{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.13969267611255237,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 6335.173723493303,
      "epoch": 0.0013969267611255239,
      "grad_norm": 0.1530565852782136,
      "learning_rate": 0.0,
      "loss": 0.0119,
      "num_tokens": 1330201.0,
      "reward": 0.3924093656241894,
      "reward_std": 0.32559600685323986,
      "rewards/reward_func": 0.3924093528517655,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "epoch": 0.0027938535222510478,
      "grad_norm": 0.15303408357738274,
      "learning_rate": 1.3888888888888887e-08,
      "loss": 0.0119,
      "step": 2
    },
    {
      "clip_ratio": 0.010408175710056509,
      "epoch": 0.004190780283376571,
      "grad_norm": 0.1764633199425111,
      "learning_rate": 2.7777777777777774e-08,
      "loss": 0.0125,
      "step": 3
    },
    {
      "clip_ratio": 0.010680466823812042,
      "epoch": 0.0055877070445020955,
      "grad_norm": 0.2275277428050235,
      "learning_rate": 4.166666666666666e-08,
      "loss": 0.0126,
      "step": 4
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 5737.668526785715,
      "epoch": 0.006984633805627619,
      "grad_norm": 0.19107039038527468,
      "learning_rate": 5.555555555555555e-08,
      "loss": 0.0293,
      "num_tokens": 2545372.0,
      "reward": 0.4294457712343761,
      "reward_std": 0.37008823986564365,
      "rewards/reward_func": 0.4294457531401089,
      "step": 5
    },
    {
      "clip_ratio": 0.010828850524766105,
      "epoch": 0.008381560566753142,
      "grad_norm": 0.2240354781456917,
      "learning_rate": 6.944444444444444e-08,
      "loss": 0.0301,
      "step": 6
    },
    {
      "clip_ratio": 0.010841753533376115,
      "epoch": 0.009778487327878668,
      "grad_norm": 0.2767916742737982,
      "learning_rate": 8.333333333333333e-08,
      "loss": 0.0301,
      "step": 7
    },
    {
      "clip_ratio": 0.010758176312914916,
      "epoch": 0.011175414089004191,
      "grad_norm": 0.2668705057271519,
      "learning_rate": 9.722222222222222e-08,
      "loss": 0.0301,
      "step": 8
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 6511.857439313616,
      "epoch": 0.012572340850129714,
      "grad_norm": 0.19266419886931196,
      "learning_rate": 1.111111111111111e-07,
      "loss": 0.0051,
      "num_tokens": 3911895.0,
      "reward": 0.32038674876093864,
      "reward_std": 0.3298808889729636,
      "rewards/reward_func": 0.32038673226322445,
      "step": 9
    },
    {
      "clip_ratio": 0.012134518141725234,
      "epoch": 0.013969267611255238,
      "grad_norm": 0.21591989333017872,
      "learning_rate": 1.25e-07,
      "loss": 0.0058,
      "step": 10
    },
    {
      "clip_ratio": 0.012228245115173715,
      "epoch": 0.015366194372380763,
      "grad_norm": 0.20676433229073893,
      "learning_rate": 1.3888888888888888e-07,
      "loss": 0.0058,
      "step": 11
    },
    {
      "clip_ratio": 0.012062976469418831,
      "epoch": 0.016763121133506285,
      "grad_norm": 0.2100849809088676,
      "learning_rate": 1.527777777777778e-07,
      "loss": 0.0058,
      "step": 12
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 5776.398141043527,
      "epoch": 0.01816004789463181,
      "grad_norm": 0.21639349074679098,
      "learning_rate": 1.6666666666666665e-07,
      "loss": 0.0233,
      "num_tokens": 5132268.0,
      "reward": 0.4814984883580889,
      "reward_std": 0.3754643425345421,
      "rewards/reward_func": 0.48149845749139786,
      "step": 13
    },
    {
      "clip_ratio": 0.010207080947501319,
      "epoch": 0.019556974655757335,
      "grad_norm": 0.22611722201173162,
      "learning_rate": 1.8055555555555554e-07,
      "loss": 0.0241,
      "step": 14
    },
    {
      "clip_ratio": 0.010003214618856353,
