": 0.0001,
      "reward": 1.7839286178350449,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13896
    },
    {
      "completion_length": 262.20537185668945,
      "epoch": 2.330273691269542,
      "grad_norm": 0.2800962462007646,
      "kl": 0.1329345703125,
      "learning_rate": 4.930545541270952e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571753203869,
      "rewards/format_reward_func": 1.0,
      "step": 13898
    },
    {
      "completion_length": 267.5892963409424,
      "epoch": 2.3306089945094093,
      "grad_norm": 0.31285336060043917,
      "kl": 0.133331298828125,
      "learning_rate": 4.930517459243576e-07,
      "loss": 0.0001,
      "reward": 1.7803572043776512,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7848214544355869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13900
    },
    {
      "completion_length": 262.29465770721436,
      "epoch": 2.330944297749277,
      "grad_norm": 0.2676019483729929,
      "kl": 0.142913818359375,
      "learning_rate": 4.930489371620247e-07,
      "loss": 0.0001,
      "reward": 1.7696429342031479,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7741071656346321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13902
    },
    {
      "completion_length": 249.6071538925171,
      "epoch": 2.3312796009891446,
      "grad_norm": 0.2603757129759202,
      "kl": 0.135467529296875,
      "learning_rate": 4.930461278401033e-07,
      "loss": 0.0001,
      "reward": 1.8142857551574707,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857402563095,
      "rewards/format_reward_func": 1.0,
      "step": 13904
    },
    {
      "completion_length": 255.84822463989258,
      "epoch": 2.331614904229012,
      "grad_norm": 0.17282725467474225,
      "kl": 0.143646240234375,
      "learning_rate": 4.930433179585995e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 13906
    },
    {
      "completion_length": 260.2187614440918,
      "epoch": 2.33195020746888,
      "grad_norm": 0.08766943170738209,
      "kl": 0.1429443359375,
      "learning_rate": 4.930405075175197e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 13908
    },
    {
      "completion_length": 256.4375114440918,
      "epoch": 2.332285510708747,
      "grad_norm": 0.20621206310280166,
      "kl": 0.1400146484375,
      "learning_rate": 4.930376965168707e-07,
      "loss": 0.0001,
      "reward": 1.8214286267757416,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8214285969734192,
      "rewards/format_reward_func": 1.0,
      "step": 13910
    },
    {
      "completion_length": 263.1875114440918,
      "epoch": 2.3326208139486146,
      "grad_norm": 0.20267854080407852,
      "kl": 0.13336181640625,
      "learning_rate": 4.930348849566587e-07,
      "loss": 0.0001,
      "reward": 1.7250000834465027,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000350177288,
      "rewards/format_reward_func": 1.0,
      "step": 13912
    },
    {
      "completion_length": 247.66072750091553,
      "epoch": 2.3329561171884823,
      "grad_norm": 0.149466506521507,
      "kl": 0.114837646484375,
      "learning_rate": 4.930320728368904e-07,
      "loss": 0.0001,
      "reward": 1.730357214808464,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7348214611411095,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13914
    },
    {
      "completion_length": 261.1339406967163,
      "epoch": 2.33329142042835,
      "grad_norm": 0.21434438156759789,
      "kl": 0.137481689453125,
      "learning_rate": 4.930292601575721e-07,
      "loss": 0.0001,
      "reward": 1.8035714849829674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714328289032,
      "rewards/format_reward_func": 1.0,
      "step": 13916
    },
    {
      "completion_length": 250.38393783569336,
      "epoch": 2.3336267236682176,
      "grad_norm": 0.16715115682608317,
      "kl": 0.128936767578125,
      "learning_rate": 4.930264469187103e-07,
      "loss": 0.0001,
      "reward": 1.764285758137703,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7642857432365417,
      "rewards/format_reward_func": 1.0,
      "step": 13918
    },
    {
      "completion_length": 255.82144165039062,
      "epoch": 2.333962026908085,
      "grad_norm": 0.12919460625180537,
      "kl": 0.16015625,
      "learning_rate": 4.930236331203116e-07,
      "loss": 0.0002,
      "reward": 1.789285808801651,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857380211353,
      "rewards/format_reward_func": 1.0,
      "step": 13920
    },
    {
      "completion_length": 254.96429920196533,
      "epoch": 2.3342973301479524,
      "grad_norm": 0.19431944763476283,
      "kl": 0.1590576171875,
      "learning_rate": 4.930208187623823e-07,
      "loss": 0.0002,
      "reward": 1.7892857640981674,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857529222965,
      "rewards/format_reward_func": 1.0,
      "step": 13922
    },
    {
      "completion_length": 252.14733409881592,
      "epoch": 2.33463263338782,
      "grad_norm": 0.5509848451277506,
      "kl": 0.17974853515625,
      "learning_rate": 4.93018003844929e-07,
      "loss": 0.0002,
      "reward": 1.758928619325161,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7633928880095482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13924
    },
    {
      "completion_length": 248.67858123779297,
      "epoch": 2.3349679366276876,
      "grad_norm": 0.06223604978287845,
      "kl": 0.12445068359375,
      "learning_rate": 4.930151883679582e-07,
      "loss": 0.0001,
      "reward": 1.7321429178118706,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7321428954601288,
      "rewards/format_reward_func": 1.0,
      "step": 13926
    },
    {
      "completion_length": 251.88840293884277,
      "epoch": 2.3353032398675553,
      "grad_norm": 0.2945459558718312,
      "kl": 0.14276123046875,
      "learning_rate": 4.930123723314763e-07,
      "loss": 0.0001,
      "reward": 1.7696429193019867,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7830357365310192,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13928
    },
    {
      "completion_length": 246.0223331451416,
      "epoch": 2.335638543107423,
      "grad_norm": 0.20297409443292536,
      "kl": 0.1510009765625,
      "learning_rate": 4.930095557354897e-07,
      "loss": 0.0002,
      "reward": 1.7821429446339607,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428682655096,
      "rewards/format_reward_func": 1.0,
      "step": 13930
    },
    {
      "completion_length": 252.99108505249023,
      "epoch": 2.33597384634729,
      "grad_norm": 0.17278858736936387,
      "kl": 0.183074951171875,
      "learning_rate": 4.930067385800051e-07,
      "loss": 0.0002,
      "reward": 1.785714365541935,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 13932
    },
    {
      "completion_length": 261.7634048461914,
      "epoch": 2.3363091495871577,
      "grad_norm": 0.3142320730365933,
      "kl": 0.22943115234375,
      "learning_rate": 4.93003920865029e-07,
      "loss": 0.0002,
      "reward": 1.6785715073347092,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.6785714589059353,
      "rewards/format_reward_func": 1.0,
      "step": 13934
    },
    {
      "completion_length": 263.3169755935669,
      "epoch": 2.3366444528270254,
      "grad_norm": 0.16214082641153868,
      "kl": 0.15496826171875,
      "learning_rate": 4.930011025905677e-07,
      "loss": 0.0002,
      "reward": 1.7714286223053932,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7803571727126837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13936
    },
    {
      "completion_length": 264.99554920196533,
      "epoch": 2.336979756066893,
      "grad_norm": 0.29927774244425503,
      "kl": 0.158111572265625,
      "learning_rate": 4.929982837566277e-07,
      "loss": 0.0002,
      "reward": 1.7517857626080513,
      "reward_std": 0.08838834520429373,
      "rewards/equation_reward_func": 0.7651786096394062,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13938
    },
    {
      "completion_length": 270.9241199493408,
      "epoch": 2.3373150593067606,
      "grad_norm": 0.13127494720037441,
      "kl": 0.209259033203125,
      "learning_rate": 4.929954643632156e-07,
      "loss": 0.0002,
      "reward": 1.7696429342031479,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7741071805357933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13940
    },
    {
      "completion_length": 260.0669765472412,
      "epoch": 2.3376503625466283,
      "grad_norm": 0.1928191917215741,
      "kl": 0.188751220703125,
      "learning_rate": 4.929926444103378e-07,
      "loss": 0.0002,
      "reward": 1.7767857685685158,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7812500223517418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13942
    },
    {
      "completion_length": 259.8750123977661,
      "epoch": 2.3379856657864955,
      "grad_norm": 0.07497032977190916,
      "kl": 0.191162109375,
      "learning_rate": 4.92989823898001e-07,
      "loss": 0.0002,
      "reward": 1.6946429312229156,
      "reward_std": 0.017677669413387775,
      "rewards/equation_reward_func": 0.699107188731432,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13944
    },
    {
      "completion_length": 255.8348331451416,
      "epoch": 2.338320969026363,
      "grad_norm": 0.1732593334616933,
      "kl": 0.129974365234375,
      "learning_rate": 4.929870028262113e-07,
      "loss": 0.0001,
      "reward": 1.7910714596509933,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.795535746961832,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13946
    },
    {
      "completion_length": 258.7187614440918,
      "epoch": 2.3386562722662307,
      "grad_norm": 0.17122023310010204,
      "kl": 33.64996337890625,
      "learning_rate": 4.929841811949755e-07,
      "loss": 0.0336,
      "reward": 1.7410714849829674,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7455357424914837,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13948
    },
    {
      "completion_length": 257.1428689956665,
      "epoch": 2.3389915755060984,
      "grad_norm": 0.15583217546042194,
      "kl": 0.14776611328125,
      "learning_rate": 4.929813590043001e-07,
      "loss": 0.0001,
      "reward": 1.6910715028643608,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.695535758510232,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13950
    },
    {
      "completion_length": 248.93750858306885,
      "epoch": 2.339326878745966,
      "grad_norm": 0.20410850654255108,
      "kl": 0.15264892578125,
      "learning_rate": 4.929785362541915e-07,
      "loss": 0.0002,
      "reward": 1.7214286401867867,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7303571775555611,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13952
    },
    {
      "completion_length": 246.821439743042,
      "epoch": 2.339662181985833,
      "grad_norm": 0.19591857120170567,
      "kl": 0.222259521484375,
      "learning_rate": 4.929757129446562e-07,
      "loss": 0.0002,
      "reward": 1.739285796880722,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7482143118977547,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13954
    },
    {
      "completion_length": 242.6428689956665,
      "epoch": 2.339997485225701,
      "grad_norm": 0.1094692940084371,
      "kl": 0.172698974609375,
      "learning_rate": 4.929728890757007e-07,
      "loss": 0.0002,
      "reward": 1.8017857670783997,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.806250024586916,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13956
    },
    {
      "completion_length": 267.64732933044434,
      "epoch": 2.3403327884655685,
      "grad_norm": 0.2566541790875162,
      "kl": 0.1883544921875,
      "learning_rate": 4.929700646473315e-07,
      "loss": 0.0002,
      "reward": 1.7571429312229156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 13958
    },
    {
      "completion_length": 254.52233123779297,
      "epoch": 2.340668091705436,
      "grad_norm": 0.17341243088634806,
      "kl": 0.224151611328125,
      "learning_rate": 4.929672396595553e-07,
      "loss": 0.0002,
      "reward": 1.7625000551342964,
      "reward_std": 0.07323605939745903,
      "rewards/equation_reward_func": 0.7848214488476515,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 13960
    },
    {
      "completion_length": 253.27233028411865,
      "epoch": 2.3410033949453037,
      "grad_norm": 0.22604397912429652,
      "kl": 0.131744384765625,
      "learning_rate": 4.929644141123783e-07,
      "loss": 0.0001,
      "reward": 1.7732143327593803,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7776786014437675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13962
    },
    {
      "completion_length": 251.7678680419922,
      "epoch": 2.3413386981851714,
      "grad_norm": 0.6640328142423192,
      "kl": 0.463836669921875,
      "learning_rate": 4.92961588005807e-07,
      "loss": 0.0005,
      "reward": 1.7267857789993286,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.740178607404232,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 13964
    },
    {
      "completion_length": 248.24108600616455,
      "epoch": 2.341674001425039,
      "grad_norm": 0.10695830413721986,
      "kl": 0.135406494140625,
      "learning_rate": 4.929587613398481e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 13966
    },
    {
      "completion_length": 243.7053689956665,
      "epoch": 2.342009304664906,
      "grad_norm": 0.24921575355770892,
      "kl": 0.1546630859375,
      "learning_rate": 4.929559341145082e-07,
      "loss": 0.0002,
      "reward": 1.7767857685685158,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7812500409781933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13968
    },
    {
      "completion_length": 245.3750114440918,
      "epoch": 2.342344607904774,
      "grad_norm": 0.1565350284699649,
      "kl": 0.681243896484375,
      "learning_rate": 4.929531063297936e-07,
      "loss": 0.0007,
      "reward": 1.800000049173832,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 13970
    },
    {
      "completion_length": 242.3437614440918,
      "epoch": 2.3426799111446415,
      "grad_norm": 0.15593943795376905,
      "kl": 0.13665771484375,
      "learning_rate": 4.929502779857108e-07,
      "loss": 0.0001,
      "reward": 1.7160715088248253,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7205357365310192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13972
    },
    {
      "completion_length": 246.63840293884277,
      "epoch": 2.343015214384509,
      "grad_norm": 0.13590393267799425,
      "kl": 0.175994873046875,
      "learning_rate": 4.929474490822665e-07,
      "loss": 0.0002,
      "reward": 1.825000062584877,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8250000290572643,
      "rewards/format_reward_func": 1.0,
      "step": 13974
    },
    {
      "completion_length": 234.18751049041748,
      "epoch": 2.3433505176243767,
      "grad_norm": 0.4499483960615894,
      "kl": 0.209136962890625,
      "learning_rate": 4.92944619619467e-07,
      "loss": 0.0002,
      "reward": 1.7910714894533157,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7955357357859612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13976
    },
    {
      "completion_length": 241.38394165039062,
      "epoch": 2.343685820864244,
      "grad_norm": 0.25160773088122984,
      "kl": 0.16693115234375,
      "learning_rate": 4.929417895973189e-07,
      "loss": 0.0002,
      "reward": 1.7678572237491608,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 13978
    },
    {
      "completion_length": 242.8392972946167,
      "epoch": 2.3440211241041116,
      "grad_norm": 0.19259065367274678,
      "kl": 0.203277587890625,
      "learning_rate": 4.929389590158287e-07,
      "loss": 0.0002,
      "reward": 1.719642922282219,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.724107176065445,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13980
    },
    {
      "completion_length": 234.7187614440918,
      "epoch": 2.344356427343979,
      "grad_norm": 0.16228758744483066,
      "kl": 0.11175537109375,
      "learning_rate": 4.929361278750031e-07,
      "loss": 0.0001,
      "reward": 1.8107143193483353,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143230736256,
      "rewards/format_reward_func": 1.0,
      "step": 13982
    },
    {
      "completion_length": 237.6205472946167,
      "epoch": 2.344691730583847,
      "grad_norm": 0.1776062458911018,
      "kl": 0.115142822265625,
      "learning_rate": 4.929332961748482e-07,
      "loss": 0.0001,
      "reward": 1.7910714969038963,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7955357395112514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13984
    },
    {
      "completion_length": 249.27679538726807,
      "epoch": 2.3450270338237145,
      "grad_norm": 0.26582332087857086,
      "kl": 0.185760498046875,
      "learning_rate": 4.92930463915371e-07,
      "loss": 0.0002,
      "reward": 1.7571429461240768,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7660714536905289,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13986
    },
    {
      "completion_length": 237.51786708831787,
      "epoch": 2.345362337063582,
      "grad_norm": 0.26260524688968806,
      "kl": 0.1392822265625,
      "learning_rate": 4.929276310965778e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.06060915160924196,
      "rewards/equation_reward_func": 0.7946428805589676,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 13988
    },
    {
      "completion_length": 242.7634048461914,
      "epoch": 2.3456976403034493,
      "grad_norm": 0.5846938540785703,
      "kl": 0.22698974609375,
      "learning_rate": 4.92924797718475e-07,
      "loss": 0.0002,
      "reward": 1.8017857745289803,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8062500208616257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13990
    },
    {
      "completion_length": 247.4241180419922,
      "epoch": 2.346032943543317,
      "grad_norm": 0.232436541671342,
      "kl": 0.162109375,
      "learning_rate": 4.929219637810693e-07,
      "loss": 0.0002,
      "reward": 1.848214328289032,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8526785969734192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13992
    },
    {
      "completion_length": 249.7321548461914,
      "epoch": 2.3463682467831846,
      "grad_norm": 0.29302366800324925,
      "kl": 0.1395263671875,
      "learning_rate": 4.929191292843672e-07,
      "loss": 0.0001,
      "reward": 1.7303572297096252,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7348214462399483,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13994
    },
    {
      "completion_length": 246.446439743042,
      "epoch": 2.346703550023052,
      "grad_norm": 0.18002907127930834,
      "kl": 0.11785888671875,
      "learning_rate": 4.929162942283751e-07,
      "loss": 0.0001,
      "reward": 1.7839286252856255,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7883928865194321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13996
    },
    {
      "completion_length": 248.5357255935669,
      "epoch": 2.34703885326292,
      "grad_norm": 0.30579652194045426,
      "kl": 0.1209716796875,
      "learning_rate": 4.929134586130998e-07,
      "loss": 0.0001,
      "reward": 1.758928656578064,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7633928991854191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 13998
    },
    {
      "completion_length": 249.0446538925171,
      "epoch": 2.347374156502787,
      "grad_norm": 0.2357568781576199,
      "kl": 0.103302001953125,
      "learning_rate": 4.929106224385475e-07,
      "loss": 0.0001,
      "reward": 1.7964286729693413,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 14000
    },
    {
      "completion_length": 258.7678699493408,
      "epoch": 2.3477094597426547,
      "grad_norm": 0.13462703461072748,
      "kl": 0.127471923828125,
      "learning_rate": 4.929077857047249e-07,
      "loss": 0.0001,
      "reward": 1.7250000834465027,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7339286115020514,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14002
    },
    {
      "completion_length": 253.50447368621826,
      "epoch": 2.3480447629825223,
      "grad_norm": 0.25273754623362055,
      "kl": 0.14434814453125,
      "learning_rate": 4.929049484116386e-07,
      "loss": 0.0001,
      "reward": 1.7607143446803093,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.7696428708732128,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14004
    },
    {
      "completion_length": 258.54018783569336,
      "epoch": 2.34838006622239,
      "grad_norm": 0.19787426664635152,
      "kl": 0.123138427734375,
      "learning_rate": 4.929021105592949e-07,
      "loss": 0.0001,
      "reward": 1.8232143074274063,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.827678594738245,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14006
    },
    {
      "completion_length": 271.6116199493408,
      "epoch": 2.3487153694622576,
      "grad_norm": 0.14992894677494317,
      "kl": 0.147003173828125,
      "learning_rate": 4.928992721477005e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7803571708500385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14008
    },
    {
      "completion_length": 263.29911708831787,
      "epoch": 2.349050672702125,
      "grad_norm": 0.0617426976233598,
      "kl": 0.131195068359375,
      "learning_rate": 4.928964331768621e-07,
      "loss": 0.0001,
      "reward": 1.7571429163217545,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7571428772062063,
      "rewards/format_reward_func": 1.0,
      "step": 14010
    },
    {
      "completion_length": 253.83929920196533,
      "epoch": 2.3493859759419924,
      "grad_norm": 0.07002114062756537,
      "kl": 0.154541015625,
      "learning_rate": 4.928935936467859e-07,
      "loss": 0.0002,
      "reward": 1.748214341700077,
      "reward_std": 0.012626906856894493,
      "rewards/equation_reward_func": 0.7526786122471094,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14012
    },
    {
      "completion_length": 250.7991180419922,
      "epoch": 2.34972127918186,
      "grad_norm": 0.25174814949002133,
      "kl": 0.28607177734375,
      "learning_rate": 4.928907535574786e-07,
      "loss": 0.0003,
      "reward": 1.7625000551342964,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14014
    },
    {
      "completion_length": 253.5625123977661,
      "epoch": 2.3500565824217277,
      "grad_norm": 0.17426929995654813,
      "kl": 0.119384765625,
      "learning_rate": 4.928879129089468e-07,
      "loss": 0.0001,
      "reward": 1.7732143551111221,
      "reward_std": 0.03788072057068348,
      "rewards/equation_reward_func": 0.7776786163449287,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14016
    },
    {
      "completion_length": 262.9821538925171,
      "epoch": 2.3503918856615953,
      "grad_norm": 0.1572711767458029,
      "kl": 0.310943603515625,
      "learning_rate": 4.928850717011969e-07,
      "loss": 0.0003,
      "reward": 1.6875000819563866,
      "reward_std": 0.07828682288527489,
      "rewards/equation_reward_func": 0.7008928880095482,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 14018
    },
    {
      "completion_length": 249.61608219146729,
      "epoch": 2.350727188901463,
      "grad_norm": 0.1388379776191087,
      "kl": 0.14971923828125,
      "learning_rate": 4.928822299342355e-07,
      "loss": 0.0001,
      "reward": 1.7910714969038963,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7955357395112514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14020
    },
    {
      "completion_length": 249.49554634094238,
      "epoch": 2.35106249214133,
      "grad_norm": 0.17107019366412024,
      "kl": 0.875244140625,
      "learning_rate": 4.928793876080692e-07,
      "loss": 0.0009,
      "reward": 1.7482143491506577,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7526785880327225,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14022
    },
    {
      "completion_length": 242.68304824829102,
      "epoch": 2.3513977953811978,
      "grad_norm": 0.1720167275916922,
      "kl": 0.295806884765625,
      "learning_rate": 4.928765447227044e-07,
      "loss": 0.0003,
      "reward": 1.753571480512619,
      "reward_std": 0.04545686487108469,
      "rewards/equation_reward_func": 0.7625000327825546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14024
    },
    {
      "completion_length": 244.09376430511475,
      "epoch": 2.3517330986210654,
      "grad_norm": 0.133536958377386,
      "kl": 0.123748779296875,
      "learning_rate": 4.928737012781479e-07,
      "loss": 0.0001,
      "reward": 1.737500049173832,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7419643308967352,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14026
    },
    {
      "completion_length": 246.2053680419922,
      "epoch": 2.352068401860933,
      "grad_norm": 0.26511754390693604,
      "kl": 0.18597412109375,
      "learning_rate": 4.92870857274406e-07,
      "loss": 0.0002,
      "reward": 1.7446428909897804,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7491071857511997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14028
    },
    {
      "completion_length": 239.23661708831787,
      "epoch": 2.3524037051008007,
      "grad_norm": 0.13234529295434244,
      "kl": 0.11456298828125,
      "learning_rate": 4.928680127114854e-07,
      "loss": 0.0001,
      "reward": 1.8250000178813934,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8250000197440386,
      "rewards/format_reward_func": 1.0,
      "step": 14030
    },
    {
      "completion_length": 241.6696548461914,
      "epoch": 2.3527390083406683,
      "grad_norm": 0.1214244184569556,
      "kl": 0.145416259765625,
      "learning_rate": 4.928651675893925e-07,
      "loss": 0.0001,
      "reward": 1.787500038743019,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7919643186032772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14032
    },
    {
      "completion_length": 240.49108409881592,
      "epoch": 2.3530743115805355,
      "grad_norm": 0.434751465835638,
      "kl": 1.688201904296875,
      "learning_rate": 4.92862321908134e-07,
      "loss": 0.0017,
