{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9965010496850945,
  "eval_steps": 100,
  "global_step": 178,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 840.6339149475098,
      "epoch": 0.005598320503848845,
      "grad_norm": 0.7309558391571045,
      "kl": 0.0,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4647472910583019,
      "reward_std": 0.2918772315606475,
      "rewards/random_math_reward": 0.4647472910583019,
      "step": 1
    },
    {
      "completion_length": 852.609676361084,
      "epoch": 0.01119664100769769,
      "grad_norm": 0.6802689433097839,
      "kl": 0.0002111196517944336,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.46338964626193047,
      "reward_std": 0.306376988068223,
      "rewards/random_math_reward": 0.46338964626193047,
      "step": 2
    },
    {
      "completion_length": 900.8494682312012,
      "epoch": 0.016794961511546535,
      "grad_norm": 0.11767072975635529,
      "kl": 0.00023311376571655273,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4775644950568676,
      "reward_std": 0.32149738259613514,
      "rewards/random_math_reward": 0.4775644950568676,
      "step": 3
    },
    {
      "completion_length": 851.5382461547852,
      "epoch": 0.02239328201539538,
      "grad_norm": 0.11348643153905869,
      "kl": 0.0002281665802001953,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48735272884368896,
      "reward_std": 0.30069673527032137,
      "rewards/random_math_reward": 0.48735272884368896,
      "step": 4
    },
    {
      "completion_length": 883.405590057373,
      "epoch": 0.02799160251924423,
      "grad_norm": 0.08939001709222794,
      "kl": 0.00023871660232543945,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4512794092297554,
      "reward_std": 0.27799729630351067,
      "rewards/random_math_reward": 0.4512794092297554,
      "step": 5
    },
    {
      "completion_length": 835.0280456542969,
      "epoch": 0.03358992302309307,
      "grad_norm": 0.27869054675102234,
      "kl": 0.0002474188804626465,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.47673217207193375,
      "reward_std": 0.29296134505420923,
      "rewards/random_math_reward": 0.47673217207193375,
      "step": 6
    },
    {
      "completion_length": 907.8456382751465,
      "epoch": 0.03918824352694192,
      "grad_norm": 0.862323522567749,
      "kl": 0.00039499998092651367,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.45936181023716927,
      "reward_std": 0.2918355893343687,
      "rewards/random_math_reward": 0.45936181023716927,
      "step": 7
    },
    {
      "completion_length": 859.1173267364502,
      "epoch": 0.04478656403079076,
      "grad_norm": 0.23682314157485962,
      "kl": 0.00037294626235961914,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5094553399831057,
      "reward_std": 0.28784568049013615,
      "rewards/random_math_reward": 0.5094553399831057,
      "step": 8
    },
    {
      "completion_length": 858.5267715454102,
      "epoch": 0.05038488453463961,
      "grad_norm": 0.07127245515584946,
      "kl": 0.00028765201568603516,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5017606746405363,
      "reward_std": 0.2951846132054925,
      "rewards/random_math_reward": 0.5017606746405363,
      "step": 9
    },
    {
      "completion_length": 877.473201751709,
      "epoch": 0.05598320503848846,
      "grad_norm": 0.032599493861198425,
      "kl": 0.00027692317962646484,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48044879734516144,
      "reward_std": 0.3181338291615248,
      "rewards/random_math_reward": 0.48044879734516144,
      "step": 10
    },
    {
      "completion_length": 870.3584022521973,
      "epoch": 0.0615815255423373,
      "grad_norm": 0.2817532420158386,
      "kl": 0.00034165382385253906,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4617793317884207,
      "reward_std": 0.28720600064843893,
      "rewards/random_math_reward": 0.4617793317884207,
      "step": 11
    },
    {
      "completion_length": 875.0089111328125,
      "epoch": 0.06717984604618614,
      "grad_norm": 0.04790099710226059,
      "kl": 0.0003007054328918457,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.44794472493231297,
      "reward_std": 0.27486683428287506,
      "rewards/random_math_reward": 0.44794472493231297,
      "step": 12
    },
    {
      "completion_length": 888.2474327087402,
      "epoch": 0.072778166550035,
      "grad_norm": 0.07171210646629333,
      "kl": 0.0003644227981567383,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4840338323265314,
      "reward_std": 0.2867116192355752,
      "rewards/random_math_reward": 0.4840338323265314,
      "step": 13
    },
    {
      "completion_length": 887.8545837402344,
      "epoch": 0.07837648705388384,
      "grad_norm": 0.02272365428507328,
      "kl": 0.00030791759490966797,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4695360567420721,
      "reward_std": 0.30686575919389725,
      "rewards/random_math_reward": 0.4695360567420721,
      "step": 14
    },
    {
      "completion_length": 894.8392677307129,
      "epoch": 0.08397480755773268,
      "grad_norm": 0.1592845916748047,
      "kl": 0.0007302761077880859,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5013854652643204,
      "reward_std": 0.30024987645447254,
      "rewards/random_math_reward": 0.5013854652643204,
      "step": 15
    },
    {
      "completion_length": 826.4298248291016,
      "epoch": 0.08957312806158152,
      "grad_norm": 0.11753757297992706,
      "kl": 0.00044095516204833984,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48069896548986435,
      "reward_std": 0.2729701278731227,
      "rewards/random_math_reward": 0.48069896548986435,
      "step": 16
    },
    {
      "completion_length": 902.1517715454102,
      "epoch": 0.09517144856543037,
      "grad_norm": 0.111512191593647,
      "kl": 0.0004519224166870117,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48635287396609783,
      "reward_std": 0.2879869397729635,
      "rewards/random_math_reward": 0.48635287396609783,
      "step": 17
    },
    {
      "completion_length": 893.360954284668,
      "epoch": 0.10076976906927922,
      "grad_norm": 0.2031000405550003,
      "kl": 0.0005553364753723145,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48062865249812603,
      "reward_std": 0.31339638121426105,
      "rewards/random_math_reward": 0.48062865249812603,
      "step": 18
    },
    {
      "completion_length": 838.4757537841797,
      "epoch": 0.10636808957312806,
      "grad_norm": 0.1047205924987793,
      "kl": 0.00042116641998291016,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4573891665786505,
      "reward_std": 0.283748428337276,
      "rewards/random_math_reward": 0.4573891665786505,
      "step": 19
    },
    {
      "completion_length": 853.5701332092285,
      "epoch": 0.11196641007697691,
      "grad_norm": 0.026682933792471886,
      "kl": 0.00039273500442504883,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5187842268496752,
      "reward_std": 0.2996611688286066,
      "rewards/random_math_reward": 0.5187842268496752,
      "step": 20
    },
    {
      "completion_length": 884.6606903076172,
      "epoch": 0.11756473058082575,
      "grad_norm": 0.07019887119531631,
      "kl": 0.0003604888916015625,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4664475191384554,
      "reward_std": 0.31808448024094105,
      "rewards/random_math_reward": 0.4664475191384554,
      "step": 21
    },
    {
      "completion_length": 906.0828857421875,
      "epoch": 0.1231630510846746,
      "grad_norm": 0.05280100554227829,
      "kl": 0.0005142688751220703,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4690139964222908,
      "reward_std": 0.27718855906277895,
      "rewards/random_math_reward": 0.4690139964222908,
      "step": 22
    },
    {
      "completion_length": 887.6466636657715,