      "epoch": 0.02095390141688286,
      "grad_norm": 0.21387408972131952,
      "learning_rate": 1.9444444444444445e-07,
      "loss": 0.024,
      "step": 15
    },
    {
      "clip_ratio": 0.010145484004169703,
      "epoch": 0.022350828178008382,
      "grad_norm": 0.2965931721160837,
      "learning_rate": 2.0833333333333333e-07,
      "loss": 0.024,
      "step": 16
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 6396.923740931919,
      "epoch": 0.023747754939133905,
      "grad_norm": 0.17812701051388663,
      "learning_rate": 2.222222222222222e-07,
      "loss": 0.0143,
      "num_tokens": 6474962.0,
      "reward": 0.4179248466555561,
      "reward_std": 0.3605238295027188,
      "rewards/reward_func": 0.41792482949261156,
      "step": 17
    },
    {
      "clip_ratio": 0.010267588176897593,
      "epoch": 0.02514468170025943,
      "grad_norm": 0.18587492539762027,
      "learning_rate": 2.361111111111111e-07,
      "loss": 0.015,
      "step": 18
    },
    {
      "clip_ratio": 0.010148471055020179,
      "epoch": 0.026541608461384952,
      "grad_norm": 0.1810944495339226,
      "learning_rate": 2.5e-07,
      "loss": 0.0149,
      "step": 19
    },
    {
      "clip_ratio": 0.01020466532957341,
      "epoch": 0.027938535222510476,
      "grad_norm": 0.1924193708866086,
      "learning_rate": 2.638888888888889e-07,
      "loss": 0.015,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 5085.755362374442,
      "epoch": 0.029335461983636,
      "grad_norm": 0.3165497451368565,
      "learning_rate": 2.7777777777777776e-07,
      "loss": 0.0104,
      "num_tokens": 7560523.0,
      "reward": 0.464439152606896,
      "reward_std": 0.3689843277846064,
      "rewards/reward_func": 0.46443913131952286,
      "step": 21
    },
    {
      "clip_ratio": 0.010546388570219278,
      "epoch": 0.030732388744761526,
      "grad_norm": 0.3395774959095009,
      "learning_rate": 2.916666666666667e-07,
      "loss": 0.0113,
      "step": 22
    },
    {
      "clip_ratio": 0.010336563357019,
      "epoch": 0.032129315505887046,
      "grad_norm": 0.31582946500885495,
      "learning_rate": 3.055555555555556e-07,
      "loss": 0.0112,
      "step": 23
    },
    {
      "clip_ratio": 0.010099471979109305,
      "epoch": 0.03352624226701257,
      "grad_norm": 0.36247326697578075,
      "learning_rate": 3.194444444444444e-07,
      "loss": 0.0112,
      "step": 24
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 6461.342215401785,
      "epoch": 0.03492316902813809,
      "grad_norm": 0.15567466235487185,
      "learning_rate": 3.333333333333333e-07,
      "loss": 0.0193,
      "num_tokens": 8914764.0,
      "reward": 0.38339252876383917,
      "reward_std": 0.3311073939715113,
      "rewards/reward_func": 0.3833925181201526,
      "step": 25
    },
    {
      "clip_ratio": 0.009515652732391442,
      "epoch": 0.03632009578926362,
      "grad_norm": 0.25805173133838866,
      "learning_rate": 3.472222222222222e-07,
      "loss": 0.0199,
      "step": 26
    },
    {
      "clip_ratio": 0.009263422406677688,
      "epoch": 0.03771702255038915,
      "grad_norm": 0.4090162080972072,
      "learning_rate": 3.6111111111111107e-07,
      "loss": 0.0199,
      "step": 27
    },
    {
      "clip_ratio": 0.009094293162758862,
      "epoch": 0.03911394931151467,
      "grad_norm": 0.14691760483745145,
      "learning_rate": 3.75e-07,
      "loss": 0.0198,
      "step": 28
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 5968.250279017857,
      "epoch": 0.040510876072640194,
      "grad_norm": 0.16406943154337036,
      "learning_rate": 3.888888888888889e-07,
      "loss": 0.0115,
      "num_tokens": 10173741.0,
      "reward": 0.3594382884246962,
      "reward_std": 0.32283720054796766,
      "rewards/reward_func": 0.35943826926606043,