      "reward": 1.773214340209961,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7776785865426064,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14034
    },
    {
      "completion_length": 250.0357265472412,
      "epoch": 2.353409614820403,
      "grad_norm": 0.15028585969842692,
      "kl": 0.15380859375,
      "learning_rate": 4.928594756677163e-07,
      "loss": 0.0002,
      "reward": 1.796428620815277,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 14036
    },
    {
      "completion_length": 231.30358219146729,
      "epoch": 2.3537449180602708,
      "grad_norm": 0.2420700954221213,
      "kl": 0.0980072021484375,
      "learning_rate": 4.928566288681461e-07,
      "loss": 0.0001,
      "reward": 1.8075893595814705,
      "reward_std": 0.0498762815259397,
      "rewards/equation_reward_func": 0.8133928757160902,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 14038
    },
    {
      "completion_length": 237.0892972946167,
      "epoch": 2.3540802213001384,
      "grad_norm": 0.14530807503245907,
      "kl": 0.228179931640625,
      "learning_rate": 4.928537815094299e-07,
      "loss": 0.0002,
      "reward": 1.8250000402331352,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8250000290572643,
      "rewards/format_reward_func": 1.0,
      "step": 14040
    },
    {
      "completion_length": 244.4509048461914,
      "epoch": 2.354415524540006,
      "grad_norm": 0.11180813148997293,
      "kl": 1.75067138671875,
      "learning_rate": 4.928509335915742e-07,
      "loss": 0.0017,
      "reward": 1.7803571820259094,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7848214525729418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14042
    },
    {
      "completion_length": 241.87947845458984,
      "epoch": 2.354750827779873,
      "grad_norm": 0.2982834793049277,
      "kl": 0.1754150390625,
      "learning_rate": 4.928480851145857e-07,
      "loss": 0.0002,
      "reward": 1.8321428894996643,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8321428783237934,
      "rewards/format_reward_func": 1.0,
      "step": 14044
    },
    {
      "completion_length": 242.92411708831787,
      "epoch": 2.355086131019741,
      "grad_norm": 0.08737988500493561,
      "kl": 0.383636474609375,
      "learning_rate": 4.928452360784709e-07,
      "loss": 0.0004,
      "reward": 1.751785770058632,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7562500294297934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14046
    },
    {
      "completion_length": 240.31697463989258,
      "epoch": 2.3554214342596085,
      "grad_norm": 0.20044606430797915,
      "kl": 0.116455078125,
      "learning_rate": 4.928423864832362e-07,
      "loss": 0.0001,
      "reward": 1.7428571805357933,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7517857290804386,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14048
    },
    {
      "completion_length": 235.4866189956665,
      "epoch": 2.355756737499476,
      "grad_norm": 0.2400343553634341,
      "kl": 0.223541259765625,
      "learning_rate": 4.928395363288883e-07,
      "loss": 0.0002,
      "reward": 1.8500000685453415,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.8589285761117935,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14050
    },
    {
      "completion_length": 224.12947463989258,
      "epoch": 2.3560920407393438,
      "grad_norm": 0.11322100214358818,
      "kl": 0.1439056396484375,
      "learning_rate": 4.928366856154338e-07,
      "loss": 0.0001,
      "reward": 1.7714286223053932,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7714286111295223,
      "rewards/format_reward_func": 1.0,
      "step": 14052
    },
    {
      "completion_length": 241.1250114440918,
      "epoch": 2.3564273439792114,
      "grad_norm": 0.2621128270019946,
      "kl": 0.87982177734375,
      "learning_rate": 4.928338343428791e-07,
      "loss": 0.0009,
      "reward": 1.7732143253087997,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7866071797907352,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 14054
    },
    {
      "completion_length": 237.6250114440918,
      "epoch": 2.3567626472190786,
      "grad_norm": 0.30913754789060155,
      "kl": 0.264068603515625,
      "learning_rate": 4.928309825112311e-07,
      "loss": 0.0003,
      "reward": 1.76071435213089,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143186032772,
      "rewards/format_reward_func": 1.0,
      "step": 14056
    },
    {
      "completion_length": 232.92858219146729,
      "epoch": 2.357097950458946,
      "grad_norm": 0.00804325255606672,
      "kl": 0.174468994140625,
      "learning_rate": 4.92828130120496e-07,
      "loss": 0.0002,
      "reward": 1.8035714775323868,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 14058
    },
    {
      "completion_length": 236.0357255935669,
      "epoch": 2.357433253698814,
      "grad_norm": 0.10639365932382884,
      "kl": 0.263580322265625,
      "learning_rate": 4.928252771706805e-07,
      "loss": 0.0003,
      "reward": 1.8250000476837158,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8250000178813934,
      "rewards/format_reward_func": 1.0,
      "step": 14060
    },
    {
      "completion_length": 228.81250858306885,
      "epoch": 2.3577685569386815,
      "grad_norm": 0.22752689175076363,
      "kl": 0.15313720703125,
      "learning_rate": 4.928224236617912e-07,
      "loss": 0.0002,
      "reward": 1.7928571924567223,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571775555611,
      "rewards/format_reward_func": 1.0,
      "step": 14062
    },
    {
      "completion_length": 244.99554634094238,
      "epoch": 2.358103860178549,
      "grad_norm": 0.13870489497186492,
      "kl": 0.566314697265625,
      "learning_rate": 4.928195695938346e-07,
      "loss": 0.0006,
      "reward": 1.7535714879631996,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7625000327825546,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14064
    },
    {
      "completion_length": 239.48215103149414,
      "epoch": 2.3584391634184163,
      "grad_norm": 0.28570805232389446,
      "kl": 0.2423095703125,
      "learning_rate": 4.928167149668174e-07,
      "loss": 0.0002,
      "reward": 1.8071429133415222,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428798139095,
      "rewards/format_reward_func": 1.0,
      "step": 14066
    },
    {
      "completion_length": 234.80804920196533,
      "epoch": 2.358774466658284,
      "grad_norm": 0.08342651959363287,
      "kl": 0.23736572265625,
      "learning_rate": 4.928138597807461e-07,
      "loss": 0.0002,
      "reward": 1.7357143387198448,
      "reward_std": 0.012626906856894493,
      "rewards/equation_reward_func": 0.7437500357627869,
      "rewards/format_reward_func": 0.9919642880558968,
      "step": 14068
    },
    {
      "completion_length": 242.87947368621826,
      "epoch": 2.3591097698981516,
      "grad_norm": 0.14099043974641465,
      "kl": 1.082427978515625,
      "learning_rate": 4.928110040356272e-07,
      "loss": 0.0011,
      "reward": 1.7553572058677673,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7598214484751225,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14070
    },
    {
      "completion_length": 228.77233123779297,
      "epoch": 2.359445073138019,
      "grad_norm": 0.17428381709094845,
      "kl": 0.250213623046875,
      "learning_rate": 4.928081477314674e-07,
      "loss": 0.0003,
      "reward": 1.805357202887535,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8098214529454708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14072
    },
    {
      "completion_length": 223.40179538726807,
      "epoch": 2.359780376377887,
      "grad_norm": 0.0054469211535804855,
      "kl": 0.17437744140625,
      "learning_rate": 4.928052908682731e-07,
      "loss": 0.0002,
      "reward": 1.783928595483303,
      "reward_std": 0.012626906856894493,
      "rewards/equation_reward_func": 0.7883929014205933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14074
    },
    {
      "completion_length": 228.59822273254395,
      "epoch": 2.3601156796177545,
      "grad_norm": 0.1825644958873576,
      "kl": 0.489898681640625,
      "learning_rate": 4.928024334460513e-07,
      "loss": 0.0005,
      "reward": 1.7553572058677673,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7598214671015739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14076
    },
    {
      "completion_length": 227.62947273254395,
      "epoch": 2.3604509828576217,
      "grad_norm": 0.28957667120115893,
      "kl": 0.1666259765625,
      "learning_rate": 4.92799575464808e-07,
      "loss": 0.0002,
      "reward": 1.825000062584877,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8250000216066837,
      "rewards/format_reward_func": 1.0,
      "step": 14078
    },
    {
      "completion_length": 228.77233123779297,
      "epoch": 2.3607862860974893,
      "grad_norm": 0.15762237281826855,
      "kl": 0.22186279296875,
      "learning_rate": 4.9279671692455e-07,
      "loss": 0.0002,
      "reward": 1.7964286133646965,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964286003261805,
      "rewards/format_reward_func": 1.0,
      "step": 14080
    },
    {
      "completion_length": 231.45536708831787,
      "epoch": 2.361121589337357,
      "grad_norm": 0.1459108784338708,
      "kl": 0.273040771484375,
      "learning_rate": 4.927938578252841e-07,
      "loss": 0.0003,
      "reward": 1.7892857939004898,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 14082
    },
    {
      "completion_length": 233.0803689956665,
      "epoch": 2.3614568925772246,
      "grad_norm": 0.35305294761911016,
      "kl": 0.149017333984375,
      "learning_rate": 4.927909981670166e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857413738966,
      "rewards/format_reward_func": 1.0,
      "step": 14084
    },
    {
      "completion_length": 242.602689743042,
      "epoch": 2.361792195817092,
      "grad_norm": 0.13617110402336785,
      "kl": 0.170318603515625,
      "learning_rate": 4.927881379497544e-07,
      "loss": 0.0002,
      "reward": 1.7357143685221672,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 14086
    },
    {
      "completion_length": 237.3884038925171,
      "epoch": 2.3621274990569594,
      "grad_norm": 0.135181646646026,
      "kl": 0.1388092041015625,
      "learning_rate": 4.927852771735037e-07,
      "loss": 0.0001,
      "reward": 1.775000050663948,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 14088
    },
    {
      "completion_length": 227.25000953674316,
      "epoch": 2.362462802296827,
      "grad_norm": 0.17809936062049062,
      "kl": 0.260589599609375,
      "learning_rate": 4.927824158382713e-07,
      "loss": 0.0003,
      "reward": 1.8053572103381157,
      "reward_std": 0.0328299580141902,
      "rewards/equation_reward_func": 0.8098214454948902,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14090
    },
    {
      "completion_length": 231.6562614440918,
      "epoch": 2.3627981055366947,
      "grad_norm": 0.2369996115166829,
      "kl": 0.130584716796875,
      "learning_rate": 4.927795539440638e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 14092
    },
    {
      "completion_length": 243.64286613464355,
      "epoch": 2.3631334087765623,
      "grad_norm": 0.15600298133011206,
      "kl": 0.2369384765625,
      "learning_rate": 4.927766914908878e-07,
      "loss": 0.0002,
      "reward": 1.7928571850061417,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 14094
    },
    {
      "completion_length": 229.9241189956665,
      "epoch": 2.36346871201643,
      "grad_norm": 0.2560262655058894,
      "kl": 0.256927490234375,
      "learning_rate": 4.927738284787497e-07,
      "loss": 0.0003,
      "reward": 1.7446429207921028,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7491071689873934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14096
    },
    {
      "completion_length": 238.24108409881592,
      "epoch": 2.3638040152562976,
      "grad_norm": 0.13041247601855738,
      "kl": 0.255828857421875,
      "learning_rate": 4.927709649076564e-07,
      "loss": 0.0003,
      "reward": 1.8035714849829674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714514553547,
      "rewards/format_reward_func": 1.0,
      "step": 14098
    },
    {
      "completion_length": 244.14286994934082,
      "epoch": 2.364139318496165,
      "grad_norm": 0.18896033080219826,
      "kl": 0.119232177734375,
      "learning_rate": 4.927681007776142e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000223517418,
      "rewards/format_reward_func": 1.0,
      "step": 14100
    },
    {
      "completion_length": 230.97322463989258,
      "epoch": 2.3644746217360324,
      "grad_norm": 0.2740426705203649,
      "kl": 0.307281494140625,
      "learning_rate": 4.927652360886299e-07,
      "loss": 0.0003,
      "reward": 1.7375000566244125,
      "reward_std": 0.017677669413387775,
      "rewards/equation_reward_func": 0.74196432903409,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14102
    },
    {
      "completion_length": 243.58036994934082,
      "epoch": 2.3648099249759,
      "grad_norm": 0.28152166534803,
      "kl": 0.393951416015625,
      "learning_rate": 4.927623708407099e-07,
      "loss": 0.0004,
      "reward": 1.7339286357164383,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.738392885774374,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14104
    },
    {
      "completion_length": 231.9553689956665,
      "epoch": 2.3651452282157677,
      "grad_norm": 0.1982005460353846,
      "kl": 0.279541015625,
      "learning_rate": 4.92759505033861e-07,
      "loss": 0.0003,
      "reward": 1.750000074505806,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7500000186264515,
      "rewards/format_reward_func": 1.0,
      "step": 14106
    },
    {
      "completion_length": 236.05358123779297,
      "epoch": 2.3654805314556353,
      "grad_norm": 0.3090752866343735,
      "kl": 0.13604736328125,
      "learning_rate": 4.927566386680897e-07,
      "loss": 0.0001,
      "reward": 1.739285796880722,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7482143118977547,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14108
    },
    {
      "completion_length": 236.45090579986572,
      "epoch": 2.365815834695503,
      "grad_norm": 0.3107913359321936,
      "kl": 0.2520751953125,
      "learning_rate": 4.927537717434027e-07,
      "loss": 0.0003,
      "reward": 1.7785714864730835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714529454708,
      "rewards/format_reward_func": 1.0,
      "step": 14110
    },
    {
      "completion_length": 236.11161994934082,
      "epoch": 2.36615113793537,
      "grad_norm": 0.29293999141656574,
      "kl": 0.412841796875,
      "learning_rate": 4.927509042598064e-07,
      "loss": 0.0004,
      "reward": 1.7250000834465027,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7250000201165676,
      "rewards/format_reward_func": 1.0,
      "step": 14112
    },
    {
      "completion_length": 225.22768783569336,
      "epoch": 2.3664864411752378,
      "grad_norm": 0.37355280432683863,
      "kl": 0.17962646484375,
      "learning_rate": 4.927480362173075e-07,
      "loss": 0.0002,
      "reward": 1.8000000715255737,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 14114
    },
    {
      "completion_length": 237.2276906967163,
      "epoch": 2.3668217444151054,
      "grad_norm": 0.0930373437809911,
      "kl": 0.1688232421875,
      "learning_rate": 4.927451676159127e-07,
      "loss": 0.0002,
      "reward": 1.785714328289032,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 14116
    },
    {
      "completion_length": 238.4107255935669,
      "epoch": 2.367157047654973,
      "grad_norm": 0.17907423202000464,
      "kl": 0.161224365234375,
      "learning_rate": 4.927422984556284e-07,
      "loss": 0.0002,
      "reward": 1.785714328289032,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7946428991854191,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14118
    },
    {
      "completion_length": 230.6384038925171,
      "epoch": 2.3674923508948407,
      "grad_norm": 0.21644035641179135,
      "kl": 0.115264892578125,
      "learning_rate": 4.927394287364614e-07,
      "loss": 0.0001,
      "reward": 1.7892857640981674,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857491970062,
      "rewards/format_reward_func": 1.0,
      "step": 14120
    },
    {
      "completion_length": 230.62947463989258,
      "epoch": 2.3678276541347083,
      "grad_norm": 0.1731059367453474,
      "kl": 0.129791259765625,
      "learning_rate": 4.927365584584184e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.764285746961832,
      "rewards/format_reward_func": 1.0,
      "step": 14122
    },
    {
      "completion_length": 239.2678680419922,
      "epoch": 2.3681629573745755,
      "grad_norm": 0.1786554966406942,
      "kl": 0.126251220703125,
      "learning_rate": 4.927336876215055e-07,
      "loss": 0.0001,
      "reward": 1.7767857760190964,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7812500149011612,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14124
    },
    {
      "completion_length": 235.62500858306885,
      "epoch": 2.368498260614443,
      "grad_norm": 0.15925308523575013,
      "kl": 0.12420654296875,
      "learning_rate": 4.927308162257299e-07,
      "loss": 0.0001,
      "reward": 1.7678571790456772,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571827709675,
      "rewards/format_reward_func": 1.0,
      "step": 14126
    },
    {
      "completion_length": 230.16965293884277,
      "epoch": 2.3688335638543108,
      "grad_norm": 0.08282613935284787,
      "kl": 0.15203857421875,
      "learning_rate": 4.927279442710979e-07,
      "loss": 0.0002,
      "reward": 1.7892857789993286,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7892857417464256,
      "rewards/format_reward_func": 1.0,
      "step": 14128
    },
    {
      "completion_length": 235.19197368621826,
      "epoch": 2.3691688670941784,
      "grad_norm": 0.2480821388756342,
      "kl": 0.123291015625,
      "learning_rate": 4.927250717576161e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 14130
    },
    {
      "completion_length": 226.73661708831787,
      "epoch": 2.369504170334046,
      "grad_norm": 0.17368345542346247,
      "kl": 0.1526947021484375,
      "learning_rate": 4.927221986852913e-07,
      "loss": 0.0002,
      "reward": 1.7357143387198448,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143238186836,
      "rewards/format_reward_func": 1.0,
      "step": 14132
    },
    {
      "completion_length": 239.01340293884277,
      "epoch": 2.3698394735739132,
      "grad_norm": 0.20367427730227183,
      "kl": 0.244476318359375,
      "learning_rate": 4.927193250541299e-07,
      "loss": 0.0002,
      "reward": 1.7482143491506577,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7526785992085934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14134
    },
    {
      "completion_length": 237.95536613464355,
      "epoch": 2.370174776813781,
      "grad_norm": 0.0815353276461526,
      "kl": 0.142181396484375,
      "learning_rate": 4.927164508641388e-07,
      "loss": 0.0001,
      "reward": 1.7500000596046448,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000447034836,
      "rewards/format_reward_func": 1.0,
      "step": 14136
    },
    {
      "completion_length": 229.46429538726807,
      "epoch": 2.3705100800536485,
      "grad_norm": 0.10713861064394091,
      "kl": 0.175323486328125,
      "learning_rate": 4.927135761153242e-07,
      "loss": 0.0002,
      "reward": 1.7767857909202576,
      "reward_std": 0.0328299580141902,
      "rewards/equation_reward_func": 0.7812500260770321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14138
    },
    {
      "completion_length": 225.665189743042,
      "epoch": 2.370845383293516,
      "grad_norm": 0.17673572626422573,
      "kl": 0.131622314453125,
      "learning_rate": 4.927107008076932e-07,
      "loss": 0.0001,
      "reward": 1.7160715013742447,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7205357495695353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14140
    },
    {
      "completion_length": 227.540189743042,
      "epoch": 2.3711806865333838,
      "grad_norm": 0.33349945659697183,
      "kl": 0.13836669921875,
      "learning_rate": 4.92707824941252e-07,
      "loss": 0.0001,
      "reward": 1.7839286476373672,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14142
    },
    {
      "completion_length": 220.33929347991943,
      "epoch": 2.3715159897732514,
      "grad_norm": 0.1168489542544372,
      "kl": 0.1265869140625,
      "learning_rate": 4.927049485160075e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 14144
    },
    {
      "completion_length": 238.7634048461914,
      "epoch": 2.3718512930131186,
      "grad_norm": 0.1929672755226574,
      "kl": 0.138336181640625,
      "learning_rate": 4.927020715319662e-07,
      "loss": 0.0001,
      "reward": 1.8267857506871223,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8312500268220901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14146
    },
    {
      "completion_length": 226.50893783569336,
      "epoch": 2.3721865962529862,
      "grad_norm": 0.22830195961372327,
      "kl": 0.1256103515625,
      "learning_rate": 4.926991939891346e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428962051868,
      "rewards/format_reward_func": 1.0,
      "step": 14148
    },
    {
      "completion_length": 223.10268878936768,
      "epoch": 2.372521899492854,
      "grad_norm": 0.1226224587835622,
      "kl": 0.162078857421875,
      "learning_rate": 4.926963158875197e-07,
      "loss": 0.0002,
      "reward": 1.8214286044239998,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8214286081492901,
      "rewards/format_reward_func": 1.0,
      "step": 14150
    },
    {
      "completion_length": 239.80358123779297,
      "epoch": 2.3728572027327215,
      "grad_norm": 0.13037067304308758,
      "kl": 0.135223388671875,
      "learning_rate": 4.926934372271277e-07,
      "loss": 0.0001,
      "reward": 1.7625000849366188,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7669643275439739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14152
    },
    {
      "completion_length": 234.99554538726807,
      "epoch": 2.373192505972589,
      "grad_norm": 0.2723590433917922,
      "kl": 0.140380859375,
      "learning_rate": 4.926905580079655e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7892857305705547,
      "rewards/format_reward_func": 1.0,
      "step": 14154
    },
    {
      "completion_length": 249.51786613464355,
      "epoch": 2.3735278092124563,
      "grad_norm": 0.2768983996081548,
      "kl": 0.42657470703125,
      "learning_rate": 4.926876782300397e-07,
      "loss": 0.0004,
      "reward": 1.76071435213089,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7696428894996643,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14156
    },
    {
      "completion_length": 228.89733028411865,
      "epoch": 2.373863112452324,
      "grad_norm": 0.2700517762375239,
      "kl": 0.125091552734375,
      "learning_rate": 4.926847978933568e-07,
      "loss": 0.0001,
      "reward": 1.760714367032051,
      "reward_std": 0.04545686487108469,
      "rewards/equation_reward_func": 0.7696428839117289,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14158
    },
    {
      "completion_length": 228.62947463989258,
      "epoch": 2.3741984156921916,
      "grad_norm": 0.16048736519894619,
      "kl": 0.1554718017578125,
      "learning_rate": 4.926819169979235e-07,
      "loss": 0.0002,
      "reward": 1.7428572252392769,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571674972773,
      "rewards/format_reward_func": 1.0,
      "step": 14160
    },
    {
      "completion_length": 227.23215293884277,
      "epoch": 2.3745337189320592,
      "grad_norm": 0.18227184233243968,
      "kl": 0.13006591796875,
      "learning_rate": 4.926790355437466e-07,
      "loss": 0.0001,
      "reward": 1.7125000655651093,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7169643193483353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14162
    },
    {
      "completion_length": 241.8660831451416,
      "epoch": 2.374869022171927,
      "grad_norm": 0.289783282635767,
      "kl": 0.160186767578125,
      "learning_rate": 4.926761535308324e-07,
      "loss": 0.0002,
      "reward": 1.844642885029316,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.849107176065445,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14164
    },
    {
      "completion_length": 232.65179443359375,
      "epoch": 2.3752043254117945,
      "grad_norm": 0.1962872977790209,
      "kl": 0.124542236328125,
      "learning_rate": 4.926732709591878e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 14166
    },
    {
      "completion_length": 240.821439743042,
      "epoch": 2.3755396286516617,
      "grad_norm": 0.22684733865237583,
      "kl": 0.1270751953125,
      "learning_rate": 4.926703878288194e-07,
      "loss": 0.0001,
      "reward": 1.712500087916851,
      "reward_std": 0.07323605846613646,
      "rewards/equation_reward_func": 0.716964315623045,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14168
    },
    {
      "completion_length": 239.41518878936768,
      "epoch": 2.3758749318915293,
      "grad_norm": 0.21397832178290516,
      "kl": 0.120819091796875,
      "learning_rate": 4.926675041397338e-07,
      "loss": 0.0001,
      "reward": 1.7339286282658577,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7383928783237934,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14170
    },
    {
      "completion_length": 232.6428689956665,
      "epoch": 2.376210235131397,
      "grad_norm": 0.18256904282426623,
      "kl": 0.129730224609375,
      "learning_rate": 4.926646198919375e-07,
      "loss": 0.0001,
      "reward": 1.7821428999304771,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428887546062,
      "rewards/format_reward_func": 1.0,
      "step": 14172
    },
    {
      "completion_length": 226.29911613464355,
      "epoch": 2.3765455383712646,
      "grad_norm": 0.2511379796899202,
      "kl": 0.1105499267578125,
      "learning_rate": 4.926617350854373e-07,
      "loss": 0.0001,
      "reward": 1.728571504354477,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714596509933,
