{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.009424101711759565,
  "eval_steps": 500,
  "global_step": 1100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 117.2125,
      "epoch": 8.567365192508696e-05,
      "grad_norm": 1.5428545475006104,
      "kl": 7.6057333103563e-06,
      "learning_rate": 9.999742979044224e-07,
      "loss": 0.0,
      "reward": -304.1875,
      "reward_std": 100.79027481079102,
      "rewards/reward_len": -304.1875,
      "step": 10
    },
    {
      "completion_length": 106.55,
      "epoch": 0.00017134730385017391,
      "grad_norm": 1.7018674612045288,
      "kl": 4.6649108844576405e-05,
      "learning_rate": 9.999457400204474e-07,
      "loss": 0.0,
      "reward": -268.175,
      "reward_std": 85.13249053955079,
      "rewards/reward_len": -268.175,
      "step": 20
    },
    {
      "completion_length": 114.1,
      "epoch": 0.00025702095577526086,
      "grad_norm": 1.4646943807601929,
      "kl": 0.00014046951691852881,
      "learning_rate": 9.999171821364723e-07,
      "loss": 0.0,
      "reward": -295.625,
      "reward_std": 92.77591400146484,
      "rewards/reward_len": -295.625,
      "step": 30
    },
    {
      "completion_length": 112.45,
      "epoch": 0.00034269460770034783,
      "grad_norm": 1.4721007347106934,
      "kl": 0.0002871266377042048,
      "learning_rate": 9.998886242524973e-07,
      "loss": 0.0,
      "reward": -287.4375,
      "reward_std": 98.72801208496094,
      "rewards/reward_len": -287.4375,
      "step": 40
    },
    {
      "completion_length": 103.3375,
      "epoch": 0.0004283682596254348,
      "grad_norm": 1.5164705514907837,
      "kl": 0.00045935034868307414,
      "learning_rate": 9.998600663685224e-07,
      "loss": 0.0,
      "reward": -254.0625,
      "reward_std": 92.05970458984375,
      "rewards/reward_len": -254.0625,
      "step": 50
    },
    {
      "completion_length": 100.4375,
      "epoch": 0.0005140419115505217,
      "grad_norm": 1.267320156097412,
      "kl": 0.0006870983401313424,
      "learning_rate": 9.998315084845474e-07,
      "loss": 0.0,
      "reward": -229.725,
      "reward_std": 107.81015129089356,
      "rewards/reward_len": -229.725,
      "step": 60
    },
    {
      "completion_length": 104.0375,
      "epoch": 0.0005997155634756087,
      "grad_norm": 1.2804882526397705,
      "kl": 0.0007431535341311246,
      "learning_rate": 9.998029506005722e-07,
      "loss": 0.0,
      "reward": -276.275,
      "reward_std": 99.2921049118042,
      "rewards/reward_len": -276.275,
      "step": 70
    },
    {
      "completion_length": 102.2875,
      "epoch": 0.0006853892154006957,
      "grad_norm": 1.4880214929580688,
      "kl": 0.0008916158636566252,
      "learning_rate": 9.997743927165973e-07,
      "loss": 0.0,
      "reward": -266.025,
      "reward_std": 89.98284072875977,
      "rewards/reward_len": -266.025,
      "step": 80
    },
    {
      "completion_length": 124.3625,
      "epoch": 0.0007710628673257826,
      "grad_norm": 1.6408586502075195,
      "kl": 0.0010400827683042734,
      "learning_rate": 9.997458348326221e-07,
      "loss": 0.0,
      "reward": -318.0375,
      "reward_std": 99.00584564208984,
      "rewards/reward_len": -318.0375,
      "step": 90
    },
    {
      "completion_length": 92.075,
      "epoch": 0.0008567365192508696,
      "grad_norm": 1.535640835762024,
      "kl": 0.001465974550228566,
      "learning_rate": 9.997172769486472e-07,
      "loss": 0.0001,
      "reward": -206.9125,
      "reward_std": 85.6998176574707,
      "rewards/reward_len": -206.9125,
      "step": 100
    },
    {
      "completion_length": 99.125,
      "epoch": 0.0009424101711759565,
      "grad_norm": 2.4441370964050293,
      "kl": 0.0018776856479234994,
      "learning_rate": 9.99688719064672e-07,
      "loss": 0.0001,
      "reward": -247.55,
      "reward_std": 92.91195907592774,
      "rewards/reward_len": -247.55,
      "step": 110
    },
    {
      "completion_length": 104.35,
      "epoch": 0.0010280838231010434,
      "grad_norm": 1.3777748346328735,
      "kl": 0.002180302981287241,
      "learning_rate": 9.99660161180697e-07,
      "loss": 0.0001,
      "reward": -257.2625,
      "reward_std": 95.76971626281738,
      "rewards/reward_len": -257.2625,
      "step": 120
    },
    {
      "completion_length": 94.4125,
      "epoch": 0.0011137574750261306,
      "grad_norm": 1.3118126392364502,
      "kl": 0.002771495224442333,
      "learning_rate": 9.99631603296722e-07,
      "loss": 0.0001,
      "reward": -202.2125,
      "reward_std": 84.88922119140625,
      "rewards/reward_len": -202.2125,
      "step": 130
    },
    {
      "completion_length": 89.275,
      "epoch": 0.0011994311269512175,
      "grad_norm": 1.551858901977539,