      "step": 29
    },
    {
      "clip_ratio": 0.01090411888435483,
      "epoch": 0.04190780283376572,
      "grad_norm": 0.21927196182565817,
      "learning_rate": 4.027777777777778e-07,
      "loss": 0.0124,
      "step": 30
    },
    {
      "clip_ratio": 0.010733458446338773,
      "epoch": 0.04330472959489124,
      "grad_norm": 0.3447213393816689,
      "learning_rate": 4.1666666666666667e-07,
      "loss": 0.0121,
      "step": 31
    },
    {
      "clip_ratio": 0.010630057725523199,
      "epoch": 0.044701656356016764,
      "grad_norm": 0.3620568701207513,
      "learning_rate": 4.3055555555555555e-07,
      "loss": 0.0121,
      "step": 32
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 5812.648210797991,
      "epoch": 0.04609858311714229,
      "grad_norm": 0.16987419224516487,
      "learning_rate": 4.444444444444444e-07,
      "loss": 0.0179,
      "num_tokens": 11403427.0,
      "reward": 0.3713971259338515,
      "reward_std": 0.3195132836699486,
      "rewards/reward_func": 0.3713971110326903,
      "step": 33
    },
    {
      "clip_ratio": 0.010423385953929807,
      "epoch": 0.04749550987826781,
      "grad_norm": 0.16329217057193554,
      "learning_rate": 4.5833333333333327e-07,
      "loss": 0.0185,
      "step": 34
    },
    {
      "clip_ratio": 0.010347368278806763,
      "epoch": 0.048892436639393334,
      "grad_norm": 0.15422959333602954,
      "learning_rate": 4.722222222222222e-07,
      "loss": 0.0185,
      "step": 35
    },
    {
      "clip_ratio": 0.010140369646251202,
      "epoch": 0.05028936340051886,
      "grad_norm": 0.14689712083262277,
      "learning_rate": 4.861111111111111e-07,
      "loss": 0.0184,
      "step": 36
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 5546.617606026785,
      "epoch": 0.05168629016164438,
      "grad_norm": 0.23439115342799705,
      "learning_rate": 5e-07,
      "loss": 0.0161,
      "num_tokens": 12578819.0,
      "reward": 0.4297656629766737,
      "reward_std": 0.357943703021322,
      "rewards/reward_func": 0.42976564009274754,
      "step": 37
    },
    {
      "clip_ratio": 0.010506943932601384,
      "epoch": 0.053083216922769905,
      "grad_norm": 0.28362735010406454,
      "learning_rate": 5.138888888888889e-07,
      "loss": 0.0168,
      "step": 38
    },
    {
      "clip_ratio": 0.010338681018246072,
      "epoch": 0.05448014368389543,
      "grad_norm": 0.22707404637725537,
      "learning_rate": 5.277777777777777e-07,
      "loss": 0.0167,
      "step": 39
    },
    {
      "clip_ratio": 0.010024449788033962,
      "epoch": 0.05587707044502095,
      "grad_norm": 0.17811231341381648,
      "learning_rate": 5.416666666666666e-07,
      "loss": 0.0165,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 6501.444231305803,
      "epoch": 0.057273997206146475,
      "grad_norm": 0.19673275542337104,
      "learning_rate": 5.555555555555555e-07,
      "loss": 0.0203,
      "num_tokens": 13942392.0,
      "reward": 0.39034509126629147,
      "reward_std": 0.3664030634931156,
      "rewards/reward_func": 0.3903450625283377,
      "step": 41
    },
    {
      "clip_ratio": 0.010402361529746227,
      "epoch": 0.058670923967272,
      "grad_norm": 0.46687244963414065,
      "learning_rate": 5.694444444444444e-07,
      "loss": 0.0211,
      "step": 42
    },
    {
      "clip_ratio": 0.010496154893189669,
      "epoch": 0.06006785072839753,
      "grad_norm": 0.19428194405187207,
      "learning_rate": 5.833333333333334e-07,
      "loss": 0.0209,
      "step": 43
    },
    {
      "clip_ratio": 0.010158670766811286,
      "epoch": 0.06146477748952305,
      "grad_norm": 0.15306555166745645,
      "learning_rate": 5.972222222222222e-07,
      "loss": 0.0208,
      "step": 44