      "rewards/format_reward_func": 1.0,
      "step": 14174
    },
    {
      "completion_length": 234.5357265472412,
      "epoch": 2.3768808416111322,
      "grad_norm": 0.11663380199458673,
      "kl": 0.11688232421875,
      "learning_rate": 4.926588497202399e-07,
      "loss": 0.0001,
      "reward": 1.7821428924798965,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 14176
    },
    {
      "completion_length": 232.5759048461914,
      "epoch": 2.3772161448509994,
      "grad_norm": 0.14246938662770156,
      "kl": 0.097503662109375,
      "learning_rate": 4.926559637963518e-07,
      "loss": 0.0001,
      "reward": 1.733928643167019,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7383928932249546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14178
    },
    {
      "completion_length": 243.15179538726807,
      "epoch": 2.377551448090867,
      "grad_norm": 0.1807404842578879,
      "kl": 0.113311767578125,
      "learning_rate": 4.926530773137798e-07,
      "loss": 0.0001,
      "reward": 1.7464286610484123,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 14180
    },
    {
      "completion_length": 237.43750953674316,
      "epoch": 2.3778867513307347,
      "grad_norm": 0.2560356006133875,
      "kl": 0.116668701171875,
      "learning_rate": 4.926501902725303e-07,
      "loss": 0.0001,
      "reward": 1.8196429014205933,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8241071775555611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14182
    },
    {
      "completion_length": 238.04911708831787,
      "epoch": 2.3782220545706023,
      "grad_norm": 0.17475930712994994,
      "kl": 0.139495849609375,
      "learning_rate": 4.926473026726101e-07,
      "loss": 0.0001,
      "reward": 1.6910715028643608,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.6955357547849417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14184
    },
    {
      "completion_length": 245.30358123779297,
      "epoch": 2.37855735781047,
      "grad_norm": 0.17446587499780564,
      "kl": 0.1220703125,
      "learning_rate": 4.92644414514026e-07,
      "loss": 0.0001,
      "reward": 1.7517857775092125,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7562500275671482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14186
    },
    {
      "completion_length": 234.08036518096924,
      "epoch": 2.3788926610503376,
      "grad_norm": 0.27303907383285336,
      "kl": 0.19439697265625,
      "learning_rate": 4.926415257967843e-07,
      "loss": 0.0002,
      "reward": 1.7571429386734962,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 14188
    },
    {
      "completion_length": 253.4732255935669,
      "epoch": 2.3792279642902048,
      "grad_norm": 0.14374585376478768,
      "kl": 0.106475830078125,
      "learning_rate": 4.92638636520892e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8035714700818062,
      "rewards/format_reward_func": 1.0,
      "step": 14190
    },
    {
      "completion_length": 238.92411613464355,
      "epoch": 2.3795632675300724,
      "grad_norm": 0.09037984547445888,
      "kl": 0.112701416015625,
      "learning_rate": 4.926357466863556e-07,
      "loss": 0.0001,
      "reward": 1.7750000432133675,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7750000208616257,
      "rewards/format_reward_func": 1.0,
      "step": 14192
    },
    {
      "completion_length": 250.32143878936768,
      "epoch": 2.37989857076994,
      "grad_norm": 0.2056104027875523,
      "kl": 0.109954833984375,
      "learning_rate": 4.926328562931816e-07,
      "loss": 0.0001,
      "reward": 1.712500087916851,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7169643193483353,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14194
    },
    {
      "completion_length": 254.5714406967163,
      "epoch": 2.3802338740098077,
      "grad_norm": 0.21454443708666152,
      "kl": 0.124755859375,
      "learning_rate": 4.92629965341377e-07,
      "loss": 0.0001,
      "reward": 1.800000049173832,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 14196
    },
    {
      "completion_length": 245.61161708831787,
      "epoch": 2.3805691772496753,
      "grad_norm": 0.21721769675845912,
      "kl": 0.1088714599609375,
      "learning_rate": 4.92627073830948e-07,
      "loss": 0.0001,
      "reward": 1.796428643167019,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 14198
    },
    {
      "completion_length": 251.32590675354004,
      "epoch": 2.3809044804895425,
      "grad_norm": 0.14698313947537994,
      "kl": 0.118255615234375,
      "learning_rate": 4.926241817619017e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 14200
    },
    {
      "completion_length": 247.36161518096924,
      "epoch": 2.38123978372941,
      "grad_norm": 0.2984387094846763,
      "kl": 0.11102294921875,
      "learning_rate": 4.926212891342445e-07,
      "loss": 0.0001,
      "reward": 1.8160714954137802,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.820535734295845,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14202
    },
    {
      "completion_length": 246.8259038925171,
      "epoch": 2.3815750869692778,
      "grad_norm": 0.19331540875940945,
      "kl": 0.115264892578125,
      "learning_rate": 4.926183959479833e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.753571467474103,
      "rewards/format_reward_func": 1.0,
      "step": 14204
    },
    {
      "completion_length": 245.6026906967163,
      "epoch": 2.3819103902091454,
      "grad_norm": 0.0032176377544802333,
      "kl": 0.115203857421875,
      "learning_rate": 4.926155022031244e-07,
      "loss": 0.0001,
      "reward": 1.7928571924567223,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 14206
    },
    {
      "completion_length": 240.30357933044434,
      "epoch": 2.382245693449013,
      "grad_norm": 0.22095254078246251,
      "kl": 0.111480712890625,
      "learning_rate": 4.926126078996747e-07,
      "loss": 0.0001,
      "reward": 1.8017857745289803,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8062500283122063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14208
    },
    {
      "completion_length": 252.48661613464355,
      "epoch": 2.3825809966888807,
      "grad_norm": 0.268309403775508,
      "kl": 0.125274658203125,
      "learning_rate": 4.926097130376409e-07,
      "loss": 0.0001,
      "reward": 1.7196429297327995,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7241071723401546,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14210
    },
    {
      "completion_length": 244.227689743042,
      "epoch": 2.382916299928748,
      "grad_norm": 0.1499611496941687,
      "kl": 0.1090087890625,
      "learning_rate": 4.926068176170295e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714285962283611,
      "rewards/format_reward_func": 1.0,
      "step": 14212
    },
    {
      "completion_length": 250.77233123779297,
      "epoch": 2.3832516031686155,
      "grad_norm": 0.108530994446496,
      "kl": 0.17041015625,
      "learning_rate": 4.926039216378472e-07,
      "loss": 0.0002,
      "reward": 1.7696429267525673,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7741071619093418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14214
    },
    {
      "completion_length": 239.0535831451416,
      "epoch": 2.383586906408483,
      "grad_norm": 0.14041276773962685,
      "kl": 0.10272216796875,
      "learning_rate": 4.926010251001008e-07,
      "loss": 0.0001,
      "reward": 1.8321429044008255,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8321428745985031,
      "rewards/format_reward_func": 1.0,
      "step": 14216
    },
    {
      "completion_length": 237.2901906967163,
      "epoch": 2.3839222096483508,
      "grad_norm": 0.12151716109331748,
      "kl": 0.14166259765625,
      "learning_rate": 4.92598128003797e-07,
      "loss": 0.0001,
      "reward": 1.7196429297327995,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7241071742027998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14218
    },
    {
      "completion_length": 240.2455472946167,
      "epoch": 2.3842575128882184,
      "grad_norm": 0.16364630154217485,
      "kl": 0.126220703125,
      "learning_rate": 4.925952303489422e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7535714693367481,
      "rewards/format_reward_func": 1.0,
      "step": 14220
    },
    {
      "completion_length": 233.61608219146729,
      "epoch": 2.3845928161280856,
      "grad_norm": 0.17529658175337104,
      "kl": 0.133697509765625,
      "learning_rate": 4.925923321355433e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143092900515,
      "rewards/format_reward_func": 1.0,
      "step": 14222
    },
    {
      "completion_length": 241.4732255935669,
      "epoch": 2.3849281193679532,
      "grad_norm": 0.09252615676509375,
      "kl": 0.133056640625,
      "learning_rate": 4.925894333636069e-07,
      "loss": 0.0001,
      "reward": 1.7285714969038963,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7285714671015739,
      "rewards/format_reward_func": 1.0,
      "step": 14224
    },
    {
      "completion_length": 234.6294765472412,
      "epoch": 2.385263422607821,
      "grad_norm": 0.12747518818637713,
      "kl": 0.110015869140625,
      "learning_rate": 4.925865340331398e-07,
      "loss": 0.0001,
      "reward": 1.7214286550879478,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7214286122471094,
      "rewards/format_reward_func": 1.0,
      "step": 14226
    },
    {
      "completion_length": 242.12947463989258,
      "epoch": 2.3855987258476885,
      "grad_norm": 0.23217568953704362,
      "kl": 0.135894775390625,
      "learning_rate": 4.925836341441484e-07,
      "loss": 0.0001,
      "reward": 1.6642858162522316,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.6642857529222965,
      "rewards/format_reward_func": 1.0,
      "step": 14228
    },
    {
      "completion_length": 241.42858409881592,
      "epoch": 2.385934029087556,
      "grad_norm": 0.32283599449486605,
      "kl": 0.107147216796875,
      "learning_rate": 4.925807336966395e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 14230
    },
    {
      "completion_length": 241.62054824829102,
      "epoch": 2.3862693323274238,
      "grad_norm": 0.4361940214247699,
      "kl": 0.099517822265625,
      "learning_rate": 4.925778326906198e-07,
      "loss": 0.0001,
      "reward": 1.7821429371833801,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428757160902,
      "rewards/format_reward_func": 1.0,
      "step": 14232
    },
    {
      "completion_length": 241.95983409881592,
      "epoch": 2.3866046355672914,
      "grad_norm": 0.15672447960175115,
      "kl": 0.114654541015625,
      "learning_rate": 4.925749311260962e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 14234
    },
    {
      "completion_length": 247.38840293884277,
      "epoch": 2.3869399388071586,
      "grad_norm": 0.12357675091633724,
      "kl": 0.127593994140625,
      "learning_rate": 4.92572029003075e-07,
      "loss": 0.0001,
      "reward": 1.8392857760190964,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8392857313156128,
      "rewards/format_reward_func": 1.0,
      "step": 14236
    },
    {
      "completion_length": 238.7009048461914,
      "epoch": 2.3872752420470262,
      "grad_norm": 0.34950113473575367,
      "kl": 0.118927001953125,
      "learning_rate": 4.92569126321563e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143059372902,
      "rewards/format_reward_func": 1.0,
      "step": 14238
    },
    {
      "completion_length": 238.56250762939453,
      "epoch": 2.387610545286894,
      "grad_norm": 0.005908223008530719,
      "kl": 0.1171875,
      "learning_rate": 4.925662230815671e-07,
      "loss": 0.0001,
      "reward": 1.703571505844593,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7035714685916901,
      "rewards/format_reward_func": 1.0,
      "step": 14240
    },
    {
      "completion_length": 238.3705472946167,
      "epoch": 2.3879458485267615,
      "grad_norm": 0.23473823838715496,
      "kl": 0.10906982421875,
      "learning_rate": 4.925633192830937e-07,
      "loss": 0.0001,
      "reward": 1.7660714909434319,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7705357410013676,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14242
    },
    {
      "completion_length": 234.7678680419922,
      "epoch": 2.3882811517666287,
      "grad_norm": 0.26786727999529003,
      "kl": 0.112030029296875,
      "learning_rate": 4.925604149261497e-07,
      "loss": 0.0001,
      "reward": 1.7464286461472511,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 14244
    },
    {
      "completion_length": 233.0223331451416,
      "epoch": 2.3886164550064963,
      "grad_norm": 0.18724417602482607,
      "kl": 0.10772705078125,
      "learning_rate": 4.925575100107415e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714286185801029,
      "rewards/format_reward_func": 1.0,
      "step": 14246
    },
    {
      "completion_length": 240.4776906967163,
      "epoch": 2.388951758246364,
      "grad_norm": 0.1619747174711402,
      "kl": 0.1134033203125,
      "learning_rate": 4.925546045368762e-07,
      "loss": 0.0001,
      "reward": 1.850000038743019,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8500000201165676,
      "rewards/format_reward_func": 1.0,
      "step": 14248
    },
    {
      "completion_length": 235.22322463989258,
      "epoch": 2.3892870614862316,
      "grad_norm": 0.14819214172171727,
      "kl": 0.11566162109375,
      "learning_rate": 4.925516985045601e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 14250
    },
    {
      "completion_length": 238.0446538925171,
      "epoch": 2.3896223647260992,
      "grad_norm": 0.14841388129257704,
      "kl": 0.229644775390625,
      "learning_rate": 4.925487919138001e-07,
      "loss": 0.0002,
      "reward": 1.7928571924567223,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571850061417,
      "rewards/format_reward_func": 1.0,
      "step": 14252
    },
    {
      "completion_length": 227.77679634094238,
      "epoch": 2.389957667965967,
      "grad_norm": 0.20086124023025853,
      "kl": 0.114776611328125,
      "learning_rate": 4.925458847646029e-07,
      "loss": 0.0001,
      "reward": 1.8107143566012383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.810714315623045,
      "rewards/format_reward_func": 1.0,
      "step": 14254
    },
    {
      "completion_length": 233.6250114440918,
      "epoch": 2.3902929712058345,
      "grad_norm": 0.2715224088342795,
      "kl": 0.108551025390625,
      "learning_rate": 4.925429770569751e-07,
      "loss": 0.0001,
      "reward": 1.7839286252856255,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.788392897695303,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14256
    },
    {
      "completion_length": 227.0446538925171,
      "epoch": 2.3906282744457017,
      "grad_norm": 0.18961583525723497,
      "kl": 0.1087646484375,
      "learning_rate": 4.925400687909235e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143294066191,
      "rewards/format_reward_func": 1.0,
      "step": 14258
    },
    {
      "completion_length": 240.30804443359375,
      "epoch": 2.3909635776855693,
      "grad_norm": 0.20396875238100765,
      "kl": 0.1153564453125,
      "learning_rate": 4.925371599664547e-07,
      "loss": 0.0001,
      "reward": 1.8125000298023224,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.816964328289032,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14260
    },
    {
      "completion_length": 235.96429538726807,
      "epoch": 2.391298880925437,
      "grad_norm": 1.6884351687415977,
      "kl": 0.130615234375,
      "learning_rate": 4.925342505835754e-07,
      "loss": 0.0001,
      "reward": 1.8107143118977547,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.810714328661561,
      "rewards/format_reward_func": 1.0,
      "step": 14262
    },
    {
      "completion_length": 245.05804538726807,
      "epoch": 2.3916341841653046,
      "grad_norm": 0.1875430097990074,
      "kl": 0.109161376953125,
      "learning_rate": 4.925313406422924e-07,
      "loss": 0.0001,
      "reward": 1.7642858028411865,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857413738966,
      "rewards/format_reward_func": 1.0,
      "step": 14264
    },
    {
      "completion_length": 240.56250953674316,
      "epoch": 2.3919694874051722,
      "grad_norm": 0.17942496787693438,
      "kl": 0.10748291015625,
      "learning_rate": 4.925284301426122e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7785714454948902,
      "rewards/format_reward_func": 1.0,
      "step": 14266
    },
    {
      "completion_length": 237.4821538925171,
      "epoch": 2.3923047906450394,
      "grad_norm": 0.09619269901935326,
      "kl": 0.11822509765625,
      "learning_rate": 4.925255190845418e-07,
      "loss": 0.0001,
      "reward": 1.7892857491970062,
      "reward_std": 0.005050762556493282,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 14268
    },
    {
      "completion_length": 235.8928680419922,
      "epoch": 2.392640093884907,
      "grad_norm": 0.11474742306725783,
      "kl": 0.108551025390625,
      "learning_rate": 4.925226074680876e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7500000223517418,
      "rewards/format_reward_func": 1.0,
      "step": 14270
    },
    {
      "completion_length": 247.57143878936768,
      "epoch": 2.3929753971247747,
      "grad_norm": 0.15741634440553448,
      "kl": 0.1328125,
      "learning_rate": 4.925196952932565e-07,
      "loss": 0.0001,
      "reward": 1.792857214808464,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 14272
    },
    {
      "completion_length": 251.08929920196533,
      "epoch": 2.3933107003646423,
      "grad_norm": 0.10766437431018219,
      "kl": 0.12640380859375,
      "learning_rate": 4.925167825600552e-07,
      "loss": 0.0001,
      "reward": 1.7750000432133675,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.775000037625432,
      "rewards/format_reward_func": 1.0,
      "step": 14274
    },
    {
      "completion_length": 257.86609077453613,
      "epoch": 2.39364600360451,
      "grad_norm": 0.16834165502722323,
      "kl": 0.108551025390625,
      "learning_rate": 4.925138692684902e-07,
      "loss": 0.0001,
      "reward": 1.8214286267757416,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8214286006987095,
      "rewards/format_reward_func": 1.0,
      "step": 14276
    },
    {
      "completion_length": 247.4509048461914,
      "epoch": 2.3939813068443776,
      "grad_norm": 0.17944890412209027,
      "kl": 0.102783203125,
      "learning_rate": 4.925109554185685e-07,
      "loss": 0.0001,
      "reward": 1.801785759627819,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8062500283122063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14278
    },
    {
      "completion_length": 251.76340198516846,
      "epoch": 2.394316610084245,
      "grad_norm": 0.11806261551808396,
      "kl": 0.233489990234375,
      "learning_rate": 4.925080410102965e-07,
      "loss": 0.0002,
      "reward": 1.6642857864499092,
      "reward_std": 0.03030457627028227,
      "rewards/equation_reward_func": 0.6732143275439739,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14280
    },
    {
      "completion_length": 256.3169755935669,
      "epoch": 2.3946519133241124,
      "grad_norm": 0.282034937595845,
      "kl": 0.12908935546875,
      "learning_rate": 4.925051260436812e-07,
      "loss": 0.0001,
      "reward": 1.7750000581145287,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000283122063,
      "rewards/format_reward_func": 1.0,
      "step": 14282
    },
    {
      "completion_length": 252.32144260406494,
      "epoch": 2.39498721656398,
      "grad_norm": 0.17036559088687567,
      "kl": 0.13946533203125,
      "learning_rate": 4.925022105187291e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714790225029,
      "rewards/format_reward_func": 1.0,
      "step": 14284
    },
    {
      "completion_length": 258.7276887893677,
      "epoch": 2.3953225198038477,
      "grad_norm": 0.1503807619374923,
      "kl": 0.144378662109375,
      "learning_rate": 4.924992944354472e-07,
      "loss": 0.0001,
      "reward": 1.7892857640981674,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 14286
    },
    {
      "completion_length": 244.72768783569336,
      "epoch": 2.3956578230437153,
      "grad_norm": 0.21012577005227273,
      "kl": 0.1162109375,
      "learning_rate": 4.924963777938417e-07,
      "loss": 0.0001,
      "reward": 1.8000000715255737,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 14288
    },
    {
      "completion_length": 248.89733219146729,
      "epoch": 2.3959931262835825,
      "grad_norm": 0.3899478536470825,
      "kl": 0.423553466796875,
      "learning_rate": 4.924934605939199e-07,
      "loss": 0.0004,
      "reward": 1.7821428924798965,
      "reward_std": 0.045456865802407265,
      "rewards/equation_reward_func": 0.7910714633762836,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14290
    },
    {
      "completion_length": 246.04018688201904,
      "epoch": 2.39632842952345,
      "grad_norm": 0.13040493195453964,
      "kl": 0.135955810546875,
      "learning_rate": 4.924905428356882e-07,
      "loss": 0.0001,
      "reward": 1.7732143476605415,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7776786088943481,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14292
    },
    {
      "completion_length": 256.4910840988159,
      "epoch": 2.396663732763318,
      "grad_norm": 0.22800534481410561,
      "kl": 0.128997802734375,
      "learning_rate": 4.924876245191532e-07,
      "loss": 0.0001,
      "reward": 1.7839286178350449,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7883928902447224,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14294
    },
    {
      "completion_length": 260.1651887893677,
      "epoch": 2.3969990360031854,
      "grad_norm": 0.25404098138978654,
      "kl": 0.384521484375,
      "learning_rate": 4.924847056443219e-07,
      "loss": 0.0004,
      "reward": 1.8250000774860382,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8250000216066837,
      "rewards/format_reward_func": 1.0,
      "step": 14296
    },
    {
      "completion_length": 250.84376335144043,
      "epoch": 2.397334339243053,
      "grad_norm": 0.2945103278601156,
      "kl": 0.115875244140625,
      "learning_rate": 4.924817862112008e-07,
      "loss": 0.0001,
      "reward": 1.751785784959793,
      "reward_std": 0.05808377172797918,
      "rewards/equation_reward_func": 0.7651786096394062,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 14298
    },
    {
      "completion_length": 262.4776916503906,
      "epoch": 2.3976696424829207,
      "grad_norm": 0.13702960543474418,
      "kl": 0.113006591796875,
      "learning_rate": 4.924788662197968e-07,
      "loss": 0.0001,
      "reward": 1.7053572237491608,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7098214775323868,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14300
    },
    {
      "completion_length": 260.9151906967163,
      "epoch": 2.398004945722788,
      "grad_norm": 0.16906523532274226,
      "kl": 0.54217529296875,
      "learning_rate": 4.924759456701166e-07,
      "loss": 0.0005,
      "reward": 1.7392857745289803,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7482143081724644,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14302
    },
    {
      "completion_length": 258.808048248291,
      "epoch": 2.3983402489626555,
      "grad_norm": 0.37236756985967784,
      "kl": 0.363616943359375,
      "learning_rate": 4.924730245621669e-07,
      "loss": 0.0004,
      "reward": 1.8000000417232513,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8000000342726707,
      "rewards/format_reward_func": 1.0,
      "step": 14304
    },
    {
      "completion_length": 255.62054538726807,
      "epoch": 2.398675552202523,
      "grad_norm": 0.0919150041745982,
      "kl": 0.917205810546875,
      "learning_rate": 4.924701028959543e-07,
      "loss": 0.0009,
      "reward": 1.7750000432133675,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7750000394880772,
      "rewards/format_reward_func": 1.0,
      "step": 14306
    },
    {
      "completion_length": 247.10269165039062,
      "epoch": 2.399010855442391,
      "grad_norm": 0.20273143904873656,
      "kl": 0.12310791015625,
      "learning_rate": 4.924671806714856e-07,
      "loss": 0.0001,
      "reward": 1.7589286342263222,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928917348385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14308
    },
    {
      "completion_length": 246.0580472946167,
      "epoch": 2.3993461586822584,
      "grad_norm": 0.19527974290438713,
      "kl": 0.1595458984375,
      "learning_rate": 4.924642578887675e-07,
      "loss": 0.0002,
      "reward": 1.8339286148548126,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8383928760886192,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14310
    },
    {
      "completion_length": 260.1919746398926,
      "epoch": 2.3996814619221256,
      "grad_norm": 0.4782948238589906,
      "kl": 0.147247314453125,
      "learning_rate": 4.924613345478069e-07,
      "loss": 0.0001,
      "reward": 1.6857143640518188,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.6857143305242062,
      "rewards/format_reward_func": 1.0,
      "step": 14312
    },
    {
      "completion_length": 261.4419775009155,
      "epoch": 2.4000167651619932,
      "grad_norm": 0.5121912858089605,
      "kl": 2.307708740234375,
      "learning_rate": 4.924584106486105e-07,
      "loss": 0.0023,
      "reward": 1.7446429207921028,
      "reward_std": 0.06818529684096575,
      "rewards/equation_reward_func": 0.7580357529222965,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 14314
    },
    {
      "completion_length": 245.25447463989258,
      "epoch": 2.400352068401861,
      "grad_norm": 0.18907697095639242,
      "kl": 0.214569091796875,
      "learning_rate": 4.924554861911847e-07,
      "loss": 0.0002,
      "reward": 1.7571429312229156,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428902447224,
      "rewards/format_reward_func": 1.0,
      "step": 14316
    },
    {
      "completion_length": 242.28125953674316,