      "kl": 0.0034235142404213546,
      "learning_rate": 9.996030454127471e-07,
      "loss": 0.0001,
      "reward": -211.7375,
      "reward_std": 80.14296417236328,
      "rewards/reward_len": -211.7375,
      "step": 140
    },
    {
      "completion_length": 115.8,
      "epoch": 0.0012851047788763044,
      "grad_norm": 1.333330750465393,
      "kl": 0.0028527263493742794,
      "learning_rate": 9.99574487528772e-07,
      "loss": 0.0001,
      "reward": -279.3625,
      "reward_std": 72.2537956237793,
      "rewards/reward_len": -279.3625,
      "step": 150
    },
    {
      "completion_length": 113.575,
      "epoch": 0.0013707784308013913,
      "grad_norm": 1.7015572786331177,
      "kl": 0.0033493154449388387,
      "learning_rate": 9.99545929644797e-07,
      "loss": 0.0001,
      "reward": -274.15,
      "reward_std": 101.25760879516602,
      "rewards/reward_len": -274.15,
      "step": 160
    },
    {
      "completion_length": 106.65,
      "epoch": 0.0014564520827264782,
      "grad_norm": 1.95174241065979,
      "kl": 0.004216998419724405,
      "learning_rate": 9.995173717608219e-07,
      "loss": 0.0002,
      "reward": -263.525,
      "reward_std": 89.78757858276367,
      "rewards/reward_len": -263.525,
      "step": 170
    },
    {
      "completion_length": 97.4875,
      "epoch": 0.0015421257346515652,
      "grad_norm": 1.5441001653671265,
      "kl": 0.0053229076787829396,
      "learning_rate": 9.99488813876847e-07,
      "loss": 0.0002,
      "reward": -236.775,
      "reward_std": 93.90848999023437,
      "rewards/reward_len": -236.775,
      "step": 180
    },
    {
      "completion_length": 102.7125,
      "epoch": 0.0016277993865766523,
      "grad_norm": 1.435225009918213,
      "kl": 0.005796084133908153,
      "learning_rate": 9.99460255992872e-07,
      "loss": 0.0002,
      "reward": -243.8375,
      "reward_std": 95.30473594665527,
      "rewards/reward_len": -243.8375,
      "step": 190
    },
    {
      "completion_length": 112.8,
      "epoch": 0.0017134730385017392,
      "grad_norm": 1.3907208442687988,
      "kl": 0.006385851232334971,
      "learning_rate": 9.994316981088968e-07,
      "loss": 0.0003,
      "reward": -292.475,
      "reward_std": 99.54066009521485,
      "rewards/reward_len": -292.475,
      "step": 200
    },
    {
      "completion_length": 106.2875,
      "epoch": 0.0017991466904268261,
      "grad_norm": 1.0377769470214844,
      "kl": 0.00682140497956425,
      "learning_rate": 9.994031402249218e-07,
      "loss": 0.0003,
      "reward": -263.15,
      "reward_std": 81.53659133911133,
      "rewards/reward_len": -263.15,
      "step": 210
    },
    {
      "completion_length": 94.15,
      "epoch": 0.001884820342351913,
      "grad_norm": 1.9323745965957642,
      "kl": 0.007357935793697834,
      "learning_rate": 9.993745823409469e-07,
      "loss": 0.0003,
      "reward": -217.4625,
      "reward_std": 90.16817970275879,
      "rewards/reward_len": -217.4625,
      "step": 220
    },
    {
      "completion_length": 92.8,
      "epoch": 0.001970493994277,
      "grad_norm": 1.8935438394546509,
      "kl": 0.008906404627487064,
      "learning_rate": 9.993460244569717e-07,
      "loss": 0.0004,
      "reward": -220.675,
      "reward_std": 74.1481372833252,
      "rewards/reward_len": -220.675,
      "step": 230
    },
    {
      "completion_length": 103.4875,
      "epoch": 0.002056167646202087,
      "grad_norm": 1.4551963806152344,
      "kl": 0.008559457911178469,
      "learning_rate": 9.993174665729968e-07,
      "loss": 0.0003,
      "reward": -242.5,
      "reward_std": 96.66918106079102,
      "rewards/reward_len": -242.5,
      "step": 240
    },
    {
      "completion_length": 96.7375,
      "epoch": 0.002141841298127174,
      "grad_norm": 1.3759669065475464,
      "kl": 0.010590891446918249,
      "learning_rate": 9.992889086890218e-07,
      "loss": 0.0004,
      "reward": -240.6125,
      "reward_std": 107.75033836364746,
      "rewards/reward_len": -240.6125,
      "step": 250
    },
    {
      "completion_length": 95.9125,
      "epoch": 0.002227514950052261,
      "grad_norm": 1.7361980676651,
      "kl": 0.009502861043438315,
      "learning_rate": 9.992603508050467e-07,
      "loss": 0.0004,
      "reward": -219.2125,
      "reward_std": 86.29145584106445,
      "rewards/reward_len": -219.2125,
      "step": 260
    },
    {
      "completion_length": 91.35,
      "epoch": 0.002313188601977348,
      "grad_norm": 1.6296714544296265,
      "kl": 0.012998782657086849,
      "learning_rate": 9.992317929210717e-07,
      "loss": 0.0005,
      "reward": -205.9,
      "reward_std": 104.51896514892579,
      "rewards/reward_len": -205.9,
      "step": 270
    },
    {