      "epoch": 0.12876137158852344,
      "grad_norm": 0.04366298019886017,
      "kl": 0.0004309415817260742,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.43934059888124466,
      "reward_std": 0.27285193372517824,
      "rewards/random_math_reward": 0.43934059888124466,
      "step": 23
    },
    {
      "completion_length": 884.6441116333008,
      "epoch": 0.13435969209237228,
      "grad_norm": 0.24076521396636963,
      "kl": 0.0005651712417602539,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4910389892756939,
      "reward_std": 0.28644737135618925,
      "rewards/random_math_reward": 0.4910389892756939,
      "step": 24
    },
    {
      "completion_length": 853.1759986877441,
      "epoch": 0.13995801259622112,
      "grad_norm": 0.27850207686424255,
      "kl": 0.00044476985931396484,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4575974587351084,
      "reward_std": 0.30237106420099735,
      "rewards/random_math_reward": 0.4575974587351084,
      "step": 25
    },
    {
      "completion_length": 894.9030342102051,
      "epoch": 0.14555633310007,
      "grad_norm": 0.049030691385269165,
      "kl": 0.00048810243606567383,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4729502145200968,
      "reward_std": 0.285190143622458,
      "rewards/random_math_reward": 0.4729502145200968,
      "step": 26
    },
    {
      "completion_length": 854.3010063171387,
      "epoch": 0.15115465360391883,
      "grad_norm": 0.07516992837190628,
      "kl": 0.0005003213882446289,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4500812850892544,
      "reward_std": 0.32013510540127754,
      "rewards/random_math_reward": 0.4500812850892544,
      "step": 27
    },
    {
      "completion_length": 858.9030418395996,
      "epoch": 0.15675297410776767,
      "grad_norm": 0.20776700973510742,
      "kl": 0.0009773969650268555,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5247392375022173,
      "reward_std": 0.32408637553453445,
      "rewards/random_math_reward": 0.5247392375022173,
      "step": 28
    },
    {
      "completion_length": 844.9770317077637,
      "epoch": 0.16235129461161651,
      "grad_norm": 0.0495678074657917,
      "kl": 0.0009752511978149414,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.43325162678956985,
      "reward_std": 0.29721821192651987,
      "rewards/random_math_reward": 0.43325162678956985,
      "step": 29
    },
    {
      "completion_length": 908.6619758605957,
      "epoch": 0.16794961511546536,
      "grad_norm": 0.053116437047719955,
      "kl": 0.0007257461547851562,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.510849991813302,
      "reward_std": 0.2977478625252843,
      "rewards/random_math_reward": 0.510849991813302,
      "step": 30
    },
    {
      "completion_length": 895.2946281433105,
      "epoch": 0.1735479356193142,
      "grad_norm": 0.01952669955790043,
      "kl": 0.0004551410675048828,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4997572433203459,
      "reward_std": 0.2973903976380825,
      "rewards/random_math_reward": 0.4997572433203459,
      "step": 31
    },
    {
      "completion_length": 917.7333984375,
      "epoch": 0.17914625612316304,
      "grad_norm": 0.1707552671432495,
      "kl": 0.0006129741668701172,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.465914161875844,
      "reward_std": 0.30206145346164703,
      "rewards/random_math_reward": 0.465914161875844,
      "step": 32
    },
    {
      "completion_length": 870.7665557861328,
      "epoch": 0.1847445766270119,
      "grad_norm": 0.035584185272455215,
      "kl": 0.0004489421844482422,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.45434157736599445,
      "reward_std": 0.30155808478593826,
      "rewards/random_math_reward": 0.45434157736599445,
      "step": 33
    },
    {
      "completion_length": 873.6504936218262,
      "epoch": 0.19034289713086075,
      "grad_norm": 0.046750955283641815,
      "kl": 0.0005222558975219727,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.49102505296468735,
      "reward_std": 0.2892821617424488,
      "rewards/random_math_reward": 0.49102505296468735,
      "step": 34
    },
    {
      "completion_length": 885.6441230773926,
      "epoch": 0.1959412176347096,
      "grad_norm": 0.02094712294638157,
      "kl": 0.0004401206970214844,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.49038008227944374,
      "reward_std": 0.2766480268910527,
      "rewards/random_math_reward": 0.49038008227944374,
      "step": 35
    },
    {
      "completion_length": 857.9094200134277,
      "epoch": 0.20153953813855843,
      "grad_norm": 0.32986631989479065,
      "kl": 0.0016863346099853516,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5011246297508478,
      "reward_std": 0.2916059549897909,
      "rewards/random_math_reward": 0.5011246297508478,
      "step": 36
    },
    {
      "completion_length": 839.3214149475098,
      "epoch": 0.20713785864240727,
      "grad_norm": 0.04805277660489082,
      "kl": 0.0015134811401367188,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.47658345103263855,
      "reward_std": 0.26997081749141216,
      "rewards/random_math_reward": 0.47658345103263855,
      "step": 37
    },
    {
      "completion_length": 888.7257461547852,
      "epoch": 0.21273617914625612,
      "grad_norm": 0.026755526661872864,
      "kl": 0.0005905628204345703,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4430251754820347,
      "reward_std": 0.28255612775683403,
      "rewards/random_math_reward": 0.4430251754820347,
      "step": 38
    },
    {
      "completion_length": 879.1823692321777,
      "epoch": 0.21833449965010496,
      "grad_norm": 0.11245640367269516,
      "kl": 0.000995039939880371,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4720071740448475,
      "reward_std": 0.27762684039771557,
      "rewards/random_math_reward": 0.4720071740448475,
      "step": 39
    },
    {
      "completion_length": 930.487232208252,
      "epoch": 0.22393282015395383,
      "grad_norm": 0.053661637008190155,
      "kl": 0.0007688403129577637,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4835520200431347,
      "reward_std": 0.28671225160360336,
      "rewards/random_math_reward": 0.4835520200431347,
      "step": 40
    },
    {
      "completion_length": 893.3328971862793,
      "epoch": 0.22953114065780267,
      "grad_norm": 0.5499238967895508,
      "kl": 0.003184080123901367,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4463555943220854,
      "reward_std": 0.2914181677624583,
      "rewards/random_math_reward": 0.4463555943220854,
      "step": 41
    },
    {
      "completion_length": 898.3367156982422,
      "epoch": 0.2351294611616515,
      "grad_norm": 0.046797916293144226,
      "kl": 0.0005197525024414062,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.45641600526869297,
      "reward_std": 0.2792764212936163,
      "rewards/random_math_reward": 0.45641600526869297,
      "step": 42
    },
    {
      "completion_length": 845.0535507202148,
      "epoch": 0.24072778166550035,
      "grad_norm": 0.0507148802280426,
      "kl": 0.0009500980377197266,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4459559340029955,
      "reward_std": 0.29090235754847527,
      "rewards/random_math_reward": 0.4459559340029955,
      "step": 43
    },
    {
      "completion_length": 876.5892677307129,
      "epoch": 0.2463261021693492,
      "grad_norm": 0.04941030591726303,
      "kl": 0.000649571418762207,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4653678424656391,
      "reward_std": 0.28301908634603024,
      "rewards/random_math_reward": 0.4653678424656391,
      "step": 44
    },
    {
      "completion_length": 888.4846801757812,
      "epoch": 0.25192442267319803,
      "grad_norm": 0.026715300977230072,
      "kl": 0.0008295774459838867,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.45214173197746277,