      "epoch": 2.4006873716417285,
      "grad_norm": 0.24829432582435484,
      "kl": 1.02117919921875,
      "learning_rate": 4.924525611755366e-07,
      "loss": 0.001,
      "reward": 1.8035714849829674,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.8125000260770321,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14318
    },
    {
      "completion_length": 246.0178689956665,
      "epoch": 2.401022674881596,
      "grad_norm": 0.5520300313907123,
      "kl": 0.346954345703125,
      "learning_rate": 4.924496356016729e-07,
      "loss": 0.0003,
      "reward": 1.7267857864499092,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7401785887777805,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 14320
    },
    {
      "completion_length": 255.9821538925171,
      "epoch": 2.401357978121464,
      "grad_norm": 0.23108229551636367,
      "kl": 0.32366943359375,
      "learning_rate": 4.924467094696001e-07,
      "loss": 0.0003,
      "reward": 1.728571504354477,
      "reward_std": 0.06060915347188711,
      "rewards/equation_reward_func": 0.7375000342726707,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14322
    },
    {
      "completion_length": 258.02679538726807,
      "epoch": 2.401693281361331,
      "grad_norm": 0.5556203153502393,
      "kl": 1.0614013671875,
      "learning_rate": 4.924437827793251e-07,
      "loss": 0.0011,
      "reward": 1.705357201397419,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.718750037252903,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 14324
    },
    {
      "completion_length": 250.02233505249023,
      "epoch": 2.4020285846011986,
      "grad_norm": 0.33785567885457496,
      "kl": 0.931396484375,
      "learning_rate": 4.924408555308547e-07,
      "loss": 0.0009,
      "reward": 1.7732143327593803,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7866071797907352,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 14326
    },
    {
      "completion_length": 257.00894260406494,
      "epoch": 2.4023638878410662,
      "grad_norm": 0.5425677862198246,
      "kl": 3.427093505859375,
      "learning_rate": 4.924379277241955e-07,
      "loss": 0.0034,
      "reward": 1.7357143685221672,
      "reward_std": 0.0505076264962554,
      "rewards/equation_reward_func": 0.7446428909897804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14328
    },
    {
      "completion_length": 241.6562614440918,
      "epoch": 2.402699191080934,
      "grad_norm": 0.08042267566696397,
      "kl": 0.248321533203125,
      "learning_rate": 4.924349993593543e-07,
      "loss": 0.0002,
      "reward": 1.7821428924798965,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.7910714633762836,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14330
    },
    {
      "completion_length": 250.99108409881592,
      "epoch": 2.4030344943208015,
      "grad_norm": 0.23087088388512614,
      "kl": 0.36553955078125,
      "learning_rate": 4.924320704363378e-07,
      "loss": 0.0004,
      "reward": 1.7357143461704254,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 14332
    },
    {
      "completion_length": 253.00447750091553,
      "epoch": 2.4033697975606687,
      "grad_norm": 0.32966255660926874,
      "kl": 0.310394287109375,
      "learning_rate": 4.924291409551529e-07,
      "loss": 0.0003,
      "reward": 1.717857226729393,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7267857417464256,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14334
    },
    {
      "completion_length": 240.29465103149414,
      "epoch": 2.4037051008005363,
      "grad_norm": 0.23779095458913324,
      "kl": 0.32281494140625,
      "learning_rate": 4.924262109158061e-07,
      "loss": 0.0003,
      "reward": 1.7839286252856255,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7973214611411095,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 14336
    },
    {
      "completion_length": 241.95536994934082,
      "epoch": 2.404040404040404,
      "grad_norm": 0.1600356620901426,
      "kl": 0.13165283203125,
      "learning_rate": 4.924232803183044e-07,
      "loss": 0.0001,
      "reward": 1.762500062584877,
      "reward_std": 0.0328299580141902,
      "rewards/equation_reward_func": 0.7669643200933933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14338
    },
    {
      "completion_length": 250.74108219146729,
      "epoch": 2.4043757072802716,
      "grad_norm": 0.24163101976938872,
      "kl": 0.1605224609375,
      "learning_rate": 4.924203491626544e-07,
      "loss": 0.0002,
      "reward": 1.7589286267757416,
      "reward_std": 0.07828682102262974,
      "rewards/equation_reward_func": 0.7723214514553547,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 14340
    },
    {
      "completion_length": 239.1517972946167,
      "epoch": 2.4047110105201392,
      "grad_norm": 0.2275927010975339,
      "kl": 0.12322998046875,
      "learning_rate": 4.924174174488628e-07,
      "loss": 0.0001,
      "reward": 1.7982143387198448,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8026785999536514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14342
    },
    {
      "completion_length": 230.55804538726807,
      "epoch": 2.405046313760007,
      "grad_norm": 0.18734378399333992,
      "kl": 0.105255126953125,
      "learning_rate": 4.924144851769365e-07,
      "loss": 0.0001,
      "reward": 1.7732143551111221,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7776786051690578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14344
    },
    {
      "completion_length": 238.5267972946167,
      "epoch": 2.405381616999874,
      "grad_norm": 0.18580847734440029,
      "kl": 0.16204833984375,
      "learning_rate": 4.924115523468821e-07,
      "loss": 0.0002,
      "reward": 1.716071493923664,
      "reward_std": 0.05808377172797918,
      "rewards/equation_reward_func": 0.7294643204659224,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 14346
    },
    {
      "completion_length": 231.34375953674316,
      "epoch": 2.4057169202397417,
      "grad_norm": 0.19578708489691765,
      "kl": 0.121368408203125,
      "learning_rate": 4.924086189587065e-07,
      "loss": 0.0001,
      "reward": 1.7910714969038963,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7955357283353806,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14348
    },
    {
      "completion_length": 231.26786994934082,
      "epoch": 2.4060522234796093,
      "grad_norm": 0.19892274262116755,
      "kl": 0.140625,
      "learning_rate": 4.924056850124164e-07,
      "loss": 0.0001,
      "reward": 1.719642922282219,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7241071686148643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14350
    },
    {
      "completion_length": 241.85715579986572,
      "epoch": 2.406387526719477,
      "grad_norm": 0.1743829897833669,
      "kl": 0.119781494140625,
      "learning_rate": 4.924027505080185e-07,
      "loss": 0.0001,
      "reward": 1.7857143357396126,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143245637417,
      "rewards/format_reward_func": 1.0,
      "step": 14352
    },
    {
      "completion_length": 239.40626049041748,
      "epoch": 2.4067228299593446,
      "grad_norm": 0.39051117823511666,
      "kl": 0.1546630859375,
      "learning_rate": 4.923998154455195e-07,
      "loss": 0.0002,
      "reward": 1.7732143104076385,
      "reward_std": 0.06818529777228832,
      "rewards/equation_reward_func": 0.7955357357859612,
      "rewards/format_reward_func": 0.977678582072258,
      "step": 14354
    },
    {
      "completion_length": 240.5580472946167,
      "epoch": 2.407058133199212,
      "grad_norm": 0.003754627012270124,
      "kl": 0.13525390625,
      "learning_rate": 4.923968798249264e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 14356
    },
    {
      "completion_length": 234.4821538925171,
      "epoch": 2.4073934364390794,
      "grad_norm": 0.1461006787847886,
      "kl": 0.112762451171875,
      "learning_rate": 4.923939436462457e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 14358
    },
    {
      "completion_length": 248.38393878936768,
      "epoch": 2.407728739678947,
      "grad_norm": 0.24628964012318919,
      "kl": 0.124847412109375,
      "learning_rate": 4.923910069094843e-07,
      "loss": 0.0001,
      "reward": 1.7504465207457542,
      "reward_std": 0.04987628059461713,
      "rewards/equation_reward_func": 0.7562500201165676,
      "rewards/format_reward_func": 0.9941964335739613,
      "step": 14360
    },
    {
      "completion_length": 235.34376049041748,
      "epoch": 2.4080640429188147,
      "grad_norm": 0.2245625890808432,
      "kl": 0.107025146484375,
      "learning_rate": 4.92388069614649e-07,
      "loss": 0.0001,
      "reward": 1.7107143700122833,
      "reward_std": 0.05555838905274868,
      "rewards/equation_reward_func": 0.7196428887546062,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14362
    },
    {
      "completion_length": 249.2187623977661,
      "epoch": 2.4083993461586823,
      "grad_norm": 0.17260432031337294,
      "kl": 0.115936279296875,
      "learning_rate": 4.923851317617464e-07,
      "loss": 0.0001,
      "reward": 1.751785784959793,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7562500350177288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14364
    },
    {
      "completion_length": 244.25447463989258,
      "epoch": 2.40873464939855,
      "grad_norm": 0.2208763728434164,
      "kl": 0.113006591796875,
      "learning_rate": 4.923821933507834e-07,
      "loss": 0.0001,
      "reward": 1.796428643167019,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.796428594738245,
      "rewards/format_reward_func": 1.0,
      "step": 14366
    },
    {
      "completion_length": 245.88394165039062,
      "epoch": 2.4090699526384176,
      "grad_norm": 0.23033703151636636,
      "kl": 0.10723876953125,
      "learning_rate": 4.923792543817667e-07,
      "loss": 0.0001,
      "reward": 1.7785715013742447,
      "reward_std": 0.04040610138326883,
      "rewards/equation_reward_func": 0.7875000312924385,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14368
    },
    {
      "completion_length": 236.25894165039062,
      "epoch": 2.409405255878285,
      "grad_norm": 0.2851470594450002,
      "kl": 0.143096923828125,
      "learning_rate": 4.923763148547031e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 14370
    },
    {
      "completion_length": 235.01786708831787,
      "epoch": 2.4097405591181524,
      "grad_norm": 0.1443877591575158,
      "kl": 0.10845947265625,
      "learning_rate": 4.923733747695994e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428999304771,
      "rewards/format_reward_func": 1.0,
      "step": 14372
    },
    {
      "completion_length": 236.65179538726807,
      "epoch": 2.41007586235802,
      "grad_norm": 0.0027849364925823456,
      "kl": 0.105072021484375,
      "learning_rate": 4.923704341264623e-07,
      "loss": 0.0001,
      "reward": 1.776785783469677,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7812500335276127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14374
    },
    {
      "completion_length": 236.2455472946167,
      "epoch": 2.4104111655978877,
      "grad_norm": 0.2298392056406199,
      "kl": 0.108001708984375,
      "learning_rate": 4.923674929252985e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714581608772,
      "rewards/format_reward_func": 1.0,
      "step": 14376
    },
    {
      "completion_length": 234.21876049041748,
      "epoch": 2.410746468837755,
      "grad_norm": 0.1903379262753221,
      "kl": 0.114990234375,
      "learning_rate": 4.92364551166115e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.803571455180645,
      "rewards/format_reward_func": 1.0,
      "step": 14378
    },
    {
      "completion_length": 238.33036613464355,
      "epoch": 2.4110817720776225,
      "grad_norm": 0.2296659492073621,
      "kl": 0.106414794921875,
      "learning_rate": 4.923616088489183e-07,
      "loss": 0.0001,
      "reward": 1.787500061094761,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7919643074274063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14380
    },
    {
      "completion_length": 240.4509048461914,
      "epoch": 2.41141707531749,
      "grad_norm": 0.1183545544888929,
      "kl": 0.10882568359375,
      "learning_rate": 4.923586659737154e-07,
      "loss": 0.0001,
      "reward": 1.7696429044008255,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7741071879863739,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14382
    },
    {
      "completion_length": 241.94197463989258,
      "epoch": 2.411752378557358,
      "grad_norm": 0.2451140468106907,
      "kl": 0.1093902587890625,
      "learning_rate": 4.92355722540513e-07,
      "loss": 0.0001,
      "reward": 1.7875000536441803,
      "reward_std": 0.06818529590964317,
      "rewards/equation_reward_func": 0.7919643186032772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14384
    },
    {
      "completion_length": 231.56251049041748,
      "epoch": 2.4120876817972254,
      "grad_norm": 0.15612636244581551,
      "kl": 0.103759765625,
      "learning_rate": 4.923527785493179e-07,
      "loss": 0.0001,
      "reward": 1.8178571835160255,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8178571686148643,
      "rewards/format_reward_func": 1.0,
      "step": 14386
    },
    {
      "completion_length": 237.1205472946167,
      "epoch": 2.412422985037093,
      "grad_norm": 0.22794551959835807,
      "kl": 0.105438232421875,
      "learning_rate": 4.923498340001369e-07,
      "loss": 0.0001,
      "reward": 1.7839286103844643,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7883928939700127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14388
    },
    {
      "completion_length": 241.57590675354004,
      "epoch": 2.4127582882769607,
      "grad_norm": 0.21870437344172014,
      "kl": 0.112030029296875,
      "learning_rate": 4.923468888929766e-07,
      "loss": 0.0001,
      "reward": 1.7321429401636124,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428842842579,
      "rewards/format_reward_func": 1.0,
      "step": 14390
    },
    {
      "completion_length": 237.82143878936768,
      "epoch": 2.413093591516828,
      "grad_norm": 0.3153485246471543,
      "kl": 0.113677978515625,
      "learning_rate": 4.923439432278439e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 14392
    },
    {
      "completion_length": 240.8303680419922,
      "epoch": 2.4134288947566955,
      "grad_norm": 0.1720802940204395,
      "kl": 0.1099853515625,
      "learning_rate": 4.923409970047456e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571846336126,
      "rewards/format_reward_func": 1.0,
      "step": 14394
    },
    {
      "completion_length": 242.5312614440918,
      "epoch": 2.413764197996563,
      "grad_norm": 0.2575036019942336,
      "kl": 0.109466552734375,
      "learning_rate": 4.923380502236885e-07,
      "loss": 0.0001,
      "reward": 1.7625000774860382,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14396
    },
    {
      "completion_length": 249.15179824829102,
      "epoch": 2.414099501236431,
      "grad_norm": 0.12303614393479674,
      "kl": 0.11273193359375,
      "learning_rate": 4.923351028846794e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000342726707,
      "rewards/format_reward_func": 1.0,
      "step": 14398
    },
    {
      "completion_length": 237.24554634094238,
      "epoch": 2.4144348044762984,
      "grad_norm": 0.22113499831589165,
      "kl": 0.114288330078125,
      "learning_rate": 4.92332154987725e-07,
      "loss": 0.0001,
      "reward": 1.7803572416305542,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.784821443259716,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14400
    },
    {
      "completion_length": 246.93304920196533,
      "epoch": 2.4147701077161656,
      "grad_norm": 0.34530621108268905,
      "kl": 0.130767822265625,
      "learning_rate": 4.923292065328321e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 14402
    },
    {
      "completion_length": 253.29018878936768,
      "epoch": 2.4151054109560333,
      "grad_norm": 0.22128352179118405,
      "kl": 0.125274658203125,
      "learning_rate": 4.923262575200076e-07,
      "loss": 0.0001,
      "reward": 1.7946428805589676,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.7991071753203869,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14404
    },
    {
      "completion_length": 251.0848331451416,
      "epoch": 2.415440714195901,
      "grad_norm": 0.06485660401128394,
      "kl": 0.122039794921875,
      "learning_rate": 4.923233079492582e-07,
      "loss": 0.0001,
      "reward": 1.758928619325161,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7633928693830967,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14406
    },
    {
      "completion_length": 249.37501430511475,
      "epoch": 2.4157760174357685,
      "grad_norm": 0.11814345401091123,
      "kl": 0.10894775390625,
      "learning_rate": 4.923203578205906e-07,
      "loss": 0.0001,
      "reward": 1.792857214808464,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 14408
    },
    {
      "completion_length": 246.2678689956665,
      "epoch": 2.416111320675636,
      "grad_norm": 0.044181690154128005,
      "kl": 0.11151123046875,
      "learning_rate": 4.923174071340117e-07,
      "loss": 0.0001,
      "reward": 1.7892857640981674,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7892857566475868,
      "rewards/format_reward_func": 1.0,
      "step": 14410
    },
    {
      "completion_length": 254.5000123977661,
      "epoch": 2.416446623915504,
      "grad_norm": 0.12192966153626056,
      "kl": 0.104949951171875,
      "learning_rate": 4.923144558895283e-07,
      "loss": 0.0001,
      "reward": 1.805357187986374,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8098214603960514,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14412
    },
    {
      "completion_length": 250.5625123977661,
      "epoch": 2.416781927155371,
      "grad_norm": 0.18429756344803158,
      "kl": 0.102203369140625,
      "learning_rate": 4.923115040871472e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 14414
    },
    {
      "completion_length": 249.3169755935669,
      "epoch": 2.4171172303952386,
      "grad_norm": 0.15038284538850447,
      "kl": 0.105133056640625,
      "learning_rate": 4.923085517268752e-07,
      "loss": 0.0001,
      "reward": 1.805357187986374,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.8187500275671482,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 14416
    },
    {
      "completion_length": 254.67411994934082,
      "epoch": 2.4174525336351063,
      "grad_norm": 0.05553958299680993,
      "kl": 0.1148681640625,
      "learning_rate": 4.92305598808719e-07,
      "loss": 0.0001,
      "reward": 1.7785714715719223,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714752972126,
      "rewards/format_reward_func": 1.0,
      "step": 14418
    },
    {
      "completion_length": 251.93751049041748,
      "epoch": 2.417787836874974,
      "grad_norm": 0.219178777856766,
      "kl": 0.1243896484375,
      "learning_rate": 4.923026453326856e-07,
      "loss": 0.0001,
      "reward": 1.739285759627819,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.739285746589303,
      "rewards/format_reward_func": 1.0,
      "step": 14420
    },
    {
      "completion_length": 250.5267972946167,
      "epoch": 2.4181231401148415,
      "grad_norm": 0.21583952375718596,
      "kl": 0.101226806640625,
      "learning_rate": 4.922996912987815e-07,
      "loss": 0.0001,
      "reward": 1.7857143580913544,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143171131611,
      "rewards/format_reward_func": 1.0,
      "step": 14422
    },
    {
      "completion_length": 252.9821548461914,
      "epoch": 2.4184584433547087,
      "grad_norm": 0.14685803789461777,
      "kl": 0.110504150390625,
      "learning_rate": 4.922967367070139e-07,
      "loss": 0.0001,
      "reward": 1.7428572177886963,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7428571619093418,
      "rewards/format_reward_func": 1.0,
      "step": 14424
    },
    {
      "completion_length": 250.14286613464355,
      "epoch": 2.4187937465945764,
      "grad_norm": 0.19287115760743048,
      "kl": 0.1021728515625,
      "learning_rate": 4.922937815573892e-07,
      "loss": 0.0001,
      "reward": 1.7232143580913544,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7276786081492901,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14426
    },
    {
      "completion_length": 253.7053689956665,
      "epoch": 2.419129049834444,
      "grad_norm": 0.17641124181431558,
      "kl": 0.116424560546875,
      "learning_rate": 4.922908258499144e-07,
      "loss": 0.0001,
      "reward": 1.74642863124609,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7464286033064127,
      "rewards/format_reward_func": 1.0,
      "step": 14428
    },
    {
      "completion_length": 246.64733505249023,
      "epoch": 2.4194643530743116,
      "grad_norm": 0.24730402074427776,
      "kl": 0.0968017578125,
      "learning_rate": 4.922878695845964e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.07071067858487368,
      "rewards/equation_reward_func": 0.7589286118745804,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14430
    },
    {
      "completion_length": 242.6205472946167,
      "epoch": 2.4197996563141793,
      "grad_norm": 0.1609843188328104,
      "kl": 0.107147216796875,
      "learning_rate": 4.922849127614417e-07,
      "loss": 0.0001,
      "reward": 1.8071429058909416,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8071428872644901,
      "rewards/format_reward_func": 1.0,
      "step": 14432
    },
    {
      "completion_length": 246.95090293884277,
      "epoch": 2.420134959554047,
      "grad_norm": 0.08023298552883207,
      "kl": 0.1016845703125,
      "learning_rate": 4.922819553804575e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143081724644,
      "rewards/format_reward_func": 1.0,
      "step": 14434
    },
    {
      "completion_length": 258.5803689956665,
      "epoch": 2.420470262793914,
      "grad_norm": 0.1708562641566658,
      "kl": 0.09808349609375,
      "learning_rate": 4.922789974416503e-07,
      "loss": 0.0001,
      "reward": 1.805357187986374,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8098214417695999,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14436
    },
    {
      "completion_length": 253.1294755935669,
      "epoch": 2.4208055660337817,
      "grad_norm": 0.1869148624482686,
      "kl": 0.091583251953125,
      "learning_rate": 4.92276038945027e-07,
      "loss": 0.0001,
      "reward": 1.7892857864499092,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857529222965,
      "rewards/format_reward_func": 1.0,
      "step": 14438
    },
    {
      "completion_length": 245.75893878936768,
      "epoch": 2.4211408692736494,
      "grad_norm": 0.23055645746875422,
      "kl": 0.1028900146484375,
      "learning_rate": 4.922730798905944e-07,
      "loss": 0.0001,
      "reward": 1.732142947614193,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321428898721933,
      "rewards/format_reward_func": 1.0,
      "step": 14440
    },
    {
      "completion_length": 245.9330472946167,
      "epoch": 2.421476172513517,
      "grad_norm": 0.2640415440209392,
      "kl": 0.10809326171875,
      "learning_rate": 4.922701202783593e-07,
      "loss": 0.0001,
      "reward": 1.8107143491506577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 14442
    },
    {
      "completion_length": 249.7053689956665,
      "epoch": 2.4218114757533846,
      "grad_norm": 0.16123161749216006,
      "kl": 0.100982666015625,
      "learning_rate": 4.922671601083287e-07,
      "loss": 0.0001,
      "reward": 1.7857143580913544,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143208384514,
      "rewards/format_reward_func": 1.0,
      "step": 14444
    },
    {
      "completion_length": 243.96429443359375,
      "epoch": 2.422146778993252,
      "grad_norm": 0.19344738804761819,
      "kl": 0.101409912109375,
      "learning_rate": 4.922641993805092e-07,
      "loss": 0.0001,
      "reward": 1.7214286550879478,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7214285954833031,
      "rewards/format_reward_func": 1.0,
      "step": 14446
    },
    {
      "completion_length": 243.48662090301514,
      "epoch": 2.4224820822331194,
      "grad_norm": 0.1864402544792117,
      "kl": 0.108001708984375,
      "learning_rate": 4.922612380949077e-07,
      "loss": 0.0001,
      "reward": 1.800000049173832,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 14448
    },
    {
      "completion_length": 254.58483123779297,
      "epoch": 2.422817385472987,
      "grad_norm": 0.18771169862166331,
      "kl": 0.095550537109375,
      "learning_rate": 4.922582762515311e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143193483353,
      "rewards/format_reward_func": 1.0,
      "step": 14450
    },
    {
      "completion_length": 256.90626335144043,
      "epoch": 2.4231526887128547,
      "grad_norm": 0.152067826141475,
      "kl": 0.112213134765625,
      "learning_rate": 4.92255313850386e-07,
      "loss": 0.0001,
      "reward": 1.7500000447034836,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 14452
    },
    {
      "completion_length": 244.7232255935669,
      "epoch": 2.4234879919527224,
      "grad_norm": 0.15337086305242245,
      "kl": 0.1212158203125,
      "learning_rate": 4.922523508914794e-07,
      "loss": 0.0001,
      "reward": 1.7321429401636124,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321428693830967,
      "rewards/format_reward_func": 1.0,
      "step": 14454
    },
    {
      "completion_length": 249.39733123779297,
      "epoch": 2.42382329519259,
      "grad_norm": 0.19277169712502537,
      "kl": 0.10882568359375,
      "learning_rate": 4.92249387374818e-07,
      "loss": 0.0001,
      "reward": 1.7285715118050575,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7285714447498322,
      "rewards/format_reward_func": 1.0,
      "step": 14456
    },
    {
      "completion_length": 239.508939743042,
      "epoch": 2.424158598432457,
      "grad_norm": 0.23544722794422707,
      "kl": 0.1253662109375,
      "learning_rate": 4.922464233004088e-07,