      "completion_length": 92.825,
      "epoch": 0.002398862253902435,
      "grad_norm": 2.636399269104004,
      "kl": 0.013483931683003903,
      "learning_rate": 9.992032350370967e-07,
      "loss": 0.0005,
      "reward": -211.0375,
      "reward_std": 72.49085922241211,
      "rewards/reward_len": -211.0375,
      "step": 280
    },
    {
      "completion_length": 96.475,
      "epoch": 0.002484535905827522,
      "grad_norm": 1.7585633993148804,
      "kl": 0.013156655337661504,
      "learning_rate": 9.991746771531216e-07,
      "loss": 0.0005,
      "reward": -233.4,
      "reward_std": 77.94148178100586,
      "rewards/reward_len": -233.4,
      "step": 290
    },
    {
      "completion_length": 78.575,
      "epoch": 0.002570209557752609,
      "grad_norm": 1.8656781911849976,
      "kl": 0.019020712468773127,
      "learning_rate": 9.991461192691466e-07,
      "loss": 0.0008,
      "reward": -165.3625,
      "reward_std": 83.65277671813965,
      "rewards/reward_len": -165.3625,
      "step": 300
    },
    {
      "completion_length": 80.9,
      "epoch": 0.0026558832096776957,
      "grad_norm": 1.886581540107727,
      "kl": 0.0151671776548028,
      "learning_rate": 9.991175613851715e-07,
      "loss": 0.0006,
      "reward": -153.625,
      "reward_std": 61.95796432495117,
      "rewards/reward_len": -153.625,
      "step": 310
    },
    {
      "completion_length": 79.9,
      "epoch": 0.0027415568616027826,
      "grad_norm": 3.363229990005493,
      "kl": 0.018376798275858162,
      "learning_rate": 9.990890035011965e-07,
      "loss": 0.0007,
      "reward": -170.775,
      "reward_std": 83.77405395507813,
      "rewards/reward_len": -170.775,
      "step": 320
    },
    {
      "completion_length": 90.9625,
      "epoch": 0.0028272305135278696,
      "grad_norm": 3.4404351711273193,
      "kl": 0.018607683992013336,
      "learning_rate": 9.990604456172216e-07,
      "loss": 0.0007,
      "reward": -189.1625,
      "reward_std": 93.11449737548828,
      "rewards/reward_len": -189.1625,
      "step": 330
    },
    {
      "completion_length": 90.8375,
      "epoch": 0.0029129041654529565,
      "grad_norm": 1.3329404592514038,
      "kl": 0.01699499851092696,
      "learning_rate": 9.990318877332464e-07,
      "loss": 0.0007,
      "reward": -203.4,
      "reward_std": 93.15925168991089,
      "rewards/reward_len": -203.4,
      "step": 340
    },
    {
      "completion_length": 93.7125,
      "epoch": 0.0029985778173780434,
      "grad_norm": 2.384962797164917,
      "kl": 0.022056785970926286,
      "learning_rate": 9.990033298492714e-07,
      "loss": 0.0009,
      "reward": -209.9,
      "reward_std": 95.43540840148925,
      "rewards/reward_len": -209.9,
      "step": 350
    },
    {
      "completion_length": 77.975,
      "epoch": 0.0030842514693031303,
      "grad_norm": 2.1530113220214844,
      "kl": 0.02068719444796443,
      "learning_rate": 9.989747719652965e-07,
      "loss": 0.0008,
      "reward": -145.9,
      "reward_std": 78.50229835510254,
      "rewards/reward_len": -145.9,
      "step": 360
    },
    {
      "completion_length": 77.175,
      "epoch": 0.0031699251212282177,
      "grad_norm": 2.1676974296569824,
      "kl": 0.028167350962758065,
      "learning_rate": 9.989462140813213e-07,
      "loss": 0.0011,
      "reward": -152.125,
      "reward_std": 77.3315586090088,
      "rewards/reward_len": -152.125,
      "step": 370
    },
    {
      "completion_length": 93.9,
      "epoch": 0.0032555987731533046,
      "grad_norm": 2.6751251220703125,
      "kl": 0.028882794454693793,
      "learning_rate": 9.989176561973464e-07,
      "loss": 0.0012,
      "reward": -192.5875,
      "reward_std": 93.17231407165528,
      "rewards/reward_len": -192.5875,
      "step": 380
    },
    {
      "completion_length": 62.6375,
      "epoch": 0.0033412724250783915,
      "grad_norm": 1.9541772603988647,
      "kl": 0.03603251725435257,
      "learning_rate": 9.988890983133714e-07,
      "loss": 0.0014,
      "reward": -95.625,
      "reward_std": 76.96504859924316,
      "rewards/reward_len": -95.625,
      "step": 390
    },
    {
      "completion_length": 76.5625,
      "epoch": 0.0034269460770034784,
      "grad_norm": 2.578866720199585,
      "kl": 0.036808890849351884,
      "learning_rate": 9.988605404293962e-07,
      "loss": 0.0015,
      "reward": -159.45,
      "reward_std": 81.1762622833252,
      "rewards/reward_len": -159.45,
      "step": 400
    },
    {
      "completion_length": 72.0875,
      "epoch": 0.0035126197289285653,
      "grad_norm": 2.2174038887023926,
      "kl": 0.03971981406211853,
      "learning_rate": 9.988319825454213e-07,
      "loss": 0.0016,
      "reward": -125.9625,
      "reward_std": 70.27665119171142,