      "reward_std": 0.3066752403974533,
      "rewards/random_math_reward": 0.45214173197746277,
      "step": 45
    },
    {
      "completion_length": 846.2231979370117,
      "epoch": 0.2575227431770469,
      "grad_norm": 0.024737877771258354,
      "kl": 0.0005326271057128906,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5093872379511595,
      "reward_std": 0.28799473866820335,
      "rewards/random_math_reward": 0.5093872379511595,
      "step": 46
    },
    {
      "completion_length": 865.0395202636719,
      "epoch": 0.2631210636808957,
      "grad_norm": 0.37511223554611206,
      "kl": 0.001316070556640625,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4726740214973688,
      "reward_std": 0.27686448488384485,
      "rewards/random_math_reward": 0.4726740214973688,
      "step": 47
    },
    {
      "completion_length": 884.132640838623,
      "epoch": 0.26871938418474456,
      "grad_norm": 0.03645370528101921,
      "kl": 0.0006043910980224609,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4929856266826391,
      "reward_std": 0.2844850402325392,
      "rewards/random_math_reward": 0.4929856266826391,
      "step": 48
    },
    {
      "completion_length": 941.7487030029297,
      "epoch": 0.2743177046885934,
      "grad_norm": 0.04691855236887932,
      "kl": 0.0006322860717773438,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.46845594607293606,
      "reward_std": 0.2635832289233804,
      "rewards/random_math_reward": 0.46845594607293606,
      "step": 49
    },
    {
      "completion_length": 886.7040634155273,
      "epoch": 0.27991602519244224,
      "grad_norm": 0.01480669155716896,
      "kl": 0.0007061958312988281,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.49290234968066216,
      "reward_std": 0.28575755935162306,
      "rewards/random_math_reward": 0.49290234968066216,
      "step": 50
    },
    {
      "completion_length": 881.2589111328125,
      "epoch": 0.28551434569629114,
      "grad_norm": 0.18569053709506989,
      "kl": 0.0008254051208496094,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.470220735296607,
      "reward_std": 0.3022102378308773,
      "rewards/random_math_reward": 0.470220735296607,
      "step": 51
    },
    {
      "completion_length": 858.8775253295898,
      "epoch": 0.29111266620014,
      "grad_norm": 0.06353636085987091,
      "kl": 0.0007264614105224609,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5083950478583574,
      "reward_std": 0.29449230805039406,
      "rewards/random_math_reward": 0.5083950478583574,
      "step": 52
    },
    {
      "completion_length": 886.2869720458984,
      "epoch": 0.2967109867039888,
      "grad_norm": 0.035578541457653046,
      "kl": 0.0007152557373046875,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4716050550341606,
      "reward_std": 0.30308040603995323,
      "rewards/random_math_reward": 0.4716050550341606,
      "step": 53
    },
    {
      "completion_length": 889.1351852416992,
      "epoch": 0.30230930720783766,
      "grad_norm": 0.027262764051556587,
      "kl": 0.0015797615051269531,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.49637374840676785,
      "reward_std": 0.30440937727689743,
      "rewards/random_math_reward": 0.49637374840676785,
      "step": 54
    },
    {
      "completion_length": 859.5165557861328,
      "epoch": 0.3079076277116865,
      "grad_norm": 0.013877746649086475,
      "kl": 0.0005941390991210938,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4874963089823723,
      "reward_std": 0.2997851762920618,
      "rewards/random_math_reward": 0.4874963089823723,
      "step": 55
    },
    {
      "completion_length": 886.5854339599609,
      "epoch": 0.31350594821553535,
      "grad_norm": 0.024866046383976936,
      "kl": 0.000647425651550293,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.46163152530789375,
      "reward_std": 0.2887133536860347,
      "rewards/random_math_reward": 0.46163152530789375,
      "step": 56
    },
    {
      "completion_length": 863.1759986877441,
      "epoch": 0.3191042687193842,
      "grad_norm": 0.023259082809090614,
      "kl": 0.0006357431411743164,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4782671481370926,
      "reward_std": 0.29867998976260424,
      "rewards/random_math_reward": 0.4782671481370926,
      "step": 57
    },
    {
      "completion_length": 895.409423828125,
      "epoch": 0.32470258922323303,
      "grad_norm": 0.02394162304699421,
      "kl": 0.0007455348968505859,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4420376904308796,
      "reward_std": 0.30142530612647533,
      "rewards/random_math_reward": 0.4420376904308796,
      "step": 58
    },
    {
      "completion_length": 933.6810913085938,
      "epoch": 0.33030090972708187,
      "grad_norm": 0.022286556661128998,
      "kl": 0.0007117986679077148,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48266960494220257,
      "reward_std": 0.2849938729777932,
      "rewards/random_math_reward": 0.48266960494220257,
      "step": 59
    },
    {
      "completion_length": 906.8890190124512,
      "epoch": 0.3358992302309307,
      "grad_norm": 0.018822669982910156,
      "kl": 0.000682830810546875,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.487277016043663,
      "reward_std": 0.27959814574569464,
      "rewards/random_math_reward": 0.487277016043663,
      "step": 60
    },
    {
      "completion_length": 864.5459022521973,
      "epoch": 0.34149755073477955,
      "grad_norm": 0.07483907043933868,
      "kl": 0.0009076595306396484,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5487107969820499,
      "reward_std": 0.29574476182460785,
      "rewards/random_math_reward": 0.5487107969820499,
      "step": 61
    },
    {
      "completion_length": 946.0178413391113,
      "epoch": 0.3470958712386284,
      "grad_norm": 0.04614647850394249,
      "kl": 0.0015153884887695312,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5009770151227713,
      "reward_std": 0.29892334900796413,
      "rewards/random_math_reward": 0.5009770151227713,
      "step": 62
    },
    {
      "completion_length": 929.1989631652832,
      "epoch": 0.35269419174247724,
      "grad_norm": 0.02183196134865284,
      "kl": 0.0007126331329345703,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4849752429872751,
      "reward_std": 0.29463311191648245,
      "rewards/random_math_reward": 0.4849752429872751,
      "step": 63
    },
    {
      "completion_length": 911.9004974365234,
      "epoch": 0.3582925122463261,
      "grad_norm": 0.02410600334405899,
      "kl": 0.0010197162628173828,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4673476442694664,
      "reward_std": 0.27815896179527044,
      "rewards/random_math_reward": 0.4673476442694664,
      "step": 64
    },
    {
      "completion_length": 885.8609504699707,
      "epoch": 0.363890832750175,
      "grad_norm": 0.013846021145582199,
      "kl": 0.0008647441864013672,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.47511172853410244,
      "reward_std": 0.29065939225256443,
      "rewards/random_math_reward": 0.47511172853410244,
      "step": 65
    },
    {
      "completion_length": 896.8316116333008,
      "epoch": 0.3694891532540238,
      "grad_norm": 0.058244168758392334,
      "kl": 0.0007791519165039062,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.49584819190204144,
      "reward_std": 0.28557164780795574,
      "rewards/random_math_reward": 0.49584819190204144,
      "step": 66
    },
    {
      "completion_length": 915.7550888061523,
      "epoch": 0.37508747375787266,
      "grad_norm": 0.026778079569339752,
      "kl": 0.0007394552230834961,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4666522778570652,
      "reward_std": 0.2651451360434294,
      "rewards/random_math_reward": 0.4666522778570652,
      "step": 67