      "loss": 0.0001,
      "reward": 1.8000000789761543,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000081956387,
      "rewards/format_reward_func": 1.0,
      "step": 14458
    },
    {
      "completion_length": 246.9285831451416,
      "epoch": 2.424493901672325,
      "grad_norm": 0.16456192966586017,
      "kl": 0.104736328125,
      "learning_rate": 4.922434586682584e-07,
      "loss": 0.0001,
      "reward": 1.7607143372297287,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7607142981141806,
      "rewards/format_reward_func": 1.0,
      "step": 14460
    },
    {
      "completion_length": 252.95983219146729,
      "epoch": 2.4248292049121924,
      "grad_norm": 0.22480382634898416,
      "kl": 0.097076416015625,
      "learning_rate": 4.922404934783738e-07,
      "loss": 0.0001,
      "reward": 1.7321429252624512,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7321428898721933,
      "rewards/format_reward_func": 1.0,
      "step": 14462
    },
    {
      "completion_length": 250.0044755935669,
      "epoch": 2.42516450815206,
      "grad_norm": 0.11925795390631981,
      "kl": 0.1038055419921875,
      "learning_rate": 4.922375277307618e-07,
      "loss": 0.0001,
      "reward": 1.7553572207689285,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7598214522004128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14464
    },
    {
      "completion_length": 234.99554538726807,
      "epoch": 2.4254998113919277,
      "grad_norm": 0.15510926176807355,
      "kl": 0.11212158203125,
      "learning_rate": 4.922345614254292e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571827709675,
      "rewards/format_reward_func": 1.0,
      "step": 14466
    },
    {
      "completion_length": 239.2634048461914,
      "epoch": 2.425835114631795,
      "grad_norm": 0.26056520083745166,
      "kl": 0.114044189453125,
      "learning_rate": 4.922315945623827e-07,
      "loss": 0.0001,
      "reward": 1.7500000894069672,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000335276127,
      "rewards/format_reward_func": 1.0,
      "step": 14468
    },
    {
      "completion_length": 241.0669755935669,
      "epoch": 2.4261704178716625,
      "grad_norm": 0.1362923694488133,
      "kl": 0.10235595703125,
      "learning_rate": 4.922286271416294e-07,
      "loss": 0.0001,
      "reward": 1.7892857566475868,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 14470
    },
    {
      "completion_length": 245.40626335144043,
      "epoch": 2.42650572111153,
      "grad_norm": 0.1893996972953304,
      "kl": 0.101776123046875,
      "learning_rate": 4.92225659163176e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 14472
    },
    {
      "completion_length": 239.88393878936768,
      "epoch": 2.426841024351398,
      "grad_norm": 0.2687254884525047,
      "kl": 0.107421875,
      "learning_rate": 4.922226906270292e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7750000357627869,
      "rewards/format_reward_func": 1.0,
      "step": 14474
    },
    {
      "completion_length": 240.14733219146729,
      "epoch": 2.4271763275912654,
      "grad_norm": 0.192068117999007,
      "kl": 0.109466552734375,
      "learning_rate": 4.92219721533196e-07,
      "loss": 0.0001,
      "reward": 1.8178571835160255,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8178571686148643,
      "rewards/format_reward_func": 1.0,
      "step": 14476
    },
    {
      "completion_length": 241.20983219146729,
      "epoch": 2.427511630831133,
      "grad_norm": 0.26444603268357253,
      "kl": 0.10882568359375,
      "learning_rate": 4.922167518816832e-07,
      "loss": 0.0001,
      "reward": 1.8000000640749931,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8000000193715096,
      "rewards/format_reward_func": 1.0,
      "step": 14478
    },
    {
      "completion_length": 239.41072368621826,
      "epoch": 2.4278469340710003,
      "grad_norm": 0.09661934193422012,
      "kl": 0.1143798828125,
      "learning_rate": 4.922137816724976e-07,
      "loss": 0.0001,
      "reward": 1.7303572073578835,
      "reward_std": 0.017677669413387775,
      "rewards/equation_reward_func": 0.7348214648663998,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14480
    },
    {
      "completion_length": 249.3214406967163,
      "epoch": 2.428182237310868,
      "grad_norm": 0.11849886046639548,
      "kl": 0.1033935546875,
      "learning_rate": 4.922108109056461e-07,
      "loss": 0.0001,
      "reward": 1.769642896950245,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7741071619093418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14482
    },
    {
      "completion_length": 241.2544765472412,
      "epoch": 2.4285175405507355,
      "grad_norm": 0.19818905557945873,
      "kl": 0.1063232421875,
      "learning_rate": 4.922078395811355e-07,
      "loss": 0.0001,
      "reward": 1.7928571924567223,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571738302708,
      "rewards/format_reward_func": 1.0,
      "step": 14484
    },
    {
      "completion_length": 257.62501430511475,
      "epoch": 2.428852843790603,
      "grad_norm": 0.16333273125407996,
      "kl": 0.12158203125,
      "learning_rate": 4.922048676989728e-07,
      "loss": 0.0001,
      "reward": 1.691071517765522,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.6955357491970062,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14486
    },
    {
      "completion_length": 244.50894260406494,
      "epoch": 2.429188147030471,
      "grad_norm": 0.06860305031557319,
      "kl": 0.102630615234375,
      "learning_rate": 4.922018952591645e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7535714656114578,
      "rewards/format_reward_func": 1.0,
      "step": 14488
    },
    {
      "completion_length": 255.4285831451416,
      "epoch": 2.429523450270338,
      "grad_norm": 0.18951155606865122,
      "kl": 0.11529541015625,
      "learning_rate": 4.921989222617177e-07,
      "loss": 0.0001,
      "reward": 1.7714286372065544,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7714285999536514,
      "rewards/format_reward_func": 1.0,
      "step": 14490
    },
    {
      "completion_length": 243.508939743042,
      "epoch": 2.4298587535102056,
      "grad_norm": 0.06729385120696377,
      "kl": 0.10260009765625,
      "learning_rate": 4.921959487066392e-07,
      "loss": 0.0001,
      "reward": 1.7535715103149414,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714600235224,
      "rewards/format_reward_func": 1.0,
      "step": 14492
    },
    {
      "completion_length": 243.26340198516846,
      "epoch": 2.4301940567500733,
      "grad_norm": 0.21808525209582078,
      "kl": 0.1015625,
      "learning_rate": 4.921929745939357e-07,
      "loss": 0.0001,
      "reward": 1.8214286267757416,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8214286044239998,
      "rewards/format_reward_func": 1.0,
      "step": 14494
    },
    {
      "completion_length": 242.9866189956665,
      "epoch": 2.430529359989941,
      "grad_norm": 0.10459009473345975,
      "kl": 0.10888671875,
      "learning_rate": 4.921899999236143e-07,
      "loss": 0.0001,
      "reward": 1.7535714954137802,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 14496
    },
    {
      "completion_length": 243.4062614440918,
      "epoch": 2.4308646632298085,
      "grad_norm": 0.20468565131128358,
      "kl": 0.09014892578125,
      "learning_rate": 4.921870246956817e-07,
      "loss": 0.0001,
      "reward": 1.7857143729925156,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 14498
    },
    {
      "completion_length": 240.4196548461914,
      "epoch": 2.431199966469676,
      "grad_norm": 0.2839771802697649,
      "kl": 0.102447509765625,
      "learning_rate": 4.921840489101447e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571827709675,
      "rewards/format_reward_func": 1.0,
      "step": 14500
    },
    {
      "completion_length": 243.91965293884277,
      "epoch": 2.431535269709544,
      "grad_norm": 0.13837876458820203,
      "kl": 0.09967041015625,
      "learning_rate": 4.921810725670104e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7964285984635353,
      "rewards/format_reward_func": 1.0,
      "step": 14502
    },
    {
      "completion_length": 248.0759048461914,
      "epoch": 2.431870572949411,
      "grad_norm": 0.0625324599377449,
      "kl": 0.10888671875,
      "learning_rate": 4.921780956662853e-07,
      "loss": 0.0001,
      "reward": 1.7357143461704254,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7357143275439739,
      "rewards/format_reward_func": 1.0,
      "step": 14504
    },
    {
      "completion_length": 240.58929824829102,
      "epoch": 2.4322058761892786,
      "grad_norm": 0.14144479185833814,
      "kl": 0.116973876953125,
      "learning_rate": 4.921751182079765e-07,
      "loss": 0.0001,
      "reward": 1.7750000655651093,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 14506
    },
    {
      "completion_length": 238.99554920196533,
      "epoch": 2.4325411794291463,
      "grad_norm": 0.12906166659910961,
      "kl": 0.115478515625,
      "learning_rate": 4.921721401920907e-07,
      "loss": 0.0001,
      "reward": 1.7464286461472511,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464286051690578,
      "rewards/format_reward_func": 1.0,
      "step": 14508
    },
    {
      "completion_length": 244.4062614440918,
      "epoch": 2.432876482669014,
      "grad_norm": 0.19870262101296146,
      "kl": 0.1026611328125,
      "learning_rate": 4.921691616186349e-07,
      "loss": 0.0001,
      "reward": 1.7892857491970062,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 14510
    },
    {
      "completion_length": 244.85269165039062,
      "epoch": 2.433211785908881,
      "grad_norm": 0.34484337934409914,
      "kl": 0.0943756103515625,
      "learning_rate": 4.921661824876161e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143022119999,
      "rewards/format_reward_func": 1.0,
      "step": 14512
    },
    {
      "completion_length": 238.5848331451416,
      "epoch": 2.4335470891487487,
      "grad_norm": 0.16552693092879947,
      "kl": 0.10479736328125,
      "learning_rate": 4.921632027990408e-07,
      "loss": 0.0001,
      "reward": 1.7625000551342964,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7669643238186836,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14514
    },
    {
      "completion_length": 233.008939743042,
      "epoch": 2.4338823923886164,
      "grad_norm": 0.23178198865499708,
      "kl": 0.109375,
      "learning_rate": 4.921602225529159e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 14516
    },
    {
      "completion_length": 241.25447940826416,
      "epoch": 2.434217695628484,
      "grad_norm": 0.23617625683615337,
      "kl": 0.1003265380859375,
      "learning_rate": 4.921572417492485e-07,
      "loss": 0.0001,
      "reward": 1.7500000968575478,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000260770321,
      "rewards/format_reward_func": 1.0,
      "step": 14518
    },
    {
      "completion_length": 236.50447463989258,
      "epoch": 2.4345529988683516,
      "grad_norm": 0.09500009038811856,
      "kl": 0.108428955078125,
      "learning_rate": 4.921542603880453e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000223517418,
      "rewards/format_reward_func": 1.0,
      "step": 14520
    },
    {
      "completion_length": 241.37947368621826,
      "epoch": 2.4348883021082193,
      "grad_norm": 0.10477360638602254,
      "kl": 0.10235595703125,
      "learning_rate": 4.921512784693132e-07,
      "loss": 0.0001,
      "reward": 1.8017857521772385,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8062500394880772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14522
    },
    {
      "completion_length": 244.4375123977661,
      "epoch": 2.435223605348087,
      "grad_norm": 0.5641491334390114,
      "kl": 0.11566162109375,
      "learning_rate": 4.921482959930592e-07,
      "loss": 0.0001,
      "reward": 1.7464286461472511,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7464286126196384,
      "rewards/format_reward_func": 1.0,
      "step": 14524
    },
    {
      "completion_length": 239.9107265472412,
      "epoch": 2.435558908587954,
      "grad_norm": 0.18693819338494735,
      "kl": 0.114654541015625,
      "learning_rate": 4.9214531295929e-07,
      "loss": 0.0001,
      "reward": 1.741071492433548,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7455357313156128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14526
    },
    {
      "completion_length": 233.2009048461914,
      "epoch": 2.4358942118278217,
      "grad_norm": 0.20661760679939017,
      "kl": 0.104034423828125,
      "learning_rate": 4.921423293680124e-07,
      "loss": 0.0001,
      "reward": 1.8250000551342964,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.825000025331974,
      "rewards/format_reward_func": 1.0,
      "step": 14528
    },
    {
      "completion_length": 230.52679634094238,
      "epoch": 2.4362295150676894,
      "grad_norm": 0.10369726032449242,
      "kl": 0.112823486328125,
      "learning_rate": 4.921393452192334e-07,
      "loss": 0.0001,
      "reward": 1.8035714700818062,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8035714626312256,
      "rewards/format_reward_func": 1.0,
      "step": 14530
    },
    {
      "completion_length": 240.18751049041748,
      "epoch": 2.436564818307557,
      "grad_norm": 0.3213602942407995,
      "kl": 0.10504150390625,
      "learning_rate": 4.9213636051296e-07,
      "loss": 0.0001,
      "reward": 1.7785714715719223,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.778571467846632,
      "rewards/format_reward_func": 1.0,
      "step": 14532
    },
    {
      "completion_length": 231.95983219146729,
      "epoch": 2.4369001215474246,
      "grad_norm": 0.2630550350345542,
      "kl": 0.102081298828125,
      "learning_rate": 4.921333752491987e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428924798965,
      "rewards/format_reward_func": 1.0,
      "step": 14534
    },
    {
      "completion_length": 230.3482265472412,
      "epoch": 2.437235424787292,
      "grad_norm": 0.13355432159632558,
      "kl": 0.104248046875,
      "learning_rate": 4.921303894279568e-07,
      "loss": 0.0001,
      "reward": 1.810714341700077,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8107143118977547,
      "rewards/format_reward_func": 1.0,
      "step": 14536
    },
    {
      "completion_length": 242.2857265472412,
      "epoch": 2.4375707280271595,
      "grad_norm": 0.23392933237799146,
      "kl": 0.1086273193359375,
      "learning_rate": 4.921274030492408e-07,
      "loss": 0.0001,
      "reward": 1.7178572416305542,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7178571857511997,
      "rewards/format_reward_func": 1.0,
      "step": 14538
    },
    {
      "completion_length": 233.24108505249023,
      "epoch": 2.437906031267027,
      "grad_norm": 0.23157423824280435,
      "kl": 0.10504150390625,
      "learning_rate": 4.921244161130578e-07,
      "loss": 0.0001,
      "reward": 1.7500000819563866,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7589286100119352,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14540
    },
    {
      "completion_length": 237.29911518096924,
      "epoch": 2.4382413345068947,
      "grad_norm": 0.08473106919533094,
      "kl": 0.11676025390625,
      "learning_rate": 4.921214286194147e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571678698063,
      "rewards/format_reward_func": 1.0,
      "step": 14542
    },
    {
      "completion_length": 228.8571548461914,
      "epoch": 2.4385766377467624,
      "grad_norm": 0.1442209266300916,
      "kl": 0.10284423828125,
      "learning_rate": 4.921184405683182e-07,
      "loss": 0.0001,
      "reward": 1.789285771548748,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857454717159,
      "rewards/format_reward_func": 1.0,
      "step": 14544
    },
    {
      "completion_length": 227.5669755935669,
      "epoch": 2.43891194098663,
      "grad_norm": 0.0418952892430739,
      "kl": 0.0934600830078125,
      "learning_rate": 4.921154519597753e-07,
      "loss": 0.0001,
      "reward": 1.789285771548748,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.789285734295845,
      "rewards/format_reward_func": 1.0,
      "step": 14546
    },
    {
      "completion_length": 241.20983219146729,
      "epoch": 2.439247244226497,
      "grad_norm": 0.15438437741472294,
      "kl": 0.097076416015625,
      "learning_rate": 4.921124627937929e-07,
      "loss": 0.0001,
      "reward": 1.7946429178118706,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7991071678698063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14548
    },
    {
      "completion_length": 237.89733409881592,
      "epoch": 2.439582547466365,
      "grad_norm": 0.22889737678043695,
      "kl": 0.110015869140625,
      "learning_rate": 4.921094730703778e-07,
      "loss": 0.0001,
      "reward": 1.765178643167019,
      "reward_std": 0.03914341004565358,
      "rewards/equation_reward_func": 0.7669643200933933,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 14550
    },
    {
      "completion_length": 236.3214406967163,
      "epoch": 2.4399178507062325,
      "grad_norm": 0.09313547110453636,
      "kl": 0.1150360107421875,
      "learning_rate": 4.921064827895369e-07,
      "loss": 0.0001,
      "reward": 1.8178571686148643,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8178571555763483,
      "rewards/format_reward_func": 1.0,
      "step": 14552
    },
    {
      "completion_length": 237.19197750091553,
      "epoch": 2.4402531539461,
      "grad_norm": 0.10693348598193116,
      "kl": 0.093170166015625,
      "learning_rate": 4.921034919512772e-07,
      "loss": 0.0001,
      "reward": 1.7839286103844643,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7883928827941418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14554
    },
    {
      "completion_length": 239.31697368621826,
      "epoch": 2.4405884571859677,
      "grad_norm": 0.17416553763203155,
      "kl": 0.0992889404296875,
      "learning_rate": 4.921005005556055e-07,
      "loss": 0.0001,
      "reward": 1.7750000357627869,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7750000394880772,
      "rewards/format_reward_func": 1.0,
      "step": 14556
    },
    {
      "completion_length": 233.5312614440918,
      "epoch": 2.440923760425835,
      "grad_norm": 0.5400463217476548,
      "kl": 0.1147308349609375,
      "learning_rate": 4.920975086025286e-07,
      "loss": 0.0001,
      "reward": 1.8321429193019867,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8321428708732128,
      "rewards/format_reward_func": 1.0,
      "step": 14558
    },
    {
      "completion_length": 224.7678680419922,
      "epoch": 2.4412590636657026,
      "grad_norm": 0.246319043845516,
      "kl": 0.093414306640625,
      "learning_rate": 4.920945160920535e-07,
      "loss": 0.0001,
      "reward": 1.783928632736206,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7875000312924385,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 14560
    },
    {
      "completion_length": 239.2634048461914,
      "epoch": 2.44159436690557,
      "grad_norm": 0.1861008204380244,
      "kl": 0.1091461181640625,
      "learning_rate": 4.920915230241871e-07,
      "loss": 0.0001,
      "reward": 1.7535714730620384,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714618861675,
      "rewards/format_reward_func": 1.0,
      "step": 14562
    },
    {
      "completion_length": 232.36161994934082,
      "epoch": 2.441929670145438,
      "grad_norm": 0.22969570435893646,
      "kl": 0.108428955078125,
      "learning_rate": 4.920885293989362e-07,
      "loss": 0.0001,
      "reward": 1.7428571954369545,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 14564
    },
    {
      "completion_length": 236.60268783569336,
      "epoch": 2.4422649733853055,
      "grad_norm": 0.17641325910517738,
      "kl": 0.0992279052734375,
      "learning_rate": 4.920855352163077e-07,
      "loss": 0.0001,
      "reward": 1.8000000417232513,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.8000000230967999,
      "rewards/format_reward_func": 1.0,
      "step": 14566
    },
    {
      "completion_length": 243.73661708831787,
      "epoch": 2.442600276625173,
      "grad_norm": 0.2636670701693931,
      "kl": 0.107452392578125,
      "learning_rate": 4.920825404763086e-07,
      "loss": 0.0001,
      "reward": 1.7696429342031479,
      "reward_std": 0.07323605753481388,
      "rewards/equation_reward_func": 0.7741071619093418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14568
    },
    {
      "completion_length": 235.40625858306885,
      "epoch": 2.4429355798650403,
      "grad_norm": 0.20063605323852532,
      "kl": 0.106597900390625,
      "learning_rate": 4.920795451789458e-07,
      "loss": 0.0001,
      "reward": 1.7803572192788124,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7848214507102966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14570
    },
    {
      "completion_length": 237.8928689956665,
      "epoch": 2.443270883104908,
      "grad_norm": 0.2844660939043713,
      "kl": 0.105743408203125,
      "learning_rate": 4.920765493242261e-07,
      "loss": 0.0001,
      "reward": 1.7839286401867867,
      "reward_std": 0.06313453335314989,
      "rewards/equation_reward_func": 0.7883928790688515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14572
    },
    {
      "completion_length": 237.571439743042,
      "epoch": 2.4436061863447756,
      "grad_norm": 0.14908936651878493,
      "kl": 0.09649658203125,
      "learning_rate": 4.920735529121563e-07,
      "loss": 0.0001,
      "reward": 1.7910714596509933,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.8044643010944128,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 14574
    },
    {
      "completion_length": 240.28125953674316,
      "epoch": 2.443941489584643,
      "grad_norm": 0.1813296833370229,
      "kl": 0.111358642578125,
      "learning_rate": 4.920705559427436e-07,
      "loss": 0.0001,
      "reward": 1.7339286282658577,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.738392885774374,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14576
    },
    {
      "completion_length": 245.95983219146729,
      "epoch": 2.444276792824511,
      "grad_norm": 0.08269867957004332,
      "kl": 0.124542236328125,
      "learning_rate": 4.920675584159947e-07,
      "loss": 0.0001,
      "reward": 1.7500000521540642,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 14578
    },
    {
      "completion_length": 247.1294765472412,
      "epoch": 2.444612096064378,
      "grad_norm": 0.2082926494731955,
      "kl": 0.1058349609375,
      "learning_rate": 4.920645603319164e-07,
      "loss": 0.0001,
      "reward": 1.7428571954369545,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571581840515,
      "rewards/format_reward_func": 1.0,
      "step": 14580
    },
    {
      "completion_length": 238.02233409881592,
      "epoch": 2.4449473993042456,
      "grad_norm": 0.25334244067411815,
      "kl": 0.1026611328125,
      "learning_rate": 4.920615616905158e-07,
      "loss": 0.0001,
      "reward": 1.8089286237955093,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.8133928701281548,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14582
    },
    {
      "completion_length": 244.36608219146729,
      "epoch": 2.4452827025441133,
      "grad_norm": 0.22762717588504205,
      "kl": 0.1002197265625,
      "learning_rate": 4.920585624917998e-07,
      "loss": 0.0001,
      "reward": 1.8178571984171867,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.8267857246100903,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14584
    },
    {
      "completion_length": 238.6741189956665,
      "epoch": 2.445618005783981,
      "grad_norm": 0.1654778832690156,
      "kl": 0.0996856689453125,
      "learning_rate": 4.920555627357752e-07,
      "loss": 0.0001,
      "reward": 1.7678571864962578,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571734577417,
      "rewards/format_reward_func": 1.0,
      "step": 14586
    },
    {
      "completion_length": 243.37501049041748,
      "epoch": 2.4459533090238486,
      "grad_norm": 0.2654349418206511,
      "kl": 0.099334716796875,
      "learning_rate": 4.920525624224489e-07,
      "loss": 0.0001,
      "reward": 1.76607146859169,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7705357596278191,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14588
    },
    {
      "completion_length": 250.31697463989258,
      "epoch": 2.446288612263716,
      "grad_norm": 0.20491560862196873,
      "kl": 0.113525390625,
      "learning_rate": 4.920495615518279e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 14590
    },
    {
      "completion_length": 256.28126335144043,
      "epoch": 2.4466239155035834,
      "grad_norm": 0.0950820708130051,
      "kl": 0.118316650390625,
      "learning_rate": 4.92046560123919e-07,
      "loss": 0.0001,
      "reward": 1.8107143118977547,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8107143081724644,
      "rewards/format_reward_func": 1.0,
      "step": 14592
    },
    {
      "completion_length": 253.70536613464355,
      "epoch": 2.446959218743451,
      "grad_norm": 0.15718448474235885,
      "kl": 0.112945556640625,
      "learning_rate": 4.920435581387293e-07,
      "loss": 0.0001,
      "reward": 1.805357187986374,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.809821467846632,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14594
    },
    {
      "completion_length": 259.1651906967163,
      "epoch": 2.4472945219833186,
      "grad_norm": 0.09285379066929722,
      "kl": 0.122222900390625,
      "learning_rate": 4.920405555962654e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 14596
    },
    {
      "completion_length": 256.9509057998657,
      "epoch": 2.4476298252231863,
      "grad_norm": 0.10765725585666376,
      "kl": 0.12353515625,
      "learning_rate": 4.920375524965346e-07,
      "loss": 0.0001,
      "reward": 1.7482143491506577,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7526785954833031,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14598
    },
    {
      "completion_length": 254.65179634094238,