      "rewards/reward_len": -125.9625,
      "step": 410
    },
    {
      "completion_length": 70.15,
      "epoch": 0.0035982933808536522,
      "grad_norm": 2.2227845191955566,
      "kl": 0.03956588953733444,
      "learning_rate": 9.988034246614463e-07,
      "loss": 0.0016,
      "reward": -110.5125,
      "reward_std": 68.30635490417481,
      "rewards/reward_len": -110.5125,
      "step": 420
    },
    {
      "completion_length": 66.8,
      "epoch": 0.003683967032778739,
      "grad_norm": 2.6529383659362793,
      "kl": 0.052050336450338366,
      "learning_rate": 9.987748667774712e-07,
      "loss": 0.0021,
      "reward": -110.275,
      "reward_std": 86.19376144409179,
      "rewards/reward_len": -110.275,
      "step": 430
    },
    {
      "completion_length": 71.2375,
      "epoch": 0.003769640684703826,
      "grad_norm": 2.401475429534912,
      "kl": 0.05769926495850086,
      "learning_rate": 9.987463088934962e-07,
      "loss": 0.0023,
      "reward": -128.325,
      "reward_std": 84.20119457244873,
      "rewards/reward_len": -128.325,
      "step": 440
    },
    {
      "completion_length": 90.3875,
      "epoch": 0.003855314336628913,
      "grad_norm": 2.8859000205993652,
      "kl": 0.043742291163653135,
      "learning_rate": 9.98717751009521e-07,
      "loss": 0.0017,
      "reward": -181.6,
      "reward_std": 75.12719230651855,
      "rewards/reward_len": -181.6,
      "step": 450
    },
    {
      "completion_length": 67.7625,
      "epoch": 0.003940987988554,
      "grad_norm": 2.0204262733459473,
      "kl": 0.0622777983546257,
      "learning_rate": 9.98689193125546e-07,
      "loss": 0.0025,
      "reward": -107.1125,
      "reward_std": 67.32327232360839,
      "rewards/reward_len": -107.1125,
      "step": 460
    },
    {
      "completion_length": 66.6,
      "epoch": 0.004026661640479087,
      "grad_norm": 2.5924794673919678,
      "kl": 0.056221196055412294,
      "learning_rate": 9.986606352415711e-07,
      "loss": 0.0022,
      "reward": -114.975,
      "reward_std": 75.61366195678711,
      "rewards/reward_len": -114.975,
      "step": 470
    },
    {
      "completion_length": 63.25,
      "epoch": 0.004112335292404174,
      "grad_norm": 2.5766448974609375,
      "kl": 0.06269333846867084,
      "learning_rate": 9.98632077357596e-07,
      "loss": 0.0025,
      "reward": -87.025,
      "reward_std": 75.8617115020752,
      "rewards/reward_len": -87.025,
      "step": 480
    },
    {
      "completion_length": 54.6625,
      "epoch": 0.004198008944329261,
      "grad_norm": 3.059234142303467,
      "kl": 0.06824085712432862,
      "learning_rate": 9.98603519473621e-07,
      "loss": 0.0027,
      "reward": -58.3375,
      "reward_std": 65.2177251815796,
      "rewards/reward_len": -58.3375,
      "step": 490
    },
    {
      "completion_length": 78.8125,
      "epoch": 0.004283682596254348,
      "grad_norm": 2.9885940551757812,
      "kl": 0.07582498777192086,
      "learning_rate": 9.98574961589646e-07,
      "loss": 0.003,
      "reward": -143.9,
      "reward_std": 63.89678440093994,
      "rewards/reward_len": -143.9,
      "step": 500
    },
    {
      "completion_length": 58.8375,
      "epoch": 0.004369356248179435,
      "grad_norm": 2.7045488357543945,
      "kl": 0.10633525140583515,
      "learning_rate": 9.98546403705671e-07,
      "loss": 0.0043,
      "reward": -71.625,
      "reward_std": 78.84037780761719,
      "rewards/reward_len": -71.625,
      "step": 510
    },
    {
      "completion_length": 66.35,
      "epoch": 0.004455029900104522,
      "grad_norm": 2.9650001525878906,
      "kl": 0.08462142385542393,
      "learning_rate": 9.98517845821696e-07,
      "loss": 0.0034,
      "reward": -93.275,
      "reward_std": 69.99792709350587,
      "rewards/reward_len": -93.275,
      "step": 520
    },
    {
      "completion_length": 81.0125,
      "epoch": 0.004540703552029609,
      "grad_norm": 3.4145212173461914,
      "kl": 0.08472472308203578,
      "learning_rate": 9.98489287937721e-07,
      "loss": 0.0034,
      "reward": -150.125,
      "reward_std": 64.46914005279541,
      "rewards/reward_len": -150.125,
      "step": 530
    },
    {
      "completion_length": 58.9625,
      "epoch": 0.004626377203954696,
      "grad_norm": 2.866358757019043,
      "kl": 0.11239302009344102,
      "learning_rate": 9.984607300537458e-07,
      "loss": 0.0045,
      "reward": -73.075,
      "reward_std": 66.32526473999023,
      "rewards/reward_len": -73.075,
      "step": 540
    },
    {
      "completion_length": 65.3875,
      "epoch": 0.004712050855879783,
      "grad_norm": 3.9847896099090576,
      "kl": 0.1188249222934246,
      "learning_rate": 9.984321721697709e-07,
      "loss": 0.0048,
      "reward": -104.2125,