    },
    {
      "completion_length": 817.3571281433105,
      "epoch": 0.3806857942617215,
      "grad_norm": 0.04487188532948494,
      "kl": 0.002661585807800293,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4737847317010164,
      "reward_std": 0.2716279961168766,
      "rewards/random_math_reward": 0.4737847317010164,
      "step": 68
    },
    {
      "completion_length": 967.4578742980957,
      "epoch": 0.38628411476557034,
      "grad_norm": 0.012447608634829521,
      "kl": 0.000609278678894043,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4469184931367636,
      "reward_std": 0.2811002554371953,
      "rewards/random_math_reward": 0.4469184931367636,
      "step": 69
    },
    {
      "completion_length": 932.1237030029297,
      "epoch": 0.3918824352694192,
      "grad_norm": 0.011375661008059978,
      "kl": 0.0006630420684814453,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.458757933229208,
      "reward_std": 0.3028138969093561,
      "rewards/random_math_reward": 0.458757933229208,
      "step": 70
    },
    {
      "completion_length": 871.3660583496094,
      "epoch": 0.397480755773268,
      "grad_norm": 0.010783006437122822,
      "kl": 0.0007634162902832031,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4868381731212139,
      "reward_std": 0.28059179708361626,
      "rewards/random_math_reward": 0.4868381731212139,
      "step": 71
    },
    {
      "completion_length": 880.8915634155273,
      "epoch": 0.40307907627711687,
      "grad_norm": 0.024310071021318436,
      "kl": 0.0012193918228149414,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4670300707221031,
      "reward_std": 0.28627045452594757,
      "rewards/random_math_reward": 0.4670300707221031,
      "step": 72
    },
    {
      "completion_length": 848.394115447998,
      "epoch": 0.4086773967809657,
      "grad_norm": 0.03577808663249016,
      "kl": 0.0011832714080810547,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4628731273114681,
      "reward_std": 0.2926492039114237,
      "rewards/random_math_reward": 0.4628731273114681,
      "step": 73
    },
    {
      "completion_length": 850.6415634155273,
      "epoch": 0.41427571728481455,
      "grad_norm": 0.02307233400642872,
      "kl": 0.0011417865753173828,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48236627131700516,
      "reward_std": 0.2867990182712674,
      "rewards/random_math_reward": 0.48236627131700516,
      "step": 74
    },
    {
      "completion_length": 935.357120513916,
      "epoch": 0.4198740377886634,
      "grad_norm": 0.009285389445722103,
      "kl": 0.0006787776947021484,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48846207931637764,
      "reward_std": 0.28832482267171144,
      "rewards/random_math_reward": 0.48846207931637764,
      "step": 75
    },
    {
      "completion_length": 897.8532943725586,
      "epoch": 0.42547235829251223,
      "grad_norm": 0.06958639621734619,
      "kl": 0.0009503364562988281,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5029652137309313,
      "reward_std": 0.2956245709210634,
      "rewards/random_math_reward": 0.5029652137309313,
      "step": 76
    },
    {
      "completion_length": 861.1275329589844,
      "epoch": 0.4310706787963611,
      "grad_norm": 0.019943566992878914,
      "kl": 0.001253366470336914,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5312744900584221,
      "reward_std": 0.2928897039964795,
      "rewards/random_math_reward": 0.5312744900584221,
      "step": 77
    },
    {
      "completion_length": 924.6313591003418,
      "epoch": 0.4366689993002099,
      "grad_norm": 0.01711793802678585,
      "kl": 0.0007042884826660156,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.49481588415801525,
      "reward_std": 0.29031859524548054,
      "rewards/random_math_reward": 0.49481588415801525,
      "step": 78
    },
    {
      "completion_length": 842.0395202636719,
      "epoch": 0.44226731980405876,
      "grad_norm": 0.018647639080882072,
      "kl": 0.0007382631301879883,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5070081520825624,
      "reward_std": 0.2859943201765418,
      "rewards/random_math_reward": 0.5070081520825624,
      "step": 79
    },
    {
      "completion_length": 895.3583946228027,
      "epoch": 0.44786564030790765,
      "grad_norm": 0.026943529024720192,
      "kl": 0.0007255077362060547,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48258678428828716,
      "reward_std": 0.2774320160970092,
      "rewards/random_math_reward": 0.48258678428828716,
      "step": 80
    },
    {
      "completion_length": 937.7754783630371,
      "epoch": 0.4534639608117565,
      "grad_norm": 0.015870269387960434,
      "kl": 0.0008311271667480469,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4687965828925371,
      "reward_std": 0.2817834969609976,
      "rewards/random_math_reward": 0.4687965828925371,
      "step": 81
    },
    {
      "completion_length": 835.7397804260254,
      "epoch": 0.45906228131560534,
      "grad_norm": 0.06118820607662201,
      "kl": 0.0011771917343139648,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4703396111726761,
      "reward_std": 0.29155910573899746,
      "rewards/random_math_reward": 0.4703396111726761,
      "step": 82
    },
    {
      "completion_length": 887.909423828125,
      "epoch": 0.4646606018194542,
      "grad_norm": 0.05378980189561844,
      "kl": 0.0013784170150756836,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4672347716987133,
      "reward_std": 0.316370103508234,
      "rewards/random_math_reward": 0.4672347716987133,
      "step": 83
    },
    {
      "completion_length": 818.6708984375,
      "epoch": 0.470258922323303,
      "grad_norm": 0.05784986913204193,
      "kl": 0.0008500814437866211,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5053709652274847,
      "reward_std": 0.3026046808809042,
      "rewards/random_math_reward": 0.5053709652274847,
      "step": 84
    },
    {
      "completion_length": 926.3800888061523,
      "epoch": 0.47585724282715186,
      "grad_norm": 0.021458175033330917,
      "kl": 0.0008251667022705078,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.47099834494292736,
      "reward_std": 0.30302479304373264,
      "rewards/random_math_reward": 0.47099834494292736,
      "step": 85
    },
    {
      "completion_length": 859.1288108825684,
      "epoch": 0.4814555633310007,
      "grad_norm": 0.019605984911322594,
      "kl": 0.0009077787399291992,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4812137298285961,
      "reward_std": 0.29327244497835636,
      "rewards/random_math_reward": 0.4812137298285961,
      "step": 86
    },
    {
      "completion_length": 908.1402816772461,
      "epoch": 0.48705388383484954,
      "grad_norm": 0.06367423385381699,
      "kl": 0.0011855363845825195,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4995248857885599,
      "reward_std": 0.293773477897048,
      "rewards/random_math_reward": 0.4995248857885599,
      "step": 87
    },
    {
      "completion_length": 879.8099327087402,
      "epoch": 0.4926522043386984,
      "grad_norm": 0.013439509086310863,
      "kl": 0.0009884834289550781,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4674531724303961,
      "reward_std": 0.28925504721701145,
      "rewards/random_math_reward": 0.4674531724303961,
      "step": 88
    },
    {
      "completion_length": 897.298454284668,
      "epoch": 0.4982505248425472,
      "grad_norm": 0.011524248868227005,
      "kl": 0.0013021230697631836,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.46935696713626385,
      "reward_std": 0.28805949725210667,
      "rewards/random_math_reward": 0.46935696713626385,
      "step": 89
    },
    {
      "completion_length": 862.3762626647949,
      "epoch": 0.5038488453463961,
      "grad_norm": 0.01074186060577631,