      "epoch": 2.447965128463054,
      "grad_norm": 0.09697412589677633,
      "kl": 0.116363525390625,
      "learning_rate": 4.920345488395436e-07,
      "loss": 0.0001,
      "reward": 1.744642935693264,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7491071745753288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14600
    },
    {
      "completion_length": 258.3660831451416,
      "epoch": 2.448300431702921,
      "grad_norm": 0.2110317592273284,
      "kl": 0.10479736328125,
      "learning_rate": 4.920315446252992e-07,
      "loss": 0.0001,
      "reward": 1.7214286550879478,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7392857484519482,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 14602
    },
    {
      "completion_length": 259.3482246398926,
      "epoch": 2.4486357349427887,
      "grad_norm": 0.16048095784784516,
      "kl": 0.11065673828125,
      "learning_rate": 4.920285398538085e-07,
      "loss": 0.0001,
      "reward": 1.6964286491274834,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.6964286062866449,
      "rewards/format_reward_func": 1.0,
      "step": 14604
    },
    {
      "completion_length": 265.58929347991943,
      "epoch": 2.4489710381826564,
      "grad_norm": 0.3574648813796875,
      "kl": 0.12420654296875,
      "learning_rate": 4.920255345250784e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.08081220183521509,
      "rewards/equation_reward_func": 0.7589286044239998,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14606
    },
    {
      "completion_length": 260.214298248291,
      "epoch": 2.449306341422524,
      "grad_norm": 0.22111108570420177,
      "kl": 0.118621826171875,
      "learning_rate": 4.920225286391157e-07,
      "loss": 0.0001,
      "reward": 1.740178644657135,
      "reward_std": 0.06439722329378128,
      "rewards/equation_reward_func": 0.7419643141329288,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 14608
    },
    {
      "completion_length": 254.24108123779297,
      "epoch": 2.4496416446623916,
      "grad_norm": 0.21975558457475472,
      "kl": 0.10028076171875,
      "learning_rate": 4.920195221959275e-07,
      "loss": 0.0001,
      "reward": 1.8000000640749931,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8000000193715096,
      "rewards/format_reward_func": 1.0,
      "step": 14610
    },
    {
      "completion_length": 258.1562623977661,
      "epoch": 2.4499769479022593,
      "grad_norm": 0.3037889689093855,
      "kl": 0.12811279296875,
      "learning_rate": 4.920165151955205e-07,
      "loss": 0.0001,
      "reward": 1.7446429431438446,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7491071708500385,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14612
    },
    {
      "completion_length": 251.09376430511475,
      "epoch": 2.4503122511421265,
      "grad_norm": 0.11148829311822392,
      "kl": 0.096405029296875,
      "learning_rate": 4.920135076379019e-07,
      "loss": 0.0001,
      "reward": 1.7446429058909416,
      "reward_std": 0.047982245683670044,
      "rewards/equation_reward_func": 0.7491071801632643,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14614
    },
    {
      "completion_length": 259.15626525878906,
      "epoch": 2.450647554381994,
      "grad_norm": 0.2130659164666184,
      "kl": 0.120697021484375,
      "learning_rate": 4.920104995230784e-07,
      "loss": 0.0001,
      "reward": 1.7910714969038963,
      "reward_std": 0.04293148312717676,
      "rewards/equation_reward_func": 0.7955357432365417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14616
    },
    {
      "completion_length": 253.92412185668945,
      "epoch": 2.4509828576218617,
      "grad_norm": 0.1594905819651956,
      "kl": 0.123779296875,
      "learning_rate": 4.92007490851057e-07,
      "loss": 0.0001,
      "reward": 1.725000075995922,
      "reward_std": 0.06565991509705782,
      "rewards/equation_reward_func": 0.733928594738245,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14618
    },
    {
      "completion_length": 258.1116180419922,
      "epoch": 2.4513181608617294,
      "grad_norm": 0.28001387833506153,
      "kl": 0.139801025390625,
      "learning_rate": 4.920044816218446e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.03535533882677555,
      "rewards/equation_reward_func": 0.7910714484751225,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14620
    },
    {
      "completion_length": 256.36608028411865,
      "epoch": 2.451653464101597,
      "grad_norm": 0.15087349096042746,
      "kl": 0.132415771484375,
      "learning_rate": 4.920014718354481e-07,
      "loss": 0.0001,
      "reward": 1.7517857924103737,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500238418579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14622
    },
    {
      "completion_length": 251.73661994934082,
      "epoch": 2.451988767341464,
      "grad_norm": 0.0777120217393265,
      "kl": 0.12347412109375,
      "learning_rate": 4.919984614918746e-07,
      "loss": 0.0001,
      "reward": 1.7642857879400253,
      "reward_std": 0.05050762742757797,
      "rewards/equation_reward_func": 0.7821428775787354,
      "rewards/format_reward_func": 0.9821428656578064,
      "step": 14624
    },
    {
      "completion_length": 250.46876335144043,
      "epoch": 2.452324070581332,
      "grad_norm": 0.061192203297932936,
      "kl": 0.110107421875,
      "learning_rate": 4.919954505911309e-07,
      "loss": 0.0001,
      "reward": 1.8107143342494965,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8107143193483353,
      "rewards/format_reward_func": 1.0,
      "step": 14626
    },
    {
      "completion_length": 264.4732255935669,
      "epoch": 2.4526593738211995,
      "grad_norm": 0.25103566878335815,
      "kl": 0.142913818359375,
      "learning_rate": 4.919924391332239e-07,
      "loss": 0.0001,
      "reward": 1.730357214808464,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7348214406520128,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14628
    },
    {
      "completion_length": 249.88393878936768,
      "epoch": 2.452994677061067,
      "grad_norm": 0.12985395819053508,
      "kl": 0.1202392578125,
      "learning_rate": 4.919894271181606e-07,
      "loss": 0.0001,
      "reward": 1.7892857640981674,
      "reward_std": 0.045456863939762115,
      "rewards/equation_reward_func": 0.798214316368103,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14630
    },
    {
      "completion_length": 257.48215675354004,
      "epoch": 2.4533299803009347,
      "grad_norm": 0.20780368372354519,
      "kl": 0.151947021484375,
      "learning_rate": 4.91986414545948e-07,
      "loss": 0.0002,
      "reward": 1.7464286386966705,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7464285977184772,
      "rewards/format_reward_func": 1.0,
      "step": 14632
    },
    {
      "completion_length": 256.6294765472412,
      "epoch": 2.4536652835408024,
      "grad_norm": 0.20687599287668856,
      "kl": 0.11669921875,
      "learning_rate": 4.919834014165929e-07,
      "loss": 0.0001,
      "reward": 1.807142935693264,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.8160714581608772,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14634
    },
    {
      "completion_length": 252.1294755935669,
      "epoch": 2.45400058678067,
      "grad_norm": 0.1576095082803586,
      "kl": 0.1048583984375,
      "learning_rate": 4.919803877301022e-07,
      "loss": 0.0001,
      "reward": 1.8017857819795609,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8062500208616257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14636
    },
    {
      "completion_length": 254.8571538925171,
      "epoch": 2.454335890020537,
      "grad_norm": 0.10649602704630118,
      "kl": 0.137786865234375,
      "learning_rate": 4.91977373486483e-07,
      "loss": 0.0001,
      "reward": 1.7500000596046448,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 14638
    },
    {
      "completion_length": 251.45090579986572,
      "epoch": 2.454671193260405,
      "grad_norm": 0.17848525693874534,
      "kl": 0.128631591796875,
      "learning_rate": 4.919743586857422e-07,
      "loss": 0.0001,
      "reward": 1.742857202887535,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7428571749478579,
      "rewards/format_reward_func": 1.0,
      "step": 14640
    },
    {
      "completion_length": 256.9285840988159,
      "epoch": 2.4550064965002725,
      "grad_norm": 0.1561715109998979,
      "kl": 0.1240692138671875,
      "learning_rate": 4.919713433278866e-07,
      "loss": 0.0001,
      "reward": 1.7553571835160255,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7598214596509933,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14642
    },
    {
      "completion_length": 250.8928680419922,
      "epoch": 2.45534179974014,
      "grad_norm": 0.09381988812620205,
      "kl": 0.139129638671875,
      "learning_rate": 4.919683274129234e-07,
      "loss": 0.0001,
      "reward": 1.8214286342263222,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8214285857975483,
      "rewards/format_reward_func": 1.0,
      "step": 14644
    },
    {
      "completion_length": 254.1785831451416,
      "epoch": 2.4556771029800073,
      "grad_norm": 0.1776446950481363,
      "kl": 0.1341552734375,
      "learning_rate": 4.919653109408593e-07,
      "loss": 0.0001,
      "reward": 1.776785746216774,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7812500260770321,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14646
    },
    {
      "completion_length": 251.08483600616455,
      "epoch": 2.456012406219875,
      "grad_norm": 0.16014605591091147,
      "kl": 0.126861572265625,
      "learning_rate": 4.919622939117013e-07,
      "loss": 0.0001,
      "reward": 1.7758929058909416,
      "reward_std": 0.06439722329378128,
      "rewards/equation_reward_func": 0.7776786088943481,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 14648
    },
    {
      "completion_length": 254.19197750091553,
      "epoch": 2.4563477094597426,
      "grad_norm": 0.0025472339599904464,
      "kl": 0.126617431640625,
      "learning_rate": 4.919592763254564e-07,
      "loss": 0.0001,
      "reward": 1.791071467101574,
      "reward_std": 0.022728432901203632,
      "rewards/equation_reward_func": 0.7955357506871223,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14650
    },
    {
      "completion_length": 250.99108409881592,
      "epoch": 2.45668301269961,
      "grad_norm": 0.294690839121686,
      "kl": 0.19085693359375,
      "learning_rate": 4.919562581821316e-07,
      "loss": 0.0002,
      "reward": 1.7410714626312256,
      "reward_std": 0.09343910962343216,
      "rewards/equation_reward_func": 0.7544643059372902,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 14652
    },
    {
      "completion_length": 241.83929634094238,
      "epoch": 2.457018315939478,
      "grad_norm": 0.2685941546167499,
      "kl": 0.141021728515625,
      "learning_rate": 4.919532394817338e-07,
      "loss": 0.0001,
      "reward": 1.8000000640749931,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 14654
    },
    {
      "completion_length": 249.29911994934082,
      "epoch": 2.4573536191793455,
      "grad_norm": 0.12056836743824613,
      "kl": 0.155242919921875,
      "learning_rate": 4.919502202242699e-07,
      "loss": 0.0002,
      "reward": 1.7767857685685158,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7812500316649675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14656
    },
    {
      "completion_length": 237.00893878936768,
      "epoch": 2.457688922419213,
      "grad_norm": 0.1632951800033488,
      "kl": 0.137969970703125,
      "learning_rate": 4.919472004097468e-07,
      "loss": 0.0001,
      "reward": 1.7517857551574707,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7562500350177288,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14658
    },
    {
      "completion_length": 242.84375953674316,
      "epoch": 2.4580242256590803,
      "grad_norm": 0.38433586783614954,
      "kl": 0.147674560546875,
      "learning_rate": 4.919441800381717e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.06565991323441267,
      "rewards/equation_reward_func": 0.7607143148779869,
      "rewards/format_reward_func": 1.0,
      "step": 14660
    },
    {
      "completion_length": 233.5178689956665,
      "epoch": 2.458359528898948,
      "grad_norm": 0.15426940267481715,
      "kl": 0.13128662109375,
      "learning_rate": 4.919411591095512e-07,
      "loss": 0.0001,
      "reward": 1.7357143387198448,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 14662
    },
    {
      "completion_length": 238.9330472946167,
      "epoch": 2.4586948321388156,
      "grad_norm": 0.2574320114609894,
      "kl": 0.115234375,
      "learning_rate": 4.919381376238927e-07,
      "loss": 0.0001,
      "reward": 1.8214286267757416,
      "reward_std": 0.07071067579090595,
      "rewards/equation_reward_func": 0.821428582072258,
      "rewards/format_reward_func": 1.0,
      "step": 14664
    },
    {
      "completion_length": 235.0625123977661,
      "epoch": 2.459030135378683,
      "grad_norm": 0.2578124993607625,
      "kl": 0.14154052734375,
      "learning_rate": 4.919351155812027e-07,
      "loss": 0.0001,
      "reward": 1.8017857521772385,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8062500208616257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14666
    },
    {
      "completion_length": 229.8616180419922,
      "epoch": 2.459365438618551,
      "grad_norm": 0.18994664944682665,
      "kl": 0.11993408203125,
      "learning_rate": 4.919320929814884e-07,
      "loss": 0.0001,
      "reward": 1.7571429163217545,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428883820772,
      "rewards/format_reward_func": 1.0,
      "step": 14668
    },
    {
      "completion_length": 226.6071538925171,
      "epoch": 2.459700741858418,
      "grad_norm": 0.08334989454251703,
      "kl": 0.107177734375,
      "learning_rate": 4.919290698247568e-07,
      "loss": 0.0001,
      "reward": 1.8250000402331352,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.825000025331974,
      "rewards/format_reward_func": 1.0,
      "step": 14670
    },
    {
      "completion_length": 236.5759038925171,
      "epoch": 2.4600360450982857,
      "grad_norm": 0.24064977610399677,
      "kl": 0.124664306640625,
      "learning_rate": 4.919260461110146e-07,
      "loss": 0.0001,
      "reward": 1.7428572177886963,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.742857176810503,
      "rewards/format_reward_func": 1.0,
      "step": 14672
    },
    {
      "completion_length": 228.68304538726807,
      "epoch": 2.4603713483381533,
      "grad_norm": 0.3965086922599267,
      "kl": 0.116363525390625,
      "learning_rate": 4.919230218402692e-07,
      "loss": 0.0001,
      "reward": 1.750000074505806,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7500000298023224,
      "rewards/format_reward_func": 1.0,
      "step": 14674
    },
    {
      "completion_length": 227.08483219146729,
      "epoch": 2.460706651578021,
      "grad_norm": 0.07705829929685584,
      "kl": 0.12017822265625,
      "learning_rate": 4.919199970125271e-07,
      "loss": 0.0001,
      "reward": 1.803571492433548,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8035714365541935,
      "rewards/format_reward_func": 1.0,
      "step": 14676
    },
    {
      "completion_length": 229.67411518096924,
      "epoch": 2.4610419548178886,
      "grad_norm": 0.13454988026363998,
      "kl": 0.10888671875,
      "learning_rate": 4.919169716277956e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7321428898721933,
      "rewards/format_reward_func": 1.0,
      "step": 14678
    },
    {
      "completion_length": 236.2053680419922,
      "epoch": 2.461377258057756,
      "grad_norm": 0.251892997636684,
      "kl": 0.114898681640625,
      "learning_rate": 4.919139456860814e-07,
      "loss": 0.0001,
      "reward": 1.7857143580913544,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 14680
    },
    {
      "completion_length": 227.08483028411865,
      "epoch": 2.4617125612976234,
      "grad_norm": 0.3806151196214663,
      "kl": 0.105987548828125,
      "learning_rate": 4.919109191873917e-07,
      "loss": 0.0001,
      "reward": 1.8142857551574707,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857477068901,
      "rewards/format_reward_func": 1.0,
      "step": 14682
    },
    {
      "completion_length": 240.7991189956665,
      "epoch": 2.462047864537491,
      "grad_norm": 0.15874235465177416,
      "kl": 0.12335205078125,
      "learning_rate": 4.919078921317334e-07,
      "loss": 0.0001,
      "reward": 1.7196429371833801,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.7241071946918964,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14684
    },
    {
      "completion_length": 233.2812614440918,
      "epoch": 2.4623831677773587,
      "grad_norm": 0.24433920234307233,
      "kl": 0.116546630859375,
      "learning_rate": 4.919048645191133e-07,
      "loss": 0.0001,
      "reward": 1.8196429163217545,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.8241071589291096,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14686
    },
    {
      "completion_length": 233.66518878936768,
      "epoch": 2.4627184710172263,
      "grad_norm": 0.1526931970382,
      "kl": 0.1051025390625,
      "learning_rate": 4.919018363495386e-07,
      "loss": 0.0001,
      "reward": 1.7857143580913544,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 14688
    },
    {
      "completion_length": 238.74108219146729,
      "epoch": 2.463053774257094,
      "grad_norm": 0.10079218212524813,
      "kl": 0.10546875,
      "learning_rate": 4.918988076230161e-07,
      "loss": 0.0001,
      "reward": 1.7714286297559738,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7714286036789417,
      "rewards/format_reward_func": 1.0,
      "step": 14690
    },
    {
      "completion_length": 239.29911708831787,
      "epoch": 2.463389077496961,
      "grad_norm": 0.14274660573087072,
      "kl": 0.113311767578125,
      "learning_rate": 4.91895778339553e-07,
      "loss": 0.0001,
      "reward": 1.796428605914116,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 14692
    },
    {
      "completion_length": 246.5580472946167,
      "epoch": 2.4637243807368288,
      "grad_norm": 0.20910002876872932,
      "kl": 0.101806640625,
      "learning_rate": 4.91892748499156e-07,
      "loss": 0.0001,
      "reward": 1.7517857626080513,
      "reward_std": 0.05808377079665661,
      "rewards/equation_reward_func": 0.7562500163912773,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14694
    },
    {
      "completion_length": 240.37054538726807,
      "epoch": 2.4640596839766964,
      "grad_norm": 0.3170325404349844,
      "kl": 0.119476318359375,
      "learning_rate": 4.918897181018323e-07,
      "loss": 0.0001,
      "reward": 1.762500062584877,
      "reward_std": 0.053033008240163326,
      "rewards/equation_reward_func": 0.7669643126428127,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14696
    },
    {
      "completion_length": 237.92411518096924,
      "epoch": 2.464394987216564,
      "grad_norm": 0.17860278813200464,
      "kl": 0.110260009765625,
      "learning_rate": 4.918866871475887e-07,
      "loss": 0.0001,
      "reward": 1.791071467101574,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.795535746961832,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14698
    },
    {
      "completion_length": 240.602689743042,
      "epoch": 2.4647302904564317,
      "grad_norm": 0.19296638822722686,
      "kl": 0.103668212890625,
      "learning_rate": 4.918836556364324e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 14700
    },
    {
      "completion_length": 244.63393783569336,
      "epoch": 2.4650655936962993,
      "grad_norm": 0.1819558090672466,
      "kl": 0.107757568359375,
      "learning_rate": 4.918806235683701e-07,
      "loss": 0.0001,
      "reward": 1.8017857745289803,
      "reward_std": 0.04798224475234747,
      "rewards/equation_reward_func": 0.8062500208616257,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14702
    },
    {
      "completion_length": 239.3884048461914,
      "epoch": 2.4654008969361665,
      "grad_norm": 0.16922460661830055,
      "kl": 0.110870361328125,
      "learning_rate": 4.91877590943409e-07,
      "loss": 0.0001,
      "reward": 1.8500000685453415,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.850000012665987,
      "rewards/format_reward_func": 1.0,
      "step": 14704
    },
    {
      "completion_length": 243.34822177886963,
      "epoch": 2.465736200176034,
      "grad_norm": 0.174083358236681,
      "kl": 0.11090087890625,
      "learning_rate": 4.918745577615559e-07,
      "loss": 0.0001,
      "reward": 1.758928619325161,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7633928712457418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14706
    },
    {
      "completion_length": 236.4866180419922,
      "epoch": 2.4660715034159018,
      "grad_norm": 0.18333538920013637,
      "kl": 0.104248046875,
      "learning_rate": 4.918715240228181e-07,
      "loss": 0.0001,
      "reward": 1.7321429327130318,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7321428954601288,
      "rewards/format_reward_func": 1.0,
      "step": 14708
    },
    {
      "completion_length": 249.57590198516846,
      "epoch": 2.4664068066557694,
      "grad_norm": 0.050767253993794496,
      "kl": 0.119903564453125,
      "learning_rate": 4.918684897272022e-07,
      "loss": 0.0001,
      "reward": 1.7482143491506577,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.7526785843074322,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14710
    },
    {
      "completion_length": 240.40179634094238,
      "epoch": 2.466742109895637,
      "grad_norm": 0.20012551468951106,
      "kl": 0.118896484375,
      "learning_rate": 4.918654548747154e-07,
      "loss": 0.0001,
      "reward": 1.7625000774860382,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.766964316368103,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14712
    },
    {
      "completion_length": 244.321439743042,
      "epoch": 2.467077413135504,
      "grad_norm": 0.18902951420859376,
      "kl": 0.1083984375,
      "learning_rate": 4.918624194653646e-07,
      "loss": 0.0001,
      "reward": 1.771428644657135,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.771428607404232,
      "rewards/format_reward_func": 1.0,
      "step": 14714
    },
    {
      "completion_length": 238.95983028411865,
      "epoch": 2.467412716375372,
      "grad_norm": 0.16578032130190862,
      "kl": 0.136688232421875,
      "learning_rate": 4.918593834991569e-07,
      "loss": 0.0001,
      "reward": 1.821428619325161,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.8214285895228386,
      "rewards/format_reward_func": 1.0,
      "step": 14716
    },
    {
      "completion_length": 240.70983123779297,
      "epoch": 2.4677480196152395,
      "grad_norm": 0.2250877718203696,
      "kl": 0.108428955078125,
      "learning_rate": 4.918563469760992e-07,
      "loss": 0.0001,
      "reward": 1.8000000715255737,
      "reward_std": 0.06060915254056454,
      "rewards/equation_reward_func": 0.8089285865426064,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14718
    },
    {
      "completion_length": 248.5803689956665,
      "epoch": 2.468083322855107,
      "grad_norm": 0.13311610670420726,
      "kl": 0.112213134765625,
      "learning_rate": 4.918533098961985e-07,
      "loss": 0.0001,
      "reward": 1.8142857551574707,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857439815998,
      "rewards/format_reward_func": 1.0,
      "step": 14720
    },
    {
      "completion_length": 242.0312614440918,
      "epoch": 2.4684186260949748,
      "grad_norm": 0.17924485307370552,
      "kl": 0.13079833984375,
      "learning_rate": 4.918502722594619e-07,
      "loss": 0.0001,
      "reward": 1.7642857730388641,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7642857600003481,
      "rewards/format_reward_func": 1.0,
      "step": 14722
    },
    {
      "completion_length": 249.31697940826416,
      "epoch": 2.4687539293348424,
      "grad_norm": 0.2366084108225645,
      "kl": 0.127410888671875,
      "learning_rate": 4.918472340658961e-07,
      "loss": 0.0001,
      "reward": 1.807142935693264,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.8071428798139095,
      "rewards/format_reward_func": 1.0,
      "step": 14724
    },
    {
      "completion_length": 248.94644451141357,
      "epoch": 2.4690892325747096,
      "grad_norm": 0.1999633689515607,
      "kl": 0.123321533203125,
      "learning_rate": 4.918441953155085e-07,
      "loss": 0.0001,
      "reward": 1.7500000670552254,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7500000409781933,
      "rewards/format_reward_func": 1.0,
      "step": 14726
    },
    {
      "completion_length": 251.47768878936768,
      "epoch": 2.469424535814577,
      "grad_norm": 0.1271370337954408,
      "kl": 0.123931884765625,
      "learning_rate": 4.918411560083058e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.796428594738245,
      "rewards/format_reward_func": 1.0,
      "step": 14728
    },
    {
      "completion_length": 249.41965293884277,
      "epoch": 2.469759839054445,
      "grad_norm": 0.11077971547908103,
      "kl": 0.110931396484375,
      "learning_rate": 4.918381161442951e-07,
      "loss": 0.0001,
      "reward": 1.7732143178582191,
      "reward_std": 0.017677669413387775,
      "rewards/equation_reward_func": 0.7776786051690578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14730
    },
    {
      "completion_length": 246.1651906967163,
      "epoch": 2.4700951422943125,
      "grad_norm": 0.21287808362756122,
      "kl": 0.1270751953125,
      "learning_rate": 4.918350757234834e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7821428701281548,
      "rewards/format_reward_func": 1.0,
      "step": 14732
    },
    {
      "completion_length": 250.77233409881592,
      "epoch": 2.47043044553418,
      "grad_norm": 0.1836312903043879,
      "kl": 0.114288330078125,
      "learning_rate": 4.918320347458777e-07,
      "loss": 0.0001,