      "reward_std": 84.05291709899902,
      "rewards/reward_len": -104.2125,
      "step": 550
    },
    {
      "completion_length": 55.425,
      "epoch": 0.00479772450780487,
      "grad_norm": 1.9827405214309692,
      "kl": 0.13096430897712708,
      "learning_rate": 9.984036142857957e-07,
      "loss": 0.0052,
      "reward": -52.8875,
      "reward_std": 76.0460901260376,
      "rewards/reward_len": -52.8875,
      "step": 560
    },
    {
      "completion_length": 63.4,
      "epoch": 0.004883398159729956,
      "grad_norm": 2.7263333797454834,
      "kl": 0.11985208839178085,
      "learning_rate": 9.983750564018208e-07,
      "loss": 0.0048,
      "reward": -85.3,
      "reward_std": 75.41236038208008,
      "rewards/reward_len": -85.3,
      "step": 570
    },
    {
      "completion_length": 60.8125,
      "epoch": 0.004969071811655044,
      "grad_norm": 2.395049810409546,
      "kl": 0.11624505072832107,
      "learning_rate": 9.983464985178458e-07,
      "loss": 0.0046,
      "reward": -79.4,
      "reward_std": 51.84059715270996,
      "rewards/reward_len": -79.4,
      "step": 580
    },
    {
      "completion_length": 51.7125,
      "epoch": 0.00505474546358013,
      "grad_norm": 3.295987606048584,
      "kl": 0.16594674922525882,
      "learning_rate": 9.983179406338709e-07,
      "loss": 0.0066,
      "reward": -43.55,
      "reward_std": 40.574602890014646,
      "rewards/reward_len": -43.55,
      "step": 590
    },
    {
      "completion_length": 67.3375,
      "epoch": 0.005140419115505218,
      "grad_norm": 4.078490257263184,
      "kl": 0.13739465102553367,
      "learning_rate": 9.982893827498957e-07,
      "loss": 0.0055,
      "reward": -98.5625,
      "reward_std": 72.55281829833984,
      "rewards/reward_len": -98.5625,
      "step": 600
    },
    {
      "completion_length": 46.9625,
      "epoch": 0.005226092767430304,
      "grad_norm": 4.195443153381348,
      "kl": 0.16595774739980698,
      "learning_rate": 9.982608248659207e-07,
      "loss": 0.0066,
      "reward": -25.4375,
      "reward_std": 67.01236572265626,
      "rewards/reward_len": -25.4375,
      "step": 610
    },
    {
      "completion_length": 48.8625,
      "epoch": 0.0053117664193553914,
      "grad_norm": 2.9587647914886475,
      "kl": 0.19849314391613007,
      "learning_rate": 9.982322669819456e-07,
      "loss": 0.0079,
      "reward": -32.075,
      "reward_std": 57.74148025512695,
      "rewards/reward_len": -32.075,
      "step": 620
    },
    {
      "completion_length": 55.6125,
      "epoch": 0.005397440071280479,
      "grad_norm": 3.3368797302246094,
      "kl": 0.1759088508784771,
      "learning_rate": 9.982037090979706e-07,
      "loss": 0.007,
      "reward": -60.65,
      "reward_std": 58.34819984436035,
      "rewards/reward_len": -60.65,
      "step": 630
    },
    {
      "completion_length": 51.225,
      "epoch": 0.005483113723205565,
      "grad_norm": 2.436890125274658,
      "kl": 0.19748679026961327,
      "learning_rate": 9.981751512139957e-07,
      "loss": 0.0079,
      "reward": -42.5,
      "reward_std": 70.55487613677978,
      "rewards/reward_len": -42.5,
      "step": 640
    },
    {
      "completion_length": 49.6875,
      "epoch": 0.005568787375130653,
      "grad_norm": 2.7104854583740234,
      "kl": 0.2179098792374134,
      "learning_rate": 9.981465933300205e-07,
      "loss": 0.0087,
      "reward": -41.025,
      "reward_std": 67.48074893951416,
      "rewards/reward_len": -41.025,
      "step": 650
    },
    {
      "completion_length": 45.3625,
      "epoch": 0.005654461027055739,
      "grad_norm": 2.9114573001861572,
      "kl": 0.292061947286129,
      "learning_rate": 9.981180354460456e-07,
      "loss": 0.0117,
      "reward": -22.2875,
      "reward_std": 67.80285606384277,
      "rewards/reward_len": -22.2875,
      "step": 660
    },
    {
      "completion_length": 43.25,
      "epoch": 0.0057401346789808265,
      "grad_norm": 4.684057712554932,
      "kl": 0.23551913797855378,
      "learning_rate": 9.980894775620706e-07,
      "loss": 0.0094,
      "reward": -7.725,
      "reward_std": 54.08637924194336,
      "rewards/reward_len": -7.725,
      "step": 670
    },
    {
      "completion_length": 39.75,
      "epoch": 0.005825808330905913,
      "grad_norm": 3.1786115169525146,
      "kl": 0.32834408432245255,
      "learning_rate": 9.980609196780954e-07,
      "loss": 0.0131,
      "reward": 6.0375,
      "reward_std": 58.28718643188476,
      "rewards/reward_len": 6.0375,
      "step": 680
    },
    {
      "completion_length": 49.025,
      "epoch": 0.005911481982831,
      "grad_norm": 4.3211870193481445,
      "kl": 0.30903512686491014,
      "learning_rate": 9.980323617941205e-07,
      "loss": 0.0124,