      "kl": 0.0008319616317749023,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4534935001283884,
      "reward_std": 0.29410699382424355,
      "rewards/random_math_reward": 0.4534935001283884,
      "step": 90
    },
    {
      "completion_length": 903.1083946228027,
      "epoch": 0.509447165850245,
      "grad_norm": 0.20411504805088043,
      "kl": 0.0014219284057617188,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.47694382816553116,
      "reward_std": 0.28339869249612093,
      "rewards/random_math_reward": 0.47694382816553116,
      "step": 91
    },
    {
      "completion_length": 922.6543235778809,
      "epoch": 0.5150454863540938,
      "grad_norm": 0.01232316717505455,
      "kl": 0.000966191291809082,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.44490973837673664,
      "reward_std": 0.2793145142495632,
      "rewards/random_math_reward": 0.44490973837673664,
      "step": 92
    },
    {
      "completion_length": 911.021671295166,
      "epoch": 0.5206438068579426,
      "grad_norm": 0.012167483568191528,
      "kl": 0.0007963180541992188,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48554741591215134,
      "reward_std": 0.2934918478131294,
      "rewards/random_math_reward": 0.48554741591215134,
      "step": 93
    },
    {
      "completion_length": 939.8800773620605,
      "epoch": 0.5262421273617914,
      "grad_norm": 0.009945104829967022,
      "kl": 0.0008685588836669922,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.47484579868614674,
      "reward_std": 0.30168304964900017,
      "rewards/random_math_reward": 0.47484579868614674,
      "step": 94
    },
    {
      "completion_length": 906.4119720458984,
      "epoch": 0.5318404478656403,
      "grad_norm": 0.010836634784936905,
      "kl": 0.0007404088973999023,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.47092380560934544,
      "reward_std": 0.28657434694468975,
      "rewards/random_math_reward": 0.47092380560934544,
      "step": 95
    },
    {
      "completion_length": 883.8583946228027,
      "epoch": 0.5374387683694891,
      "grad_norm": 0.035804830491542816,
      "kl": 0.0007826089859008789,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4703601971268654,
      "reward_std": 0.28724440187215805,
      "rewards/random_math_reward": 0.4703601971268654,
      "step": 96
    },
    {
      "completion_length": 880.0191192626953,
      "epoch": 0.543037088873338,
      "grad_norm": 0.06715273857116699,
      "kl": 0.0016443729400634766,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5048227999359369,
      "reward_std": 0.29030876979231834,
      "rewards/random_math_reward": 0.5048227999359369,
      "step": 97
    },
    {
      "completion_length": 898.9642677307129,
      "epoch": 0.5486354093771868,
      "grad_norm": 0.014051802456378937,
      "kl": 0.001214146614074707,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.49155336059629917,
      "reward_std": 0.2932043820619583,
      "rewards/random_math_reward": 0.49155336059629917,
      "step": 98
    },
    {
      "completion_length": 854.6913032531738,
      "epoch": 0.5542337298810357,
      "grad_norm": 0.01157184038311243,
      "kl": 0.0010590553283691406,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.45419391617178917,
      "reward_std": 0.28226044587790966,
      "rewards/random_math_reward": 0.45419391617178917,
      "step": 99
    },
    {
      "completion_length": 902.7282943725586,
      "epoch": 0.5598320503848845,
      "grad_norm": 0.11193796992301941,
      "kl": 0.0011587142944335938,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.46381357870996,
      "reward_std": 0.27888650726526976,
      "rewards/random_math_reward": 0.46381357870996,
      "step": 100
    },
    {
      "completion_length": 873.6173210144043,
      "epoch": 0.5654303708887334,
      "grad_norm": 0.02095395140349865,
      "kl": 0.0009194612503051758,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4897339139133692,
      "reward_std": 0.27832474932074547,
      "rewards/random_math_reward": 0.4897339139133692,
      "step": 101
    },
    {
      "completion_length": 914.6849250793457,
      "epoch": 0.5710286913925823,
      "grad_norm": 0.04229766130447388,
      "kl": 0.0011234283447265625,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48241828568279743,
      "reward_std": 0.2899211458861828,
      "rewards/random_math_reward": 0.48241828568279743,
      "step": 102
    },
    {
      "completion_length": 888.3277778625488,
      "epoch": 0.5766270118964311,
      "grad_norm": 0.0680854320526123,
      "kl": 0.0012063980102539062,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4864093214273453,
      "reward_std": 0.2803313685581088,
      "rewards/random_math_reward": 0.4864093214273453,
      "step": 103
    },
    {
      "completion_length": 930.9400291442871,
      "epoch": 0.58222533240028,
      "grad_norm": 0.02672196552157402,
      "kl": 0.0008995532989501953,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.47334628365933895,
      "reward_std": 0.2828089501708746,
      "rewards/random_math_reward": 0.47334628365933895,
      "step": 104
    },
    {
      "completion_length": 865.7499847412109,
      "epoch": 0.5878236529041287,
      "grad_norm": 0.010435185395181179,
      "kl": 0.0008490085601806641,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4868275187909603,
      "reward_std": 0.2970134112983942,
      "rewards/random_math_reward": 0.4868275187909603,
      "step": 105
    },
    {
      "completion_length": 860.4770202636719,
      "epoch": 0.5934219734079776,
      "grad_norm": 0.023035289719700813,
      "kl": 0.0009222030639648438,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4827886149287224,
      "reward_std": 0.2927941419184208,
      "rewards/random_math_reward": 0.4827886149287224,
      "step": 106
    },
    {
      "completion_length": 926.4259948730469,
      "epoch": 0.5990202939118264,
      "grad_norm": 0.01681309938430786,
      "kl": 0.0008640289306640625,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.46561466343700886,
      "reward_std": 0.2794105224311352,
      "rewards/random_math_reward": 0.46561466343700886,
      "step": 107
    },
    {
      "completion_length": 899.8622245788574,
      "epoch": 0.6046186144156753,
      "grad_norm": 0.010567005723714828,
      "kl": 0.0010676383972167969,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4655596222728491,
      "reward_std": 0.27886725403368473,
      "rewards/random_math_reward": 0.4655596222728491,
      "step": 108
    },
    {
      "completion_length": 939.2283058166504,
      "epoch": 0.6102169349195241,
      "grad_norm": 0.07191047817468643,
      "kl": 0.0011771917343139648,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4728291556239128,
      "reward_std": 0.2772167157381773,
      "rewards/random_math_reward": 0.4728291556239128,
      "step": 109
    },
    {
      "completion_length": 825.8698749542236,
      "epoch": 0.615815255423373,
      "grad_norm": 0.02628973312675953,
      "kl": 0.0011527538299560547,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48204794339835644,
      "reward_std": 0.29616999346762896,
      "rewards/random_math_reward": 0.48204794339835644,
      "step": 110
    },
    {
      "completion_length": 907.2818717956543,
      "epoch": 0.6214135759272218,
      "grad_norm": 0.01321073155850172,
      "kl": 0.0007474422454833984,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4843390993773937,
      "reward_std": 0.2992616593837738,
      "rewards/random_math_reward": 0.4843390993773937,
      "step": 111
    },
    {
      "completion_length": 881.9884948730469,
      "epoch": 0.6270118964310707,
      "grad_norm": 0.011079943738877773,
      "kl": 0.0010025501251220703,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5278591625392437,
      "reward_std": 0.3031477089971304,