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 6369.643048967634,
      "epoch": 0.06286170425064858,
      "grad_norm": 0.1871015916141642,
      "learning_rate": 6.111111111111112e-07,
      "loss": 0.0161,
      "num_tokens": 15279078.0,
      "reward": 0.36025070399045944,
      "reward_std": 0.35143060450042996,
      "rewards/reward_func": 0.360250677381243,
      "step": 45
    },
    {
      "clip_ratio": 0.01079715442444597,
      "epoch": 0.06425863101177409,
      "grad_norm": 0.6294866160536174,
      "learning_rate": 6.249999999999999e-07,
      "loss": 0.0169,
      "step": 46
    },
    {
      "clip_ratio": 0.010736021545848675,
      "epoch": 0.06565555777289962,
      "grad_norm": 2.137657977104326,
      "learning_rate": 6.388888888888888e-07,
      "loss": 0.0171,
      "step": 47
    },
    {
      "clip_ratio": 0.010864090407267213,
      "epoch": 0.06705248453402514,
      "grad_norm": 0.18428434234054034,
      "learning_rate": 6.527777777777777e-07,
      "loss": 0.0166,
      "step": 48
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 6655.908412388393,
      "epoch": 0.06844941129515067,
      "grad_norm": 0.16953542603782387,
      "learning_rate": 6.666666666666666e-07,
      "loss": 0.0148,
      "num_tokens": 16672949.0,
      "reward": 0.35046685327376637,
      "reward_std": 0.30536604832325664,
      "rewards/reward_func": 0.3504668373082365,
      "step": 49
    },
    {
      "clip_ratio": 0.010905002310339893,
      "epoch": 0.06984633805627619,
      "grad_norm": 1.9607105946099048,
      "learning_rate": 6.805555555555556e-07,
      "loss": 0.0153,
      "step": 50
    },
    {
      "clip_ratio": 0.011137549964977162,
      "epoch": 0.07124326481740172,
      "grad_norm": 0.16619610980435887,
      "learning_rate": 6.944444444444444e-07,
      "loss": 0.0153,
      "step": 51
    },
    {
      "clip_ratio": 0.011474365035870246,
      "epoch": 0.07264019157852725,
      "grad_norm": 0.13731779782413311,
      "learning_rate": 7.083333333333334e-07,
      "loss": 0.0151,
      "step": 52
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 6259.326799665178,
      "epoch": 0.07403711833965276,
      "grad_norm": 0.18005417578145988,
      "learning_rate": 7.222222222222221e-07,
      "loss": 0.0124,
      "num_tokens": 17987081.0,
      "reward": 0.3450017161667347,
      "reward_std": 0.3385175902928625,
      "rewards/reward_func": 0.34500169487936155,
      "step": 53
    },
    {
      "clip_ratio": 0.012061000069869416,
      "epoch": 0.0754340451007783,
      "grad_norm": 0.3653574862365132,
      "learning_rate": 7.361111111111111e-07,
      "loss": 0.0132,
      "step": 54
    },
    {
      "clip_ratio": 0.01210923127031752,
      "epoch": 0.07683097186190381,
      "grad_norm": 0.1572526876806962,
      "learning_rate": 7.5e-07,
      "loss": 0.013,
      "step": 55
    },
    {
      "clip_ratio": 0.012545851857534476,
      "epoch": 0.07822789862302934,
      "grad_norm": 0.1348151811096135,
      "learning_rate": 7.638888888888888e-07,
      "loss": 0.0128,
      "step": 56
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 6945.081944056919,
      "epoch": 0.07962482538415486,
      "grad_norm": 0.1355243529384343,
      "learning_rate": 7.777777777777778e-07,
      "loss": 0.0203,
      "num_tokens": 19437527.0,
      "reward": 0.2853341962077788,
      "reward_std": 0.28026825402464184,
      "rewards/reward_func": 0.28533417971006464,
      "step": 57
    },
    {
      "clip_ratio": 0.010294904433456915,
      "epoch": 0.08102175214528039,
      "grad_norm": 1.4215854683138216,
      "learning_rate": 7.916666666666666e-07,
      "loss": 0.0211,
      "step": 58
    },
    {
      "clip_ratio": 0.010296880932790893,
      "epoch": 0.0824186789064059,
      "grad_norm": 0.1328390444888187,