      "reward": -36.75,
      "reward_std": 74.52720184326172,
      "rewards/reward_len": -36.75,
      "step": 690
    },
    {
      "completion_length": 37.65,
      "epoch": 0.005997155634756087,
      "grad_norm": 2.111032247543335,
      "kl": 0.32604690343141557,
      "learning_rate": 9.980038039101453e-07,
      "loss": 0.013,
      "reward": 18.4125,
      "reward_std": 50.093895721435544,
      "rewards/reward_len": 18.4125,
      "step": 700
    },
    {
      "completion_length": 33.475,
      "epoch": 0.006082829286681174,
      "grad_norm": 4.225799560546875,
      "kl": 0.3358478359878063,
      "learning_rate": 9.979752460261704e-07,
      "loss": 0.0134,
      "reward": 29.3375,
      "reward_std": 45.97467594146728,
      "rewards/reward_len": 29.3375,
      "step": 710
    },
    {
      "completion_length": 36.1875,
      "epoch": 0.006168502938606261,
      "grad_norm": 3.743486166000366,
      "kl": 0.3429659426212311,
      "learning_rate": 9.979466881421954e-07,
      "loss": 0.0137,
      "reward": 14.65,
      "reward_std": 42.70287170410156,
      "rewards/reward_len": 14.65,
      "step": 720
    },
    {
      "completion_length": 36.1375,
      "epoch": 0.006254176590531348,
      "grad_norm": 3.634866952896118,
      "kl": 0.41900385320186617,
      "learning_rate": 9.979181302582205e-07,
      "loss": 0.0168,
      "reward": 21.025,
      "reward_std": 38.98911609649658,
      "rewards/reward_len": 21.025,
      "step": 730
    },
    {
      "completion_length": 38.7,
      "epoch": 0.006339850242456435,
      "grad_norm": 5.735452175140381,
      "kl": 0.3897597849369049,
      "learning_rate": 9.978895723742453e-07,
      "loss": 0.0156,
      "reward": 11.05,
      "reward_std": 49.65783863067627,
      "rewards/reward_len": 11.05,
      "step": 740
    },
    {
      "completion_length": 38.1125,
      "epoch": 0.006425523894381522,
      "grad_norm": 2.927375555038452,
      "kl": 0.435553178191185,
      "learning_rate": 9.978610144902703e-07,
      "loss": 0.0174,
      "reward": 8.8125,
      "reward_std": 55.300292015075684,
      "rewards/reward_len": 8.8125,
      "step": 750
    },
    {
      "completion_length": 29.675,
      "epoch": 0.006511197546306609,
      "grad_norm": 3.246641159057617,
      "kl": 0.43810410499572755,
      "learning_rate": 9.978324566062952e-07,
      "loss": 0.0175,
      "reward": 40.5375,
      "reward_std": 35.55714553594589,
      "rewards/reward_len": 40.5375,
      "step": 760
    },
    {
      "completion_length": 32.3125,
      "epoch": 0.006596871198231696,
      "grad_norm": 3.6894195079803467,
      "kl": 0.5961927145719528,
      "learning_rate": 9.978038987223202e-07,
      "loss": 0.0238,
      "reward": 28.3625,
      "reward_std": 40.15065488815308,
      "rewards/reward_len": 28.3625,
      "step": 770
    },
    {
      "completion_length": 33.0625,
      "epoch": 0.006682544850156783,
      "grad_norm": 4.511296272277832,
      "kl": 0.5751866042613983,
      "learning_rate": 9.977753408383453e-07,
      "loss": 0.023,
      "reward": 27.575,
      "reward_std": 46.356824684143064,
      "rewards/reward_len": 27.575,
      "step": 780
    },
    {
      "completion_length": 33.0375,
      "epoch": 0.0067682185020818695,
      "grad_norm": 3.548011302947998,
      "kl": 0.42039217352867125,
      "learning_rate": 9.9774678295437e-07,
      "loss": 0.0168,
      "reward": 26.5125,
      "reward_std": 46.602904319763184,
      "rewards/reward_len": 26.5125,
      "step": 790
    },
    {
      "completion_length": 26.85,
      "epoch": 0.006853892154006957,
      "grad_norm": 5.255430221557617,
      "kl": 0.5921719163656235,
      "learning_rate": 9.977182250703951e-07,
      "loss": 0.0237,
      "reward": 60.1375,
      "reward_std": 30.120204424858095,
      "rewards/reward_len": 60.1375,
      "step": 800
    },
    {
      "completion_length": 31.825,
      "epoch": 0.006939565805932043,
      "grad_norm": 2.8932251930236816,
      "kl": 0.5926052063703537,
      "learning_rate": 9.976896671864202e-07,
      "loss": 0.0237,
      "reward": 37.225,
      "reward_std": 37.70037169456482,
      "rewards/reward_len": 37.225,
      "step": 810
    },
    {
      "completion_length": 43.175,
      "epoch": 0.007025239457857131,
      "grad_norm": 2.886444091796875,
      "kl": 0.5511217802762985,
      "learning_rate": 9.97661109302445e-07,
      "loss": 0.022,
      "reward": -2.45,
      "reward_std": 56.478550338745116,
      "rewards/reward_len": -2.45,
      "step": 820
    },
    {
      "completion_length": 30.2625,
      "epoch": 0.007110913109782218,
      "grad_norm": 2.814298629760742,
      "kl": 0.7488752245903015,
      "learning_rate": 9.9763255141847e-07,
      "loss": 0.03,