      "rewards/random_math_reward": 0.5278591625392437,
      "step": 112
    },
    {
      "completion_length": 889.1198768615723,
      "epoch": 0.6326102169349195,
      "grad_norm": 0.015584494918584824,
      "kl": 0.0008515119552612305,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5340252239257097,
      "reward_std": 0.28527406230568886,
      "rewards/random_math_reward": 0.5340252239257097,
      "step": 113
    },
    {
      "completion_length": 890.235954284668,
      "epoch": 0.6382085374387684,
      "grad_norm": 0.03280128538608551,
      "kl": 0.000990152359008789,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5209550634026527,
      "reward_std": 0.29494220949709415,
      "rewards/random_math_reward": 0.5209550634026527,
      "step": 114
    },
    {
      "completion_length": 865.9170799255371,
      "epoch": 0.6438068579426172,
      "grad_norm": 0.059104159474372864,
      "kl": 0.0009453296661376953,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5109865833073854,
      "reward_std": 0.29917749017477036,
      "rewards/random_math_reward": 0.5109865833073854,
      "step": 115
    },
    {
      "completion_length": 876.934928894043,
      "epoch": 0.6494051784464661,
      "grad_norm": 0.01903529092669487,
      "kl": 0.0008587837219238281,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.500015264376998,
      "reward_std": 0.29237030632793903,
      "rewards/random_math_reward": 0.500015264376998,
      "step": 116
    },
    {
      "completion_length": 846.1517677307129,
      "epoch": 0.655003498950315,
      "grad_norm": 0.02293042466044426,
      "kl": 0.0009784698486328125,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48975357227027416,
      "reward_std": 0.2920750202611089,
      "rewards/random_math_reward": 0.48975357227027416,
      "step": 117
    },
    {
      "completion_length": 889.9859504699707,
      "epoch": 0.6606018194541637,
      "grad_norm": 0.018028084188699722,
      "kl": 0.0013837814331054688,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5070291068404913,
      "reward_std": 0.2938714809715748,
      "rewards/random_math_reward": 0.5070291068404913,
      "step": 118
    },
    {
      "completion_length": 891.4043159484863,
      "epoch": 0.6662001399580126,
      "grad_norm": 0.011308281682431698,
      "kl": 0.0008604526519775391,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4764967616647482,
      "reward_std": 0.3148739319294691,
      "rewards/random_math_reward": 0.4764967616647482,
      "step": 119
    },
    {
      "completion_length": 872.5395317077637,
      "epoch": 0.6717984604618614,
      "grad_norm": 0.0414012111723423,
      "kl": 0.0009286403656005859,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4626141209155321,
      "reward_std": 0.3029380030930042,
      "rewards/random_math_reward": 0.4626141209155321,
      "step": 120
    },
    {
      "completion_length": 907.8239593505859,
      "epoch": 0.6773967809657103,
      "grad_norm": 0.024403268471360207,
      "kl": 0.0008714199066162109,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.463301295414567,
      "reward_std": 0.2877810364589095,
      "rewards/random_math_reward": 0.463301295414567,
      "step": 121
    },
    {
      "completion_length": 947.8214149475098,
      "epoch": 0.6829951014695591,
      "grad_norm": 0.03146574646234512,
      "kl": 0.0012885332107543945,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.46076902747154236,
      "reward_std": 0.2793724099174142,
      "rewards/random_math_reward": 0.46076902747154236,
      "step": 122
    },
    {
      "completion_length": 909.1071243286133,
      "epoch": 0.688593421973408,
      "grad_norm": 0.01003959309309721,
      "kl": 0.0011519193649291992,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5079296790063381,
      "reward_std": 0.29765503481030464,
      "rewards/random_math_reward": 0.5079296790063381,
      "step": 123
    },
    {
      "completion_length": 861.1160545349121,
      "epoch": 0.6941917424772568,
      "grad_norm": 0.013797848485410213,
      "kl": 0.0010790824890136719,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4844306465238333,
      "reward_std": 0.2945317914709449,
      "rewards/random_math_reward": 0.4844306465238333,
      "step": 124
    },
    {
      "completion_length": 863.8520278930664,
      "epoch": 0.6997900629811057,
      "grad_norm": 0.012310285121202469,
      "kl": 0.0011372566223144531,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4732836168259382,
      "reward_std": 0.2942813113331795,
      "rewards/random_math_reward": 0.4732836168259382,
      "step": 125
    },
    {
      "completion_length": 881.832893371582,
      "epoch": 0.7053883834849545,
      "grad_norm": 0.28292930126190186,
      "kl": 0.0030515193939208984,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48056939989328384,
      "reward_std": 0.27455357648432255,
      "rewards/random_math_reward": 0.48056939989328384,
      "step": 126
    },
    {
      "completion_length": 855.8303375244141,
      "epoch": 0.7109867039888034,
      "grad_norm": 0.026784956455230713,
      "kl": 0.0010666847229003906,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.47633388079702854,
      "reward_std": 0.27308547869324684,
      "rewards/random_math_reward": 0.47633388079702854,
      "step": 127
    },
    {
      "completion_length": 872.3571243286133,
      "epoch": 0.7165850244926522,
      "grad_norm": 0.027628231793642044,
      "kl": 0.0012966394424438477,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4544145315885544,
      "reward_std": 0.2810182133689523,
      "rewards/random_math_reward": 0.4544145315885544,
      "step": 128
    },
    {
      "completion_length": 880.8622207641602,
      "epoch": 0.722183344996501,
      "grad_norm": 0.011807740665972233,
      "kl": 0.0009188652038574219,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4832920003682375,
      "reward_std": 0.2904686816036701,
      "rewards/random_math_reward": 0.4832920003682375,
      "step": 129
    },
    {
      "completion_length": 890.7397804260254,
      "epoch": 0.72778166550035,
      "grad_norm": 0.00990369077771902,
      "kl": 0.0008499622344970703,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4763722326606512,
      "reward_std": 0.3048167824745178,
      "rewards/random_math_reward": 0.4763722326606512,
      "step": 130
    },
    {
      "completion_length": 881.3290634155273,
      "epoch": 0.7333799860041987,
      "grad_norm": 0.042855340987443924,
      "kl": 0.002415895462036133,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.49098371155560017,
      "reward_std": 0.28425728902220726,
      "rewards/random_math_reward": 0.49098371155560017,
      "step": 131
    },
    {
      "completion_length": 920.1415672302246,
      "epoch": 0.7389783065080476,
      "grad_norm": 0.011931121349334717,
      "kl": 0.0009909868240356445,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.49408636428415775,
      "reward_std": 0.30311523005366325,
      "rewards/random_math_reward": 0.49408636428415775,
      "step": 132
    },
    {
      "completion_length": 928.8456420898438,
      "epoch": 0.7445766270118964,
      "grad_norm": 0.11891764402389526,
      "kl": 0.0010557174682617188,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.47669179551303387,
      "reward_std": 0.29571292363107204,
      "rewards/random_math_reward": 0.47669179551303387,
      "step": 133
    },
    {
      "completion_length": 840.7448806762695,
      "epoch": 0.7501749475157453,
      "grad_norm": 0.020906388759613037,
      "kl": 0.001192331314086914,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48422789201140404,
      "reward_std": 0.28880419582128525,
      "rewards/random_math_reward": 0.48422789201140404,
      "step": 134
    },
    {
      "completion_length": 924.7793121337891,
      "epoch": 0.7557732680195941,