      "learning_rate": 8.055555555555556e-07,
      "loss": 0.0208,
      "step": 59
    },
    {
      "clip_ratio": 0.010402131692639418,
      "epoch": 0.08381560566753143,
      "grad_norm": 0.10630427559017747,
      "learning_rate": 8.194444444444443e-07,
      "loss": 0.0207,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 5832.984967912947,
      "epoch": 0.08521253242865695,
      "grad_norm": 0.2568979639198368,
      "learning_rate": 8.333333333333333e-07,
      "loss": 0.0135,
      "num_tokens": 20669861.0,
      "reward": 0.37696379157049315,
      "reward_std": 0.3524509519338608,
      "rewards/reward_func": 0.37696376868656706,
      "step": 61
    },
    {
      "clip_ratio": 0.011475847900978156,
      "epoch": 0.08660945918978248,
      "grad_norm": 0.26568198717380825,
      "learning_rate": 8.472222222222222e-07,
      "loss": 0.0142,
      "step": 62
    },
    {
      "clip_ratio": 0.011013164584125792,
      "epoch": 0.088006385950908,
      "grad_norm": 0.5160311893436309,
      "learning_rate": 8.611111111111111e-07,
      "loss": 0.014,
      "step": 63
    },
    {
      "clip_ratio": 0.010899153238694583,
      "epoch": 0.08940331271203353,
      "grad_norm": 0.16319985648001487,
      "learning_rate": 8.75e-07,
      "loss": 0.0138,
      "step": 64
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 5130.444091796875,
      "epoch": 0.09080023947315904,
      "grad_norm": 0.28767641683692746,
      "learning_rate": 8.888888888888888e-07,
      "loss": 0.0025,
      "num_tokens": 21764127.0,
      "reward": 0.46981193338121685,
      "reward_std": 0.3410501836666039,
      "rewards/reward_func": 0.4698119152869497,
      "step": 65
    },
    {
      "clip_ratio": 0.011167634411581926,
      "epoch": 0.09219716623428457,
      "grad_norm": 1786685.0973493713,
      "learning_rate": 9.027777777777778e-07,
      "loss": 31.3025,
      "step": 66
    },
    {
      "clip_ratio": 0.011178254083331143,
      "epoch": 0.09359409299541009,
      "grad_norm": 0.2699319084211939,
      "learning_rate": 9.166666666666665e-07,
      "loss": 0.0029,
      "step": 67
    },
    {
      "clip_ratio": 0.010996093108717884,
      "epoch": 0.09499101975653562,
      "grad_norm": 0.1796173018225043,
      "learning_rate": 9.305555555555555e-07,
      "loss": 0.0026,
      "step": 68
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 5374.097150530134,
      "epoch": 0.09638794651766115,
      "grad_norm": 0.29248413368918635,
      "learning_rate": 9.444444444444444e-07,
      "loss": 0.0204,
      "num_tokens": 22905462.0,
      "reward": 0.42480308030332836,
      "reward_std": 0.3633535067949976,
      "rewards/reward_func": 0.4248030515653746,
      "step": 69
    },
    {
      "clip_ratio": 0.010870669968426228,
      "epoch": 0.09778487327878667,
      "grad_norm": 0.2916766016433164,
      "learning_rate": 9.583333333333334e-07,
      "loss": 0.0211,
      "step": 70
    },
    {
      "clip_ratio": 0.010642989072948694,
      "epoch": 0.0991818000399122,
      "grad_norm": 0.2574669164587784,
      "learning_rate": 9.722222222222222e-07,
      "loss": 0.0209,
      "step": 71
    },
    {
      "clip_ratio": 0.010603911941871047,
      "epoch": 0.10057872680103772,
      "grad_norm": 0.3923293286285201,
      "learning_rate": 9.861111111111112e-07,
      "loss": 0.0206,
      "step": 72
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 4780.178763253348,
      "epoch": 0.10197565356216325,
      "grad_norm": 0.3357433852104651,
      "learning_rate": 1e-06,
      "loss": 0.0111,
      "num_tokens": 23928872.0,
      "reward": 0.49998574065310614,
      "reward_std": 0.3773101898176329,
      "rewards/reward_func": 0.4999857119151524,
      "step": 73
    },
    {
      "clip_ratio": 0.012297127628698945,