      "reward": 43.575,
      "reward_std": 47.09467077255249,
      "rewards/reward_len": 43.575,
      "step": 830
    },
    {
      "completion_length": 25.85,
      "epoch": 0.0071965867617073045,
      "grad_norm": 5.516399383544922,
      "kl": 0.7932253420352936,
      "learning_rate": 9.97603993534495e-07,
      "loss": 0.0317,
      "reward": 63.85,
      "reward_std": 25.45775556564331,
      "rewards/reward_len": 63.85,
      "step": 840
    },
    {
      "completion_length": 27.95,
      "epoch": 0.007282260413632392,
      "grad_norm": 4.706915378570557,
      "kl": 0.6608895510435104,
      "learning_rate": 9.9757543565052e-07,
      "loss": 0.0264,
      "reward": 52.7,
      "reward_std": 39.781547927856444,
      "rewards/reward_len": 52.7,
      "step": 850
    },
    {
      "completion_length": 26.9125,
      "epoch": 0.007367934065557478,
      "grad_norm": 14.130315780639648,
      "kl": 0.6030549615621567,
      "learning_rate": 9.97546877766545e-07,
      "loss": 0.0241,
      "reward": 62.875,
      "reward_std": 28.904275512695314,
      "rewards/reward_len": 62.875,
      "step": 860
    },
    {
      "completion_length": 27.35,
      "epoch": 0.007453607717482566,
      "grad_norm": 5.187411785125732,
      "kl": 0.7092643141746521,
      "learning_rate": 9.9751831988257e-07,
      "loss": 0.0284,
      "reward": 47.6375,
      "reward_std": 32.484121799468994,
      "rewards/reward_len": 47.6375,
      "step": 870
    },
    {
      "completion_length": 46.7875,
      "epoch": 0.007539281369407652,
      "grad_norm": 2.9746391773223877,
      "kl": 0.6793952610343694,
      "learning_rate": 9.974897619985949e-07,
      "loss": 0.0272,
      "reward": -19.2625,
      "reward_std": 51.298577308654785,
      "rewards/reward_len": -19.2625,
      "step": 880
    },
    {
      "completion_length": 26.6375,
      "epoch": 0.0076249550213327395,
      "grad_norm": 4.254567623138428,
      "kl": 0.7589318662881851,
      "learning_rate": 9.9746120411462e-07,
      "loss": 0.0304,
      "reward": 57.5625,
      "reward_std": 39.20279521942139,
      "rewards/reward_len": 57.5625,
      "step": 890
    },
    {
      "completion_length": 24.1375,
      "epoch": 0.007710628673257826,
      "grad_norm": 4.250896453857422,
      "kl": 0.8811014115810394,
      "learning_rate": 9.974326462306448e-07,
      "loss": 0.0352,
      "reward": 73.8625,
      "reward_std": 19.012751007080077,
      "rewards/reward_len": 73.8625,
      "step": 900
    },
    {
      "completion_length": 25.4,
      "epoch": 0.007796302325182913,
      "grad_norm": 3.278289556503296,
      "kl": 0.8711494505405426,
      "learning_rate": 9.974040883466698e-07,
      "loss": 0.0348,
      "reward": 69.4375,
      "reward_std": 24.089942455291748,
      "rewards/reward_len": 69.4375,
      "step": 910
    },
    {
      "completion_length": 25.275,
      "epoch": 0.007881975977108,
      "grad_norm": 4.476153373718262,
      "kl": 0.8596636682748795,
      "learning_rate": 9.973755304626946e-07,
      "loss": 0.0344,
      "reward": 63.3875,
      "reward_std": 20.410217094421387,
      "rewards/reward_len": 63.3875,
      "step": 920
    },
    {
      "completion_length": 24.8625,
      "epoch": 0.007967649629033088,
      "grad_norm": 5.825425624847412,
      "kl": 1.0523226261138916,
      "learning_rate": 9.9734697257872e-07,
      "loss": 0.0421,
      "reward": 66.95,
      "reward_std": 25.993260192871094,
      "rewards/reward_len": 66.95,
      "step": 930
    },
    {
      "completion_length": 23.0875,
      "epoch": 0.008053323280958175,
      "grad_norm": 4.372145652770996,
      "kl": 0.9800844460725784,
      "learning_rate": 9.973184146947447e-07,
      "loss": 0.0392,
      "reward": 67.175,
      "reward_std": 27.855006313323976,
      "rewards/reward_len": 67.175,
      "step": 940
    },
    {
      "completion_length": 23.325,
      "epoch": 0.008138996932883261,
      "grad_norm": 5.127079010009766,
      "kl": 0.9943849921226502,
      "learning_rate": 9.972898568107698e-07,
      "loss": 0.0398,
      "reward": 78.1125,
      "reward_std": 17.191841506958006,
      "rewards/reward_len": 78.1125,
      "step": 950
    },
    {
      "completion_length": 22.9375,
      "epoch": 0.008224670584808347,
      "grad_norm": 4.183666229248047,
      "kl": 1.068257564306259,
      "learning_rate": 9.972612989267946e-07,
      "loss": 0.0427,
      "reward": 79.1625,
      "reward_std": 18.828048992156983,
      "rewards/reward_len": 79.1625,
      "step": 960
    },
    {
      "completion_length": 22.9375,
      "epoch": 0.008310344236733436,
      "grad_norm": 4.63860559463501,
      "kl": 1.0732409298419952,
      "learning_rate": 9.972327410428197e-07,
      "loss": 0.0429,