      "reward": 1.7839286476373672,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7883928790688515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14734
    },
    {
      "completion_length": 248.48661708831787,
      "epoch": 2.4707657487740473,
      "grad_norm": 0.13193506465499494,
      "kl": 0.11639404296875,
      "learning_rate": 4.91828993211485e-07,
      "loss": 0.0001,
      "reward": 1.8285714760422707,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8285714536905289,
      "rewards/format_reward_func": 1.0,
      "step": 14736
    },
    {
      "completion_length": 257.45536613464355,
      "epoch": 2.471101052013915,
      "grad_norm": 0.14764558623734747,
      "kl": 0.123046875,
      "learning_rate": 4.918259511203122e-07,
      "loss": 0.0001,
      "reward": 1.8017857670783997,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.8062500152736902,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14738
    },
    {
      "completion_length": 239.48215293884277,
      "epoch": 2.4714363552537826,
      "grad_norm": 0.22586394380238098,
      "kl": 0.103912353515625,
      "learning_rate": 4.918229084723665e-07,
      "loss": 0.0001,
      "reward": 1.82589291036129,
      "reward_std": 0.0340926474891603,
      "rewards/equation_reward_func": 0.8276785910129547,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 14740
    },
    {
      "completion_length": 254.2321548461914,
      "epoch": 2.47177165849365,
      "grad_norm": 0.25298517588656405,
      "kl": 0.115631103515625,
      "learning_rate": 4.918198652676547e-07,
      "loss": 0.0001,
      "reward": 1.8125000447034836,
      "reward_std": 0.06313453242182732,
      "rewards/equation_reward_func": 0.8169643059372902,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14742
    },
    {
      "completion_length": 259.14287090301514,
      "epoch": 2.472106961733518,
      "grad_norm": 0.16073060918881626,
      "kl": 0.110382080078125,
      "learning_rate": 4.918168215061841e-07,
      "loss": 0.0001,
      "reward": 1.7964286357164383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8053571619093418,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14744
    },
    {
      "completion_length": 260.5535821914673,
      "epoch": 2.4724422649733855,
      "grad_norm": 0.3182146734872481,
      "kl": 0.12188720703125,
      "learning_rate": 4.918137771879614e-07,
      "loss": 0.0001,
      "reward": 1.76071435213089,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143316417933,
      "rewards/format_reward_func": 1.0,
      "step": 14746
    },
    {
      "completion_length": 251.88840579986572,
      "epoch": 2.4727775682132527,
      "grad_norm": 0.3186785826320642,
      "kl": 0.12640380859375,
      "learning_rate": 4.918107323129937e-07,
      "loss": 0.0001,
      "reward": 1.7964286282658577,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 14748
    },
    {
      "completion_length": 255.2187614440918,
      "epoch": 2.4731128714531203,
      "grad_norm": 0.2865873297261211,
      "kl": 0.111236572265625,
      "learning_rate": 4.91807686881288e-07,
      "loss": 0.0001,
      "reward": 1.7892857789993286,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.7892857491970062,
      "rewards/format_reward_func": 1.0,
      "step": 14750
    },
    {
      "completion_length": 248.99554920196533,
      "epoch": 2.473448174692988,
      "grad_norm": 0.1606260793115146,
      "kl": 0.104583740234375,
      "learning_rate": 4.918046408928515e-07,
      "loss": 0.0001,
      "reward": 1.8178572207689285,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8178571611642838,
      "rewards/format_reward_func": 1.0,
      "step": 14752
    },
    {
      "completion_length": 252.3928737640381,
      "epoch": 2.4737834779328556,
      "grad_norm": 0.3196020255422196,
      "kl": 0.107391357421875,
      "learning_rate": 4.91801594347691e-07,
      "loss": 0.0001,
      "reward": 1.792857214808464,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 14754
    },
    {
      "completion_length": 254.96429824829102,
      "epoch": 2.474118781172723,
      "grad_norm": 0.13566741367227556,
      "kl": 0.116729736328125,
      "learning_rate": 4.917985472458135e-07,
      "loss": 0.0001,
      "reward": 1.7517857775092125,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7562500275671482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14756
    },
    {
      "completion_length": 254.2500114440918,
      "epoch": 2.4744540844125904,
      "grad_norm": 0.20523246500830702,
      "kl": 0.113677978515625,
      "learning_rate": 4.917954995872262e-07,
      "loss": 0.0001,
      "reward": 1.773214340209961,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7767857313156128,
      "rewards/format_reward_func": 0.9964285716414452,
      "step": 14758
    },
    {
      "completion_length": 255.88393878936768,
      "epoch": 2.474789387652458,
      "grad_norm": 0.17088202418743903,
      "kl": 0.1111907958984375,
      "learning_rate": 4.917924513719359e-07,
      "loss": 0.0001,
      "reward": 1.7500000894069672,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.750000037252903,
      "rewards/format_reward_func": 1.0,
      "step": 14760
    },
    {
      "completion_length": 248.63840770721436,
      "epoch": 2.4751246908923257,
      "grad_norm": 0.060042085703267425,
      "kl": 0.103485107421875,
      "learning_rate": 4.917894025999498e-07,
      "loss": 0.0001,
      "reward": 1.7857143580913544,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7857143059372902,
      "rewards/format_reward_func": 1.0,
      "step": 14762
    },
    {
      "completion_length": 258.9285821914673,
      "epoch": 2.4754599941321933,
      "grad_norm": 0.10379061448307643,
      "kl": 0.118896484375,
      "learning_rate": 4.917863532712748e-07,
      "loss": 0.0001,
      "reward": 1.7750000730156898,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7839286029338837,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14764
    },
    {
      "completion_length": 245.46876335144043,
      "epoch": 2.475795297372061,
      "grad_norm": 0.08166250755892202,
      "kl": 0.11309814453125,
      "learning_rate": 4.91783303385918e-07,
      "loss": 0.0001,
      "reward": 1.769642911851406,
      "reward_std": 0.022728431969881058,
      "rewards/equation_reward_func": 0.774107176810503,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14766
    },
    {
      "completion_length": 242.16965579986572,
      "epoch": 2.4761306006119286,
      "grad_norm": 0.11966890403423652,
      "kl": 0.120269775390625,
      "learning_rate": 4.917802529438863e-07,
      "loss": 0.0001,
      "reward": 1.7857143506407738,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7857143096625805,
      "rewards/format_reward_func": 1.0,
      "step": 14768
    },
    {
      "completion_length": 244.65179538726807,
      "epoch": 2.476465903851796,
      "grad_norm": 0.06438476377658822,
      "kl": 0.12042236328125,
      "learning_rate": 4.91777201945187e-07,
      "loss": 0.0001,
      "reward": 1.7535714730620384,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7535714507102966,
      "rewards/format_reward_func": 1.0,
      "step": 14770
    },
    {
      "completion_length": 248.00000953674316,
      "epoch": 2.4768012070916634,
      "grad_norm": 0.25166210607398165,
      "kl": 0.1202392578125,
      "learning_rate": 4.917741503898268e-07,
      "loss": 0.0001,
      "reward": 1.8035714849829674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8035714495927095,
      "rewards/format_reward_func": 1.0,
      "step": 14772
    },
    {
      "completion_length": 243.7455472946167,
      "epoch": 2.477136510331531,
      "grad_norm": 0.12492238999535173,
      "kl": 0.111572265625,
      "learning_rate": 4.917710982778129e-07,
      "loss": 0.0001,
      "reward": 1.7107143700122833,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7107143048197031,
      "rewards/format_reward_func": 1.0,
      "step": 14774
    },
    {
      "completion_length": 256.3303689956665,
      "epoch": 2.4774718135713987,
      "grad_norm": 0.23558559483290095,
      "kl": 0.10693359375,
      "learning_rate": 4.917680456091523e-07,
      "loss": 0.0001,
      "reward": 1.798214353621006,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8026785925030708,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14776
    },
    {
      "completion_length": 251.633939743042,
      "epoch": 2.4778071168112663,
      "grad_norm": 0.18466931698986533,
      "kl": 0.118194580078125,
      "learning_rate": 4.917649923838521e-07,
      "loss": 0.0001,
      "reward": 1.7607143595814705,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 14778
    },
    {
      "completion_length": 254.3214406967163,
      "epoch": 2.4781424200511335,
      "grad_norm": 0.18040244234467673,
      "kl": 0.122772216796875,
      "learning_rate": 4.917619386019191e-07,
      "loss": 0.0001,
      "reward": 1.7482143640518188,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7526785880327225,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14780
    },
    {
      "completion_length": 246.53572463989258,
      "epoch": 2.478477723291001,
      "grad_norm": 0.0026001646841756804,
      "kl": 0.106536865234375,
      "learning_rate": 4.917588842633605e-07,
      "loss": 0.0001,
      "reward": 1.8035714775323868,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8035714402794838,
      "rewards/format_reward_func": 1.0,
      "step": 14782
    },
    {
      "completion_length": 257.0714406967163,
      "epoch": 2.4788130265308688,
      "grad_norm": 0.2176137843266907,
      "kl": 0.1085205078125,
      "learning_rate": 4.917558293681834e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.778571454808116,
      "rewards/format_reward_func": 1.0,
      "step": 14784
    },
    {
      "completion_length": 252.01340293884277,
      "epoch": 2.4791483297707364,
      "grad_norm": 0.08079521274065427,
      "kl": 0.102691650390625,
      "learning_rate": 4.917527739163947e-07,
      "loss": 0.0001,
      "reward": 1.7750000804662704,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000320374966,
      "rewards/format_reward_func": 1.0,
      "step": 14786
    },
    {
      "completion_length": 241.19197750091553,
      "epoch": 2.479483633010604,
      "grad_norm": 0.06630944765696818,
      "kl": 0.09710693359375,
      "learning_rate": 4.917497179080013e-07,
      "loss": 0.0001,
      "reward": 1.8250000476837158,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8250000178813934,
      "rewards/format_reward_func": 1.0,
      "step": 14788
    },
    {
      "completion_length": 244.2857255935669,
      "epoch": 2.4798189362504717,
      "grad_norm": 0.15402009555722682,
      "kl": 0.108489990234375,
      "learning_rate": 4.917466613430107e-07,
      "loss": 0.0001,
      "reward": 1.7678572162985802,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 14790
    },
    {
      "completion_length": 253.34376430511475,
      "epoch": 2.4801542394903393,
      "grad_norm": 0.22679226122504897,
      "kl": 0.116607666015625,
      "learning_rate": 4.917436042214294e-07,
      "loss": 0.0001,
      "reward": 1.7464286386966705,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.755357164889574,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14792
    },
    {
      "completion_length": 248.57143878936768,
      "epoch": 2.4804895427302065,
      "grad_norm": 0.14180333609406645,
      "kl": 0.107330322265625,
      "learning_rate": 4.917405465432649e-07,
      "loss": 0.0001,
      "reward": 1.757142923772335,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428939700127,
      "rewards/format_reward_func": 1.0,
      "step": 14794
    },
    {
      "completion_length": 250.1607255935669,
      "epoch": 2.480824845970074,
      "grad_norm": 0.21500318228619053,
      "kl": 0.118072509765625,
      "learning_rate": 4.91737488308524e-07,
      "loss": 0.0001,
      "reward": 1.7821428924798965,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428813040257,
      "rewards/format_reward_func": 1.0,
      "step": 14796
    },
    {
      "completion_length": 248.2678689956665,
      "epoch": 2.4811601492099418,
      "grad_norm": 0.17072486776960497,
      "kl": 0.1063385009765625,
      "learning_rate": 4.917344295172137e-07,
      "loss": 0.0001,
      "reward": 1.8000000566244125,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8000000268220901,
      "rewards/format_reward_func": 1.0,
      "step": 14798
    },
    {
      "completion_length": 241.50447845458984,
      "epoch": 2.4814954524498094,
      "grad_norm": 0.16210839854375575,
      "kl": 0.114166259765625,
      "learning_rate": 4.917313701693412e-07,
      "loss": 0.0001,
      "reward": 1.792857214808464,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571701049805,
      "rewards/format_reward_func": 1.0,
      "step": 14800
    },
    {
      "completion_length": 245.27233695983887,
      "epoch": 2.481830755689677,
      "grad_norm": 0.18009075898418456,
      "kl": 0.11663818359375,
      "learning_rate": 4.917283102649133e-07,
      "loss": 0.0001,
      "reward": 1.7821429371833801,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 14802
    },
    {
      "completion_length": 252.86161994934082,
      "epoch": 2.4821660589295442,
      "grad_norm": 0.22212604391809182,
      "kl": 0.1253662109375,
      "learning_rate": 4.917252498039374e-07,
      "loss": 0.0001,
      "reward": 1.7589286267757416,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7633928880095482,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14804
    },
    {
      "completion_length": 247.05804824829102,
      "epoch": 2.482501362169412,
      "grad_norm": 0.17060599530084553,
      "kl": 0.10174560546875,
      "learning_rate": 4.917221887864202e-07,
      "loss": 0.0001,
      "reward": 1.776785783469677,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7812500186264515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14806
    },
    {
      "completion_length": 248.0044765472412,
      "epoch": 2.4828366654092795,
      "grad_norm": 0.19718760886727332,
      "kl": 0.108856201171875,
      "learning_rate": 4.91719127212369e-07,
      "loss": 0.0001,
      "reward": 1.7607143446803093,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143092900515,
      "rewards/format_reward_func": 1.0,
      "step": 14808
    },
    {
      "completion_length": 256.0312604904175,
      "epoch": 2.483171968649147,
      "grad_norm": 0.002740858816126789,
      "kl": 0.102996826171875,
      "learning_rate": 4.917160650817906e-07,
      "loss": 0.0001,
      "reward": 1.735714353621006,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7357143200933933,
      "rewards/format_reward_func": 1.0,
      "step": 14810
    },
    {
      "completion_length": 246.3884048461914,
      "epoch": 2.4835072718890148,
      "grad_norm": 0.24967328188320786,
      "kl": 0.11102294921875,
      "learning_rate": 4.917130023946924e-07,
      "loss": 0.0001,
      "reward": 1.7678572088479996,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7678571790456772,
      "rewards/format_reward_func": 1.0,
      "step": 14812
    },
    {
      "completion_length": 247.1696548461914,
      "epoch": 2.4838425751288824,
      "grad_norm": 0.14186057536510902,
      "kl": 0.105712890625,
      "learning_rate": 4.917099391510811e-07,
      "loss": 0.0001,
      "reward": 1.8464286103844643,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8464285992085934,
      "rewards/format_reward_func": 1.0,
      "step": 14814
    },
    {
      "completion_length": 243.70090293884277,
      "epoch": 2.4841778783687496,
      "grad_norm": 0.16875123376236464,
      "kl": 0.115570068359375,
      "learning_rate": 4.917068753509639e-07,
      "loss": 0.0001,
      "reward": 1.7821429148316383,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.782142885029316,
      "rewards/format_reward_func": 1.0,
      "step": 14816
    },
    {
      "completion_length": 251.5178680419922,
      "epoch": 2.4845131816086172,
      "grad_norm": 0.18352318078910093,
      "kl": 0.108489990234375,
      "learning_rate": 4.917038109943479e-07,
      "loss": 0.0001,
      "reward": 1.7071429267525673,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7071429006755352,
      "rewards/format_reward_func": 1.0,
      "step": 14818
    },
    {
      "completion_length": 245.8437623977661,
      "epoch": 2.484848484848485,
      "grad_norm": 0.16090355001841908,
      "kl": 0.105133056640625,
      "learning_rate": 4.917007460812401e-07,
      "loss": 0.0001,
      "reward": 1.8214286118745804,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8214285857975483,
      "rewards/format_reward_func": 1.0,
      "step": 14820
    },
    {
      "completion_length": 246.42858219146729,
      "epoch": 2.4851837880883525,
      "grad_norm": 0.1380564013172623,
      "kl": 0.09600830078125,
      "learning_rate": 4.916976806116476e-07,
      "loss": 0.0001,
      "reward": 1.8196429163217545,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8241071663796902,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14822
    },
    {
      "completion_length": 246.72322750091553,
      "epoch": 2.48551909132822,
      "grad_norm": 0.15147188409912418,
      "kl": 0.097808837890625,
      "learning_rate": 4.916946145855774e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857357859612,
      "rewards/format_reward_func": 1.0,
      "step": 14824
    },
    {
      "completion_length": 249.93304634094238,
      "epoch": 2.4858543945680873,
      "grad_norm": 0.25257574082870304,
      "kl": 0.1040191650390625,
      "learning_rate": 4.916915480030365e-07,
      "loss": 0.0001,
      "reward": 1.7464286535978317,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.746428593993187,
      "rewards/format_reward_func": 1.0,
      "step": 14826
    },
    {
      "completion_length": 247.65626049041748,
      "epoch": 2.486189697807955,
      "grad_norm": 0.09384685704779432,
      "kl": 0.09881591796875,
      "learning_rate": 4.916884808640323e-07,
      "loss": 0.0001,
      "reward": 1.7464286163449287,
      "reward_std": 0.005050762556493282,
      "rewards/equation_reward_func": 0.7464286126196384,
      "rewards/format_reward_func": 1.0,
      "step": 14828
    },
    {
      "completion_length": 248.2321548461914,
      "epoch": 2.4865250010478226,
      "grad_norm": 0.16828925861257932,
      "kl": 0.105712890625,
      "learning_rate": 4.916854131685713e-07,
      "loss": 0.0001,
      "reward": 1.755357213318348,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.7598214708268642,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14830
    },
    {
      "completion_length": 242.9285831451416,
      "epoch": 2.4868603042876902,
      "grad_norm": 0.15619959437078137,
      "kl": 0.10552978515625,
      "learning_rate": 4.916823449166611e-07,
      "loss": 0.0001,
      "reward": 1.7267857939004898,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.7312500365078449,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14832
    },
    {
      "completion_length": 254.9955472946167,
      "epoch": 2.487195607527558,
      "grad_norm": 0.2567715716716288,
      "kl": 0.0962982177734375,
      "learning_rate": 4.916792761083084e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 14834
    },
    {
      "completion_length": 255.8794755935669,
      "epoch": 2.4875309107674255,
      "grad_norm": 0.13534000071902677,
      "kl": 0.105438232421875,
      "learning_rate": 4.916762067435204e-07,
      "loss": 0.0001,
      "reward": 1.7928571999073029,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7928571812808514,
      "rewards/format_reward_func": 1.0,
      "step": 14836
    },
    {
      "completion_length": 252.67858600616455,
      "epoch": 2.4878662140072927,
      "grad_norm": 0.24524325288408574,
      "kl": 0.112060546875,
      "learning_rate": 4.916731368223042e-07,
      "loss": 0.0001,
      "reward": 1.7142857983708382,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.714285746216774,
      "rewards/format_reward_func": 1.0,
      "step": 14838
    },
    {
      "completion_length": 254.23215198516846,
      "epoch": 2.4882015172471603,
      "grad_norm": 0.21139274158098173,
      "kl": 0.104644775390625,
      "learning_rate": 4.916700663446668e-07,
      "loss": 0.0001,
      "reward": 1.7607143446803093,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7607143297791481,
      "rewards/format_reward_func": 1.0,
      "step": 14840
    },
    {
      "completion_length": 259.102689743042,
      "epoch": 2.488536820487028,
      "grad_norm": 0.14614650915831776,
      "kl": 0.117706298828125,
      "learning_rate": 4.916669953106154e-07,
      "loss": 0.0001,
      "reward": 1.7214286401867867,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7214285917580128,
      "rewards/format_reward_func": 1.0,
      "step": 14842
    },
    {
      "completion_length": 248.5491189956665,
      "epoch": 2.4888721237268956,
      "grad_norm": 0.24970700735528656,
      "kl": 0.1097412109375,
      "learning_rate": 4.916639237201568e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7357143126428127,
      "rewards/format_reward_func": 1.0,
      "step": 14844
    },
    {
      "completion_length": 250.30804538726807,
      "epoch": 2.4892074269667632,
      "grad_norm": 0.15992926840127522,
      "kl": 0.1121826171875,
      "learning_rate": 4.916608515732984e-07,
      "loss": 0.0001,
      "reward": 1.7035714834928513,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7035714685916901,
      "rewards/format_reward_func": 1.0,
      "step": 14846
    },
    {
      "completion_length": 249.04018878936768,
      "epoch": 2.4895427302066304,
      "grad_norm": 0.20425696513495342,
      "kl": 0.096221923828125,
      "learning_rate": 4.916577788700471e-07,
      "loss": 0.0001,
      "reward": 1.79464291036129,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7991071715950966,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14848
    },
    {
      "completion_length": 255.94644165039062,
      "epoch": 2.489878033446498,
      "grad_norm": 0.2703017241730625,
      "kl": 0.104248046875,
      "learning_rate": 4.9165470561041e-07,
      "loss": 0.0001,
      "reward": 1.7660714834928513,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7705357447266579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14850
    },
    {
      "completion_length": 251.53572750091553,
      "epoch": 2.4902133366863657,
      "grad_norm": 0.08715365745771998,
      "kl": 0.109375,
      "learning_rate": 4.916516317943942e-07,
      "loss": 0.0001,
      "reward": 1.7464286237955093,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7464286014437675,
      "rewards/format_reward_func": 1.0,
      "step": 14852
    },
    {
      "completion_length": 242.1071548461914,
      "epoch": 2.4905486399262333,
      "grad_norm": 0.1947782628519035,
      "kl": 0.105072021484375,
      "learning_rate": 4.916485574220066e-07,
      "loss": 0.0001,
      "reward": 1.7767857760190964,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.7812500223517418,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14854
    },
    {
      "completion_length": 244.5178689956665,
      "epoch": 2.490883943166101,
      "grad_norm": 0.17149906856996142,
      "kl": 0.099365234375,
      "learning_rate": 4.916454824932545e-07,
      "loss": 0.0001,
      "reward": 1.8321429193019867,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.8321428745985031,
      "rewards/format_reward_func": 1.0,
      "step": 14856
    },
    {
      "completion_length": 253.70536613464355,
      "epoch": 2.4912192464059686,
      "grad_norm": 0.225006882706186,
      "kl": 0.106292724609375,
      "learning_rate": 4.916424070081448e-07,
      "loss": 0.0001,
      "reward": 1.7357143685221672,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7357143089175224,
      "rewards/format_reward_func": 1.0,
      "step": 14858
    },
    {
      "completion_length": 258.6651906967163,
      "epoch": 2.491554549645836,
      "grad_norm": 0.19474515995727912,
      "kl": 0.110687255859375,
      "learning_rate": 4.916393309666849e-07,
      "loss": 0.0001,
      "reward": 1.7892857640981674,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7892857491970062,
      "rewards/format_reward_func": 1.0,
      "step": 14860
    },
    {
      "completion_length": 252.4821548461914,
      "epoch": 2.4918898528857034,
      "grad_norm": 0.25165869699775634,
      "kl": 0.0942230224609375,
      "learning_rate": 4.916362543688816e-07,
      "loss": 0.0001,
      "reward": 1.7392857819795609,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.739285746589303,
      "rewards/format_reward_func": 1.0,
      "step": 14862
    },
    {
      "completion_length": 253.31697750091553,
      "epoch": 2.492225156125571,
      "grad_norm": 0.4725974865113826,
      "kl": 0.11981201171875,
      "learning_rate": 4.91633177214742e-07,
      "loss": 0.0001,
      "reward": 1.7517857924103737,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7562500238418579,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14864
    },
    {
      "completion_length": 246.87947463989258,
      "epoch": 2.4925604593654387,
      "grad_norm": 0.16160872714609656,
      "kl": 0.10552978515625,
      "learning_rate": 4.916300995042732e-07,
      "loss": 0.0001,
      "reward": 1.8142857626080513,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857514321804,
      "rewards/format_reward_func": 1.0,
      "step": 14866
    },
    {
      "completion_length": 250.52679443359375,
      "epoch": 2.4928957626053063,
      "grad_norm": 0.09402971091803046,
      "kl": 0.097747802734375,
      "learning_rate": 4.916270212374824e-07,
      "loss": 0.0001,
      "reward": 1.803571492433548,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8035714440047741,
      "rewards/format_reward_func": 1.0,
      "step": 14868
    },
    {
      "completion_length": 256.6116189956665,
      "epoch": 2.4932310658451735,
      "grad_norm": 0.16822883694092985,
      "kl": 0.1134033203125,
      "learning_rate": 4.916239424143766e-07,
      "loss": 0.0001,
      "reward": 1.7375000640749931,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7419643253087997,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14870