      "grad_norm": 0.015085350722074509,
      "kl": 0.0009965896606445312,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48809418082237244,
      "reward_std": 0.30123256146907806,
      "rewards/random_math_reward": 0.48809418082237244,
      "step": 135
    },
    {
      "completion_length": 783.3507499694824,
      "epoch": 0.761371588523443,
      "grad_norm": 0.025510285049676895,
      "kl": 0.0011339187622070312,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4794926680624485,
      "reward_std": 0.29374638944864273,
      "rewards/random_math_reward": 0.4794926680624485,
      "step": 136
    },
    {
      "completion_length": 886.570140838623,
      "epoch": 0.7669699090272918,
      "grad_norm": 0.014409742318093777,
      "kl": 0.0010912418365478516,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.44775343872606754,
      "reward_std": 0.27853819355368614,
      "rewards/random_math_reward": 0.44775343872606754,
      "step": 137
    },
    {
      "completion_length": 875.3928413391113,
      "epoch": 0.7725682295311407,
      "grad_norm": 0.009967821650207043,
      "kl": 0.001146078109741211,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4673925694078207,
      "reward_std": 0.289134263060987,
      "rewards/random_math_reward": 0.4673925694078207,
      "step": 138
    },
    {
      "completion_length": 856.2155380249023,
      "epoch": 0.7781665500349895,
      "grad_norm": 0.012912734411656857,
      "kl": 0.0009341239929199219,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5017714705318213,
      "reward_std": 0.28041696455329657,
      "rewards/random_math_reward": 0.5017714705318213,
      "step": 139
    },
    {
      "completion_length": 894.0050811767578,
      "epoch": 0.7837648705388384,
      "grad_norm": 0.08305728435516357,
      "kl": 0.0018918514251708984,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5223593860864639,
      "reward_std": 0.2958631496876478,
      "rewards/random_math_reward": 0.5223593860864639,
      "step": 140
    },
    {
      "completion_length": 881.410701751709,
      "epoch": 0.7893631910426872,
      "grad_norm": 0.009809349663555622,
      "kl": 0.0009772777557373047,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5037720259279013,
      "reward_std": 0.2979677114635706,
      "rewards/random_math_reward": 0.5037720259279013,
      "step": 141
    },
    {
      "completion_length": 840.1135063171387,
      "epoch": 0.794961511546536,
      "grad_norm": 0.018646089360117912,
      "kl": 0.001020193099975586,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5081987045705318,
      "reward_std": 0.29782584123313427,
      "rewards/random_math_reward": 0.5081987045705318,
      "step": 142
    },
    {
      "completion_length": 876.1058502197266,
      "epoch": 0.8005598320503848,
      "grad_norm": 0.024378400295972824,
      "kl": 0.0011475086212158203,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4988156743347645,
      "reward_std": 0.29589168168604374,
      "rewards/random_math_reward": 0.4988156743347645,
      "step": 143
    },
    {
      "completion_length": 891.9885063171387,
      "epoch": 0.8061581525542337,
      "grad_norm": 0.017726967111229897,
      "kl": 0.0011355876922607422,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.49985454976558685,
      "reward_std": 0.2889725724235177,
      "rewards/random_math_reward": 0.49985454976558685,
      "step": 144
    },
    {
      "completion_length": 900.8316192626953,
      "epoch": 0.8117564730580826,
      "grad_norm": 0.01951918564736843,
      "kl": 0.0014090538024902344,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48107675835490227,
      "reward_std": 0.2852372843772173,
      "rewards/random_math_reward": 0.48107675835490227,
      "step": 145
    },
    {
      "completion_length": 899.4731903076172,
      "epoch": 0.8173547935619314,
      "grad_norm": 0.021214712411165237,
      "kl": 0.0010182857513427734,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4994645491242409,
      "reward_std": 0.2752206530421972,
      "rewards/random_math_reward": 0.4994645491242409,
      "step": 146
    },
    {
      "completion_length": 949.8775329589844,
      "epoch": 0.8229531140657803,
      "grad_norm": 0.010702384635806084,
      "kl": 0.0011107921600341797,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.47683133371174335,
      "reward_std": 0.2850389126688242,
      "rewards/random_math_reward": 0.47683133371174335,
      "step": 147
    },
    {
      "completion_length": 890.5433540344238,
      "epoch": 0.8285514345696291,
      "grad_norm": 0.014014041982591152,
      "kl": 0.0009796619415283203,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.45283058658242226,
      "reward_std": 0.2967394981533289,
      "rewards/random_math_reward": 0.45283058658242226,
      "step": 148
    },
    {
      "completion_length": 854.0229377746582,
      "epoch": 0.834149755073478,
      "grad_norm": 0.010618280619382858,
      "kl": 0.0011134147644042969,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.46891826018691063,
      "reward_std": 0.2850739639252424,
      "rewards/random_math_reward": 0.46891826018691063,
      "step": 149
    },
    {
      "completion_length": 873.9553413391113,
      "epoch": 0.8397480755773268,
      "grad_norm": 0.03690403327345848,
      "kl": 0.0010218620300292969,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5085400156676769,
      "reward_std": 0.2830808274447918,
      "rewards/random_math_reward": 0.5085400156676769,
      "step": 150
    },
    {
      "completion_length": 927.468090057373,
      "epoch": 0.8453463960811757,
      "grad_norm": 0.017428407445549965,
      "kl": 0.000989675521850586,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4990678243339062,
      "reward_std": 0.2827162565663457,
      "rewards/random_math_reward": 0.4990678243339062,
      "step": 151
    },
    {
      "completion_length": 910.8379859924316,
      "epoch": 0.8509447165850245,
      "grad_norm": 0.017853064462542534,
      "kl": 0.0009360313415527344,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.46421765722334385,
      "reward_std": 0.2790751438587904,
      "rewards/random_math_reward": 0.46421765722334385,
      "step": 152
    },
    {
      "completion_length": 886.9323768615723,
      "epoch": 0.8565430370888734,
      "grad_norm": 0.018860990181565285,
      "kl": 0.0011188983917236328,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.47205423563718796,
      "reward_std": 0.28594095539301634,
      "rewards/random_math_reward": 0.47205423563718796,
      "step": 153
    },
    {
      "completion_length": 848.5089149475098,
      "epoch": 0.8621413575927221,
      "grad_norm": 0.020478500053286552,
      "kl": 0.0010101795196533203,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4954501297324896,
      "reward_std": 0.28822094202041626,
      "rewards/random_math_reward": 0.4954501297324896,
      "step": 154
    },
    {
      "completion_length": 873.7601776123047,
      "epoch": 0.867739678096571,
      "grad_norm": 0.014679434709250927,
      "kl": 0.001070261001586914,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48937808722257614,
      "reward_std": 0.28821886610239744,
      "rewards/random_math_reward": 0.48937808722257614,
      "step": 155
    },
    {
      "completion_length": 887.0267677307129,
      "epoch": 0.8733379986004198,
      "grad_norm": 0.053716812282800674,
      "kl": 0.0013840198516845703,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48024408891797066,
      "reward_std": 0.30272081680595875,
      "rewards/random_math_reward": 0.48024408891797066,
      "step": 156
    },
    {
      "completion_length": 910.788257598877,
      "epoch": 0.8789363191042687,
      "grad_norm": 0.03143594413995743,
      "kl": 0.0013470649719238281,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.49199927039444447,