      "reward": 71.4875,
      "reward_std": 22.58247709274292,
      "rewards/reward_len": 71.4875,
      "step": 970
    },
    {
      "completion_length": 25.125,
      "epoch": 0.008396017888658522,
      "grad_norm": 5.824211597442627,
      "kl": 0.9688010692596436,
      "learning_rate": 9.972041831588445e-07,
      "loss": 0.0388,
      "reward": 67.6,
      "reward_std": 23.308015012741087,
      "rewards/reward_len": 67.6,
      "step": 980
    },
    {
      "completion_length": 22.6375,
      "epoch": 0.008481691540583609,
      "grad_norm": 2.8945517539978027,
      "kl": 1.118459904193878,
      "learning_rate": 9.971756252748695e-07,
      "loss": 0.0447,
      "reward": 81.375,
      "reward_std": 17.22281708717346,
      "rewards/reward_len": 81.375,
      "step": 990
    },
    {
      "completion_length": 21.275,
      "epoch": 0.008567365192508695,
      "grad_norm": 4.304519176483154,
      "kl": 0.9363485395908355,
      "learning_rate": 9.971470673908946e-07,
      "loss": 0.0375,
      "reward": 75.775,
      "reward_std": 23.021865463256837,
      "rewards/reward_len": 75.775,
      "step": 1000
    },
    {
      "completion_length": 18.8625,
      "epoch": 0.008653038844433783,
      "grad_norm": 4.940158843994141,
      "kl": 1.0820651412010194,
      "learning_rate": 9.971185095069196e-07,
      "loss": 0.0433,
      "reward": 76.725,
      "reward_std": 19.4085599899292,
      "rewards/reward_len": 76.725,
      "step": 1010
    },
    {
      "completion_length": 22.45,
      "epoch": 0.00873871249635887,
      "grad_norm": 6.257218837738037,
      "kl": 1.3543362498283387,
      "learning_rate": 9.970899516229445e-07,
      "loss": 0.0542,
      "reward": 73.275,
      "reward_std": 21.16908242702484,
      "rewards/reward_len": 73.275,
      "step": 1020
    },
    {
      "completion_length": 21.1875,
      "epoch": 0.008824386148283956,
      "grad_norm": 7.533606052398682,
      "kl": 0.9111672580242157,
      "learning_rate": 9.970613937389695e-07,
      "loss": 0.0364,
      "reward": 79.5375,
      "reward_std": 17.676442378759383,
      "rewards/reward_len": 79.5375,
      "step": 1030
    },
    {
      "completion_length": 17.7,
      "epoch": 0.008910059800209045,
      "grad_norm": 6.559030532836914,
      "kl": 1.283721137046814,
      "learning_rate": 9.970328358549944e-07,
      "loss": 0.0513,
      "reward": 86.375,
      "reward_std": 15.953600931167603,
      "rewards/reward_len": 86.375,
      "step": 1040
    },
    {
      "completion_length": 21.5125,
      "epoch": 0.008995733452134131,
      "grad_norm": 9.517870903015137,
      "kl": 1.2262637436389923,
      "learning_rate": 9.970042779710194e-07,
      "loss": 0.0491,
      "reward": 75.275,
      "reward_std": 21.458053398132325,
      "rewards/reward_len": 75.275,
      "step": 1050
    },
    {
      "completion_length": 21.75,
      "epoch": 0.009081407104059218,
      "grad_norm": 12.434151649475098,
      "kl": 1.324447250366211,
      "learning_rate": 9.969757200870442e-07,
      "loss": 0.053,
      "reward": 81.1125,
      "reward_std": 16.958592414855957,
      "rewards/reward_len": 81.1125,
      "step": 1060
    },
    {
      "completion_length": 22.225,
      "epoch": 0.009167080755984304,
      "grad_norm": 5.839966297149658,
      "kl": 1.3111671924591064,
      "learning_rate": 9.969471622030695e-07,
      "loss": 0.0524,
      "reward": 74.425,
      "reward_std": 27.983480072021486,
      "rewards/reward_len": 74.425,
      "step": 1070
    },
    {
      "completion_length": 20.3,
      "epoch": 0.009252754407909392,
      "grad_norm": 5.334268093109131,
      "kl": 1.2456512570381164,
      "learning_rate": 9.969186043190943e-07,
      "loss": 0.0498,
      "reward": 81.6,
      "reward_std": 18.58235924243927,
      "rewards/reward_len": 81.6,
      "step": 1080
    },
    {
      "completion_length": 22.3125,
      "epoch": 0.009338428059834479,
      "grad_norm": 4.366571426391602,
      "kl": 1.3097686409950255,
      "learning_rate": 9.968900464351194e-07,
      "loss": 0.0524,
      "reward": 77.85,
      "reward_std": 20.753051567077637,
      "rewards/reward_len": 77.85,
      "step": 1090
    },
    {
      "completion_length": 17.3875,
      "epoch": 0.009424101711759565,
      "grad_norm": 6.299209117889404,
      "kl": 1.508902084827423,
      "learning_rate": 9.968614885511442e-07,
      "loss": 0.0604,
      "reward": 83.225,
      "reward_std": 13.49931423664093,
      "rewards/reward_len": 83.225,
      "step": 1100
    }
  ],
  "logging_steps": 10,
  "max_steps": 350166,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 50,
  "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": 8,
  "trial_name": null,
  "trial_params": null
}