    },
    {
      "completion_length": 247.7544765472412,
      "epoch": 2.493566369085041,
      "grad_norm": 0.17898003366783408,
      "kl": 0.10797119140625,
      "learning_rate": 4.916208630349628e-07,
      "loss": 0.0001,
      "reward": 1.835714340209961,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.835714302957058,
      "rewards/format_reward_func": 1.0,
      "step": 14872
    },
    {
      "completion_length": 237.7410831451416,
      "epoch": 2.493901672324909,
      "grad_norm": 0.15222057489193858,
      "kl": 0.100433349609375,
      "learning_rate": 4.916177830992482e-07,
      "loss": 0.0001,
      "reward": 1.8107143267989159,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.8107143044471741,
      "rewards/format_reward_func": 1.0,
      "step": 14874
    },
    {
      "completion_length": 253.51786994934082,
      "epoch": 2.4942369755647764,
      "grad_norm": 0.16371729482573208,
      "kl": 0.10443115234375,
      "learning_rate": 4.9161470260724e-07,
      "loss": 0.0001,
      "reward": 1.7732143476605415,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7776785977184772,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14876
    },
    {
      "completion_length": 263.071439743042,
      "epoch": 2.494572278804644,
      "grad_norm": 0.19259044880307022,
      "kl": 0.119903564453125,
      "learning_rate": 4.916116215589451e-07,
      "loss": 0.0001,
      "reward": 1.7732143476605415,
      "reward_std": 0.0883883461356163,
      "rewards/equation_reward_func": 0.7776786051690578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14878
    },
    {
      "completion_length": 253.60268878936768,
      "epoch": 2.4949075820445117,
      "grad_norm": 0.13104247323147747,
      "kl": 0.109039306640625,
      "learning_rate": 4.916085399543707e-07,
      "loss": 0.0001,
      "reward": 1.7321429178118706,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7321428973227739,
      "rewards/format_reward_func": 1.0,
      "step": 14880
    },
    {
      "completion_length": 247.28126335144043,
      "epoch": 2.495242885284379,
      "grad_norm": 0.0019853055196513178,
      "kl": 0.099822998046875,
      "learning_rate": 4.916054577935238e-07,
      "loss": 0.0001,
      "reward": 1.791071467101574,
      "reward_std": 0.012626906856894493,
      "rewards/equation_reward_func": 0.7955357432365417,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14882
    },
    {
      "completion_length": 252.61608600616455,
      "epoch": 2.4955781885242465,
      "grad_norm": 0.18710983271999201,
      "kl": 0.117523193359375,
      "learning_rate": 4.916023750764116e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 14884
    },
    {
      "completion_length": 252.84375858306885,
      "epoch": 2.495913491764114,
      "grad_norm": 0.18797587280849457,
      "kl": 0.1116943359375,
      "learning_rate": 4.915992918030412e-07,
      "loss": 0.0001,
      "reward": 1.8125000596046448,
      "reward_std": 0.03282995708286762,
      "rewards/equation_reward_func": 0.8169643096625805,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14886
    },
    {
      "completion_length": 247.48662090301514,
      "epoch": 2.496248795003982,
      "grad_norm": 0.27004660458537005,
      "kl": 0.1292724609375,
      "learning_rate": 4.915962079734195e-07,
      "loss": 0.0001,
      "reward": 1.7964286357164383,
      "reward_std": 0.04545686300843954,
      "rewards/equation_reward_func": 0.7964285910129547,
      "rewards/format_reward_func": 1.0,
      "step": 14888
    },
    {
      "completion_length": 252.14733219146729,
      "epoch": 2.4965840982438494,
      "grad_norm": 0.12158380895645579,
      "kl": 0.120574951171875,
      "learning_rate": 4.915931235875538e-07,
      "loss": 0.0001,
      "reward": 1.7785714864730835,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 14890
    },
    {
      "completion_length": 261.8794765472412,
      "epoch": 2.4969194014837166,
      "grad_norm": 0.12848837493711257,
      "kl": 0.1173095703125,
      "learning_rate": 4.915900386454512e-07,
      "loss": 0.0001,
      "reward": 1.817857213318348,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.8178571723401546,
      "rewards/format_reward_func": 1.0,
      "step": 14892
    },
    {
      "completion_length": 257.508939743042,
      "epoch": 2.4972547047235842,
      "grad_norm": 0.37214306927685425,
      "kl": 0.11920166015625,
      "learning_rate": 4.915869531471188e-07,
      "loss": 0.0001,
      "reward": 1.7642857804894447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857357859612,
      "rewards/format_reward_func": 1.0,
      "step": 14894
    },
    {
      "completion_length": 257.415189743042,
      "epoch": 2.497590007963452,
      "grad_norm": 0.19139334417935533,
      "kl": 0.1249847412109375,
      "learning_rate": 4.915838670925636e-07,
      "loss": 0.0001,
      "reward": 1.778571479022503,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7785714585334063,
      "rewards/format_reward_func": 1.0,
      "step": 14896
    },
    {
      "completion_length": 258.83037090301514,
      "epoch": 2.4979253112033195,
      "grad_norm": 0.1739975764465377,
      "kl": 0.11895751953125,
      "learning_rate": 4.915807804817927e-07,
      "loss": 0.0001,
      "reward": 1.7625000402331352,
      "reward_std": 0.06313453521579504,
      "rewards/equation_reward_func": 0.775892885401845,
      "rewards/format_reward_func": 0.9866071492433548,
      "step": 14898
    },
    {
      "completion_length": 257.80358028411865,
      "epoch": 2.498260614443187,
      "grad_norm": 0.12046699172711495,
      "kl": 0.124237060546875,
      "learning_rate": 4.915776933148135e-07,
      "loss": 0.0001,
      "reward": 1.81428574770689,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8142857402563095,
      "rewards/format_reward_func": 1.0,
      "step": 14900
    },
    {
      "completion_length": 263.2187604904175,
      "epoch": 2.498595917683055,
      "grad_norm": 0.35322802369354245,
      "kl": 0.1260986328125,
      "learning_rate": 4.915746055916327e-07,
      "loss": 0.0001,
      "reward": 1.7803571969270706,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7848214488476515,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14902
    },
    {
      "completion_length": 246.6428689956665,
      "epoch": 2.4989312209229224,
      "grad_norm": 0.004772484218738324,
      "kl": 0.1209869384765625,
      "learning_rate": 4.915715173122575e-07,
      "loss": 0.0001,
      "reward": 1.7857143431901932,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 14904
    },
    {
      "completion_length": 258.6875104904175,
      "epoch": 2.4992665241627896,
      "grad_norm": 0.14919904400282405,
      "kl": 0.1072998046875,
      "learning_rate": 4.915684284766953e-07,
      "loss": 0.0001,
      "reward": 1.7785714715719223,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714641213417,
      "rewards/format_reward_func": 1.0,
      "step": 14906
    },
    {
      "completion_length": 253.6205472946167,
      "epoch": 2.4996018274026572,
      "grad_norm": 0.25196101649712355,
      "kl": 0.10772705078125,
      "learning_rate": 4.915653390849529e-07,
      "loss": 0.0001,
      "reward": 1.832142911851406,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8321428708732128,
      "rewards/format_reward_func": 1.0,
      "step": 14908
    },
    {
      "completion_length": 253.071439743042,
      "epoch": 2.499937130642525,
      "grad_norm": 0.1945168808074864,
      "kl": 0.124420166015625,
      "learning_rate": 4.915622491370376e-07,
      "loss": 0.0001,
      "reward": 1.7821429073810577,
      "reward_std": 0.055558389984071255,
      "rewards/equation_reward_func": 0.7910714522004128,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14910
    },
    {
      "completion_length": 250.1384038925171,
      "epoch": 2.5002724338823925,
      "grad_norm": 0.15781246856605685,
      "kl": 0.1234130859375,
      "learning_rate": 4.915591586329563e-07,
      "loss": 0.0001,
      "reward": 1.7928571850061417,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7928571812808514,
      "rewards/format_reward_func": 1.0,
      "step": 14912
    },
    {
      "completion_length": 255.00894165039062,
      "epoch": 2.5006077371222597,
      "grad_norm": 0.2201245062876672,
      "kl": 0.155975341796875,
      "learning_rate": 4.915560675727164e-07,
      "loss": 0.0002,
      "reward": 1.8053572103381157,
      "reward_std": 0.05303300730884075,
      "rewards/equation_reward_func": 0.8098214566707611,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14914
    },
    {
      "completion_length": 252.48215579986572,
      "epoch": 2.5009430403621273,
      "grad_norm": 0.14755556808748146,
      "kl": 0.123779296875,
      "learning_rate": 4.915529759563248e-07,
      "loss": 0.0001,
      "reward": 1.796428620815277,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286096394062,
      "rewards/format_reward_func": 1.0,
      "step": 14916
    },
    {
      "completion_length": 250.58483123779297,
      "epoch": 2.501278343601995,
      "grad_norm": 0.16455636465986687,
      "kl": 0.120697021484375,
      "learning_rate": 4.915498837837887e-07,
      "loss": 0.0001,
      "reward": 1.785714365541935,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857143133878708,
      "rewards/format_reward_func": 1.0,
      "step": 14918
    },
    {
      "completion_length": 260.75894260406494,
      "epoch": 2.5016136468418626,
      "grad_norm": 0.17272260899471706,
      "kl": 0.1383056640625,
      "learning_rate": 4.915467910551153e-07,
      "loss": 0.0001,
      "reward": 1.814285784959793,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8142857477068901,
      "rewards/format_reward_func": 1.0,
      "step": 14920
    },
    {
      "completion_length": 256.6205463409424,
      "epoch": 2.5019489500817302,
      "grad_norm": 0.19178782600827965,
      "kl": 0.25885009765625,
      "learning_rate": 4.915436977703114e-07,
      "loss": 0.0003,
      "reward": 1.8107143566012383,
      "reward_std": 0.055558388121426105,
      "rewards/equation_reward_func": 0.8107143063098192,
      "rewards/format_reward_func": 1.0,
      "step": 14922
    },
    {
      "completion_length": 258.1964416503906,
      "epoch": 2.502284253321598,
      "grad_norm": 0.39726649210975157,
      "kl": 0.156768798828125,
      "learning_rate": 4.915406039293845e-07,
      "loss": 0.0002,
      "reward": 1.7821428924798965,
      "reward_std": 0.06565991416573524,
      "rewards/equation_reward_func": 0.7910714633762836,
      "rewards/format_reward_func": 0.9910714328289032,
      "step": 14924
    },
    {
      "completion_length": 268.22769355773926,
      "epoch": 2.5026195565614655,
      "grad_norm": 0.2066904623535352,
      "kl": 0.150665283203125,
      "learning_rate": 4.915375095323417e-07,
      "loss": 0.0002,
      "reward": 1.780357226729393,
      "reward_std": 0.0681852949783206,
      "rewards/equation_reward_func": 0.7848214618861675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14926
    },
    {
      "completion_length": 257.2232246398926,
      "epoch": 2.5029548598013327,
      "grad_norm": 0.21913485169174565,
      "kl": 0.150390625,
      "learning_rate": 4.915344145791898e-07,
      "loss": 0.0002,
      "reward": 1.7142858058214188,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.7142857611179352,
      "rewards/format_reward_func": 1.0,
      "step": 14928
    },
    {
      "completion_length": 263.6607255935669,
      "epoch": 2.5032901630412003,
      "grad_norm": 0.2649583866173159,
      "kl": 0.17791748046875,
      "learning_rate": 4.915313190699362e-07,
      "loss": 0.0002,
      "reward": 1.7160714864730835,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7205357383936644,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14930
    },
    {
      "completion_length": 266.9955472946167,
      "epoch": 2.503625466281068,
      "grad_norm": 0.20459147581001738,
      "kl": 0.191192626953125,
      "learning_rate": 4.915282230045878e-07,
      "loss": 0.0002,
      "reward": 1.7732143253087997,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7776786051690578,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14932
    },
    {
      "completion_length": 255.90180015563965,
      "epoch": 2.5039607695209356,
      "grad_norm": 0.25484307197764544,
      "kl": 0.142059326171875,
      "learning_rate": 4.91525126383152e-07,
      "loss": 0.0001,
      "reward": 1.7964286357164383,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7964286021888256,
      "rewards/format_reward_func": 1.0,
      "step": 14934
    },
    {
      "completion_length": 267.54912090301514,
      "epoch": 2.504296072760803,
      "grad_norm": 0.15722467533159465,
      "kl": 0.2144775390625,
      "learning_rate": 4.915220292056359e-07,
      "loss": 0.0002,
      "reward": 1.753571517765522,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714730620384,
      "rewards/format_reward_func": 1.0,
      "step": 14936
    },
    {
      "completion_length": 250.33483219146729,
      "epoch": 2.5046313760006704,
      "grad_norm": 0.13771303946559407,
      "kl": 0.122711181640625,
      "learning_rate": 4.915189314720465e-07,
      "loss": 0.0001,
      "reward": 1.7982143387198448,
      "reward_std": 0.04293148219585419,
      "rewards/equation_reward_func": 0.802678607404232,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14938
    },
    {
      "completion_length": 247.95536708831787,
      "epoch": 2.504966679240538,
      "grad_norm": 0.15147710220354377,
      "kl": 0.14837646484375,
      "learning_rate": 4.915158331823909e-07,
      "loss": 0.0001,
      "reward": 1.796428643167019,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7964285910129547,
      "rewards/format_reward_func": 1.0,
      "step": 14940
    },
    {
      "completion_length": 252.75000953674316,
      "epoch": 2.5053019824804057,
      "grad_norm": 0.09408187522982184,
      "kl": 0.2237548828125,
      "learning_rate": 4.915127343366763e-07,
      "loss": 0.0002,
      "reward": 1.760714367032051,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7607143111526966,
      "rewards/format_reward_func": 1.0,
      "step": 14942
    },
    {
      "completion_length": 250.71430015563965,
      "epoch": 2.5056372857202733,
      "grad_norm": 0.09563232486259707,
      "kl": 0.1910400390625,
      "learning_rate": 4.915096349349098e-07,
      "loss": 0.0002,
      "reward": 1.7875000312924385,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.7919643223285675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14944
    },
    {
      "completion_length": 249.0937623977661,
      "epoch": 2.505972588960141,
      "grad_norm": 0.17560738401963297,
      "kl": 0.17327880859375,
      "learning_rate": 4.915065349770987e-07,
      "loss": 0.0002,
      "reward": 1.7428572252392769,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7428571805357933,
      "rewards/format_reward_func": 1.0,
      "step": 14946
    },
    {
      "completion_length": 247.321439743042,
      "epoch": 2.5063078922000086,
      "grad_norm": 0.4454910772698461,
      "kl": 0.14111328125,
      "learning_rate": 4.9150343446325e-07,
      "loss": 0.0001,
      "reward": 1.7571429312229156,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7571428865194321,
      "rewards/format_reward_func": 1.0,
      "step": 14948
    },
    {
      "completion_length": 249.8571538925171,
      "epoch": 2.506643195439876,
      "grad_norm": 0.24536262681171336,
      "kl": 0.170684814453125,
      "learning_rate": 4.915003333933708e-07,
      "loss": 0.0002,
      "reward": 1.7937500774860382,
      "reward_std": 0.049244935624301434,
      "rewards/equation_reward_func": 0.7955357395112514,
      "rewards/format_reward_func": 0.9982142895460129,
      "step": 14950
    },
    {
      "completion_length": 238.4687614440918,
      "epoch": 2.5069784986797434,
      "grad_norm": 0.25230609198150733,
      "kl": 0.148651123046875,
      "learning_rate": 4.914972317674683e-07,
      "loss": 0.0001,
      "reward": 1.782142885029316,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428831666708,
      "rewards/format_reward_func": 1.0,
      "step": 14952
    },
    {
      "completion_length": 250.16072368621826,
      "epoch": 2.507313801919611,
      "grad_norm": 0.1501950138335119,
      "kl": 0.25262451171875,
      "learning_rate": 4.914941295855496e-07,
      "loss": 0.0003,
      "reward": 1.7785714715719223,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 1.0,
      "step": 14954
    },
    {
      "completion_length": 233.84376335144043,
      "epoch": 2.5076491051594787,
      "grad_norm": 0.10516466220468688,
      "kl": 0.159942626953125,
      "learning_rate": 4.91491026847622e-07,
      "loss": 0.0002,
      "reward": 1.7857143431901932,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.7857142947614193,
      "rewards/format_reward_func": 1.0,
      "step": 14956
    },
    {
      "completion_length": 245.0982265472412,
      "epoch": 2.507984408399346,
      "grad_norm": 0.22351192817369384,
      "kl": 0.1419677734375,
      "learning_rate": 4.914879235536924e-07,
      "loss": 0.0001,
      "reward": 1.8250000402331352,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.8250000290572643,
      "rewards/format_reward_func": 1.0,
      "step": 14958
    },
    {
      "completion_length": 234.88840293884277,
      "epoch": 2.5083197116392135,
      "grad_norm": 0.10862469988182176,
      "kl": 0.117095947265625,
      "learning_rate": 4.914848197037681e-07,
      "loss": 0.0001,
      "reward": 1.7232143506407738,
      "reward_std": 0.02777919452637434,
      "rewards/equation_reward_func": 0.727678595110774,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14960
    },
    {
      "completion_length": 246.5982255935669,
      "epoch": 2.508655014879081,
      "grad_norm": 0.34570987382328267,
      "kl": 0.117919921875,
      "learning_rate": 4.914817152978561e-07,
      "loss": 0.0001,
      "reward": 1.7428572103381157,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7428571693599224,
      "rewards/format_reward_func": 1.0,
      "step": 14962
    },
    {
      "completion_length": 245.7232265472412,
      "epoch": 2.508990318118949,
      "grad_norm": 0.16383356087510462,
      "kl": 0.11492919921875,
      "learning_rate": 4.914786103359639e-07,
      "loss": 0.0001,
      "reward": 1.7678571939468384,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7678571902215481,
      "rewards/format_reward_func": 1.0,
      "step": 14964
    },
    {
      "completion_length": 229.80804634094238,
      "epoch": 2.5093256213588164,
      "grad_norm": 0.3562128005704095,
      "kl": 0.173583984375,
      "learning_rate": 4.914755048180981e-07,
      "loss": 0.0002,
      "reward": 1.7875000536441803,
      "reward_std": 0.058083769865334034,
      "rewards/equation_reward_func": 0.7919643074274063,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14966
    },
    {
      "completion_length": 235.73215293884277,
      "epoch": 2.509660924598684,
      "grad_norm": 0.20378825155179361,
      "kl": 0.159698486328125,
      "learning_rate": 4.914723987442664e-07,
      "loss": 0.0002,
      "reward": 1.7785715013742447,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7785714603960514,
      "rewards/format_reward_func": 1.0,
      "step": 14968
    },
    {
      "completion_length": 240.1250114440918,
      "epoch": 2.5099962278385517,
      "grad_norm": 0.0031109090255418614,
      "kl": 0.117584228515625,
      "learning_rate": 4.914692921144756e-07,
      "loss": 0.0001,
      "reward": 1.8500000312924385,
      "reward_std": 0.010101525112986565,
      "rewards/equation_reward_func": 0.8500000312924385,
      "rewards/format_reward_func": 1.0,
      "step": 14970
    },
    {
      "completion_length": 243.4732265472412,
      "epoch": 2.510331531078419,
      "grad_norm": 0.1671077710760345,
      "kl": 0.12384033203125,
      "learning_rate": 4.91466184928733e-07,
      "loss": 0.0001,
      "reward": 1.8285714611411095,
      "reward_std": 0.04040610045194626,
      "rewards/equation_reward_func": 0.8285714574158192,
      "rewards/format_reward_func": 1.0,
      "step": 14972
    },
    {
      "completion_length": 243.3616180419922,
      "epoch": 2.5106668343182865,
      "grad_norm": 0.0040697662114411014,
      "kl": 0.123992919921875,
      "learning_rate": 4.914630771870457e-07,
      "loss": 0.0001,
      "reward": 1.7607143372297287,
      "reward_std": 0.015152287669479847,
      "rewards/equation_reward_func": 0.7607143260538578,
      "rewards/format_reward_func": 1.0,
      "step": 14974
    },
    {
      "completion_length": 233.06251049041748,
      "epoch": 2.511002137558154,
      "grad_norm": 0.2969220308731453,
      "kl": 0.117645263671875,
      "learning_rate": 4.914599688894208e-07,
      "loss": 0.0001,
      "reward": 1.7928572073578835,
      "reward_std": 0.06060915067791939,
      "rewards/equation_reward_func": 0.7928571663796902,
      "rewards/format_reward_func": 1.0,
      "step": 14976
    },
    {
      "completion_length": 244.55804443359375,
      "epoch": 2.511337440798022,
      "grad_norm": 0.14675272791289684,
      "kl": 0.14947509765625,
      "learning_rate": 4.914568600358656e-07,
      "loss": 0.0001,
      "reward": 1.7750000432133675,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7750000432133675,
      "rewards/format_reward_func": 1.0,
      "step": 14978
    },
    {
      "completion_length": 242.08929824829102,
      "epoch": 2.5116727440378894,
      "grad_norm": 0.23792415652608712,
      "kl": 0.123748779296875,
      "learning_rate": 4.914537506263871e-07,
      "loss": 0.0001,
      "reward": 1.7285715118050575,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7285714522004128,
      "rewards/format_reward_func": 1.0,
      "step": 14980
    },
    {
      "completion_length": 237.70983219146729,
      "epoch": 2.5120080472777566,
      "grad_norm": 0.16721352864248729,
      "kl": 0.1259307861328125,
      "learning_rate": 4.914506406609927e-07,
      "loss": 0.0001,
      "reward": 1.7732143476605415,
      "reward_std": 0.037880719639360905,
      "rewards/equation_reward_func": 0.7776786014437675,
      "rewards/format_reward_func": 0.9955357164144516,
      "step": 14982
    },
    {
      "completion_length": 242.0669755935669,
      "epoch": 2.5123433505176243,
      "grad_norm": 0.1793158228133712,
      "kl": 0.114105224609375,
      "learning_rate": 4.914475301396893e-07,
      "loss": 0.0001,
      "reward": 1.767857201397419,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7678571715950966,
      "rewards/format_reward_func": 1.0,
      "step": 14984
    },
    {
      "completion_length": 234.45090198516846,
      "epoch": 2.512678653757492,
      "grad_norm": 0.1808318129904359,
      "kl": 0.1202392578125,
      "learning_rate": 4.914444190624842e-07,
      "loss": 0.0001,
      "reward": 1.778571493923664,
      "reward_std": 0.02020305022597313,
      "rewards/equation_reward_func": 0.7785714492201805,
      "rewards/format_reward_func": 1.0,
      "step": 14986
    },
    {
      "completion_length": 232.07590198516846,
      "epoch": 2.5130139569973595,
      "grad_norm": 0.00458665470919745,
      "kl": 0.172943115234375,
      "learning_rate": 4.914413074293845e-07,
      "loss": 0.0002,
      "reward": 1.7642857655882835,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7642857395112514,
      "rewards/format_reward_func": 1.0,
      "step": 14988
    },
    {
      "completion_length": 241.29018783569336,
      "epoch": 2.513349260237227,
      "grad_norm": 0.21982994897106903,
      "kl": 0.13336181640625,
      "learning_rate": 4.914381952403974e-07,
      "loss": 0.0001,
      "reward": 1.8142857626080513,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8142857402563095,
      "rewards/format_reward_func": 1.0,
      "step": 14990
    },
    {
      "completion_length": 230.55358123779297,
      "epoch": 2.513684563477095,
      "grad_norm": 0.23499309856862302,
      "kl": 0.129669189453125,
      "learning_rate": 4.9143508249553e-07,
      "loss": 0.0001,
      "reward": 1.7535715028643608,
      "reward_std": 0.035355337895452976,
      "rewards/equation_reward_func": 0.7535714767873287,
      "rewards/format_reward_func": 1.0,
      "step": 14992
    },
    {
      "completion_length": 240.4241180419922,
      "epoch": 2.5140198667169624,
      "grad_norm": 0.15111986044046005,
      "kl": 0.165252685546875,
      "learning_rate": 4.914319691947896e-07,
      "loss": 0.0002,
      "reward": 1.807142898440361,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.8071428798139095,
      "rewards/format_reward_func": 1.0,
      "step": 14994
    },
    {
      "completion_length": 227.11608028411865,
      "epoch": 2.5143551699568296,
      "grad_norm": 0.13801908442240357,
      "kl": 0.121490478515625,
      "learning_rate": 4.914288553381833e-07,
      "loss": 0.0001,
      "reward": 1.782142922282219,
      "reward_std": 0.02525381278246641,
      "rewards/equation_reward_func": 0.7821428906172514,
      "rewards/format_reward_func": 1.0,
      "step": 14996
    },
    {
      "completion_length": 232.95090293884277,
      "epoch": 2.5146904731966973,
      "grad_norm": 0.13381548188941406,
      "kl": 0.12237548828125,
      "learning_rate": 4.914257409257182e-07,
      "loss": 0.0001,
      "reward": 1.8071429058909416,
      "reward_std": 0.05050762556493282,
      "rewards/equation_reward_func": 0.8071428798139095,
      "rewards/format_reward_func": 1.0,
      "step": 14998
    },
    {
      "completion_length": 230.81250953674316,
      "epoch": 2.515025776436565,
      "grad_norm": 0.003928641599324006,
      "kl": 0.128692626953125,
      "learning_rate": 4.914226259574015e-07,
      "loss": 0.0001,
      "reward": 1.7357143610715866,
      "reward_std": 0.030304575338959694,
      "rewards/equation_reward_func": 0.7357143014669418,
      "rewards/format_reward_func": 1.0,
      "step": 15000
    }
  ],
  "logging_steps": 2,
  "max_steps": 134996,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 23,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}