      "reward_std": 0.30107220634818077,
      "rewards/random_math_reward": 0.49199927039444447,
      "step": 157
    },
    {
      "completion_length": 844.5101852416992,
      "epoch": 0.8845346396081175,
      "grad_norm": 0.031151605769991875,
      "kl": 0.001325845718383789,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4829933065921068,
      "reward_std": 0.2935031168162823,
      "rewards/random_math_reward": 0.4829933065921068,
      "step": 158
    },
    {
      "completion_length": 911.2563629150391,
      "epoch": 0.8901329601119664,
      "grad_norm": 0.010499164462089539,
      "kl": 0.0009677410125732422,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.46660212986171246,
      "reward_std": 0.30480547808110714,
      "rewards/random_math_reward": 0.46660212986171246,
      "step": 159
    },
    {
      "completion_length": 867.9106979370117,
      "epoch": 0.8957312806158153,
      "grad_norm": 0.039778802543878555,
      "kl": 0.0010061264038085938,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.46561445854604244,
      "reward_std": 0.2867754641920328,
      "rewards/random_math_reward": 0.46561445854604244,
      "step": 160
    },
    {
      "completion_length": 891.145378112793,
      "epoch": 0.9013296011196641,
      "grad_norm": 0.013824980705976486,
      "kl": 0.0010936260223388672,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4558851607143879,
      "reward_std": 0.277794330380857,
      "rewards/random_math_reward": 0.4558851607143879,
      "step": 161
    },
    {
      "completion_length": 869.1186027526855,
      "epoch": 0.906927921623513,
      "grad_norm": 0.01410576980561018,
      "kl": 0.0011093616485595703,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48392108641564846,
      "reward_std": 0.2871867660433054,
      "rewards/random_math_reward": 0.48392108641564846,
      "step": 162
    },
    {
      "completion_length": 884.0267677307129,
      "epoch": 0.9125262421273618,
      "grad_norm": 0.01321236602962017,
      "kl": 0.0009493827819824219,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5128381866961718,
      "reward_std": 0.30100576020777225,
      "rewards/random_math_reward": 0.5128381866961718,
      "step": 163
    },
    {
      "completion_length": 813.4183502197266,
      "epoch": 0.9181245626312107,
      "grad_norm": 0.02240627259016037,
      "kl": 0.0012166500091552734,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5056673623621464,
      "reward_std": 0.30340027436614037,
      "rewards/random_math_reward": 0.5056673623621464,
      "step": 164
    },
    {
      "completion_length": 866.1339149475098,
      "epoch": 0.9237228831350595,
      "grad_norm": 0.01877676323056221,
      "kl": 0.000985860824584961,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4957231916487217,
      "reward_std": 0.28768617659807205,
      "rewards/random_math_reward": 0.4957231916487217,
      "step": 165
    },
    {
      "completion_length": 934.8022727966309,
      "epoch": 0.9293212036389084,
      "grad_norm": 0.03419911116361618,
      "kl": 0.001092672348022461,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4862586259841919,
      "reward_std": 0.28661127388477325,
      "rewards/random_math_reward": 0.4862586259841919,
      "step": 166
    },
    {
      "completion_length": 845.2384948730469,
      "epoch": 0.9349195241427571,
      "grad_norm": 0.016880100592970848,
      "kl": 0.0010738372802734375,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.48645939491689205,
      "reward_std": 0.2838729955255985,
      "rewards/random_math_reward": 0.48645939491689205,
      "step": 167
    },
    {
      "completion_length": 897.2448921203613,
      "epoch": 0.940517844646606,
      "grad_norm": 0.03305617719888687,
      "kl": 0.0013968944549560547,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4894435331225395,
      "reward_std": 0.2995728589594364,
      "rewards/random_math_reward": 0.4894435331225395,
      "step": 168
    },
    {
      "completion_length": 870.5650291442871,
      "epoch": 0.9461161651504548,
      "grad_norm": 0.009322837926447392,
      "kl": 0.0010249614715576172,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5089151151478291,
      "reward_std": 0.30756222270429134,
      "rewards/random_math_reward": 0.5089151151478291,
      "step": 169
    },
    {
      "completion_length": 854.2805862426758,
      "epoch": 0.9517144856543037,
      "grad_norm": 0.0156830083578825,
      "kl": 0.000896453857421875,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4866387601941824,
      "reward_std": 0.29632874485105276,
      "rewards/random_math_reward": 0.4866387601941824,
      "step": 170
    },
    {
      "completion_length": 835.1109504699707,
      "epoch": 0.9573128061581525,
      "grad_norm": 0.04610283300280571,
      "kl": 0.0012865066528320312,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4992235377430916,
      "reward_std": 0.2944375704973936,
      "rewards/random_math_reward": 0.4992235377430916,
      "step": 171
    },
    {
      "completion_length": 885.7053337097168,
      "epoch": 0.9629111266620014,
      "grad_norm": 0.02425132691860199,
      "kl": 0.0011935234069824219,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5038886945694685,
      "reward_std": 0.28100786730647087,
      "rewards/random_math_reward": 0.5038886945694685,
      "step": 172
    },
    {
      "completion_length": 919.9604415893555,
      "epoch": 0.9685094471658502,
      "grad_norm": 0.011359083466231823,
      "kl": 0.0011434555053710938,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4733607657253742,
      "reward_std": 0.2893592659384012,
      "rewards/random_math_reward": 0.4733607657253742,
      "step": 173
    },
    {
      "completion_length": 816.8494758605957,
      "epoch": 0.9741077676696991,
      "grad_norm": 0.0354762077331543,
      "kl": 0.0011744499206542969,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.47691468335688114,
      "reward_std": 0.28350450936704874,
      "rewards/random_math_reward": 0.47691468335688114,
      "step": 174
    },
    {
      "completion_length": 837.5076293945312,
      "epoch": 0.979706088173548,
      "grad_norm": 0.030416017398238182,
      "kl": 0.0010030269622802734,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.49260007217526436,
      "reward_std": 0.265430293045938,
      "rewards/random_math_reward": 0.49260007217526436,
      "step": 175
    },
    {
      "completion_length": 887.9668235778809,
      "epoch": 0.9853044086773968,
      "grad_norm": 0.013739695772528648,
      "kl": 0.0010192394256591797,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.526423504576087,
      "reward_std": 0.2828856185078621,
      "rewards/random_math_reward": 0.526423504576087,
      "step": 176
    },
    {
      "completion_length": 837.7933464050293,
      "epoch": 0.9909027291812457,
      "grad_norm": 0.014401717111468315,
      "kl": 0.0010166168212890625,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.5026168543845415,
      "reward_std": 0.2906886078417301,
      "rewards/random_math_reward": 0.5026168543845415,
      "step": 177
    },
    {
      "completion_length": 915.8877296447754,
      "epoch": 0.9965010496850945,
      "grad_norm": 0.008807230740785599,
      "kl": 0.0009367465972900391,
      "learning_rate": 3e-07,
      "loss": 0.0,
      "reward": 0.4757802300155163,
      "reward_std": 0.29675872810184956,
      "rewards/random_math_reward": 0.4757802300155163,
      "step": 178
    },
    {
      "epoch": 0.9965010496850945,
      "step": 178,
      "total_flos": 0.0,
      "train_loss": 8.003707675146196e-08,
      "train_runtime": 99592.7903,
      "train_samples_per_second": 0.201,
      "train_steps_per_second": 0.002
    }
  ],
  "logging_steps": 1,
  "max_steps": 178,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}