      "epoch": 0.10337258032328876,
      "grad_norm": 0.24331462448033314,
      "learning_rate": 1e-06,
      "loss": 0.0117,
      "step": 74
    },
    {
      "clip_ratio": 0.012708199177203434,
      "epoch": 0.10476950708441429,
      "grad_norm": 0.6107245136629853,
      "learning_rate": 1e-06,
      "loss": 0.0114,
      "step": 75
    },
    {
      "clip_ratio": 0.013628917180800013,
      "epoch": 0.10616643384553981,
      "grad_norm": 0.1571665107730879,
      "learning_rate": 1e-06,
      "loss": 0.0112,
      "step": 76
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 5568.459463936942,
      "epoch": 0.10756336060666534,
      "grad_norm": 0.3303048447870775,
      "learning_rate": 1e-06,
      "loss": 0.0085,
      "num_tokens": 25108534.0,
      "reward": 0.3940093676958765,
      "reward_std": 0.32062976009079386,
      "rewards/reward_func": 0.39400935066597803,
      "step": 77
    },
    {
      "clip_ratio": 0.012926211846726281,
      "epoch": 0.10896028736779086,
      "grad_norm": 0.4586726422820888,
      "learning_rate": 1e-06,
      "loss": 0.0092,
      "step": 78
    },
    {
      "clip_ratio": 0.013429554013003196,
      "epoch": 0.11035721412891639,
      "grad_norm": 0.46850871504202196,
      "learning_rate": 1e-06,
      "loss": 0.0089,
      "step": 79
    },
    {
      "clip_ratio": 0.01421830172850085,
      "epoch": 0.1117541408900419,
      "grad_norm": 0.2694884119514331,
      "learning_rate": 1e-06,
      "loss": 0.0087,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 5048.066615513393,
      "epoch": 0.11315106765116743,
      "grad_norm": 0.4818094588291497,
      "learning_rate": 1e-06,
      "loss": 0.0003,
      "num_tokens": 26188824.0,
      "reward": 0.4587076764021601,
      "reward_std": 0.3670096014227186,
      "rewards/reward_func": 0.45870765511478695,
      "step": 81
    },
    {
      "clip_ratio": 0.014440144545265607,
      "epoch": 0.11454799441229295,
      "grad_norm": 39.00637105645605,
      "learning_rate": 1e-06,
      "loss": 0.0041,
      "step": 82
    },
    {
      "clip_ratio": 0.014427063853612967,
      "epoch": 0.11594492117341848,
      "grad_norm": 0.9150065270465777,
      "learning_rate": 1e-06,
      "loss": 0.0009,
      "step": 83
    },
    {
      "clip_ratio": 0.015046841731028897,
      "epoch": 0.117341847934544,
      "grad_norm": 0.24009263771304865,
      "learning_rate": 1e-06,
      "loss": 0.0007,
      "step": 84
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 4776.74507359096,
      "epoch": 0.11873877469566953,
      "grad_norm": 0.4772391378994009,
      "learning_rate": 1e-06,
      "loss": 0.0175,
      "num_tokens": 27213245.0,
      "reward": 0.46243097526686533,
      "reward_std": 0.3514525145292282,
      "rewards/reward_func": 0.4624309518507549,
      "step": 85
    },
    {
      "clip_ratio": 0.013508185378408857,
      "epoch": 0.12013570145679506,
      "grad_norm": 166639.4843736821,
      "learning_rate": 1e-06,
      "loss": 7.3176,
      "step": 86
    },
    {
      "clip_ratio": 0.013775874627754092,
      "epoch": 0.12153262821792057,
      "grad_norm": 27.738301939542676,
      "learning_rate": 1e-06,
      "loss": 0.019,
      "step": 87
    },
    {
      "clip_ratio": 0.014050438667514495,
      "epoch": 0.1229295549790461,
      "grad_norm": 2.3910898137109857,
      "learning_rate": 1e-06,
      "loss": 0.0182,
      "step": 88
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 4914.035984584263,
      "epoch": 0.12432648174017162,
      "grad_norm": 0.5313282819608155,
      "learning_rate": 1e-06,
      "loss": 0.0136,
      "num_tokens": 28265422.0,
      "reward": 0.493380460355963,
      "reward_std": 0.39978101530245375,
      "rewards/reward_func": 0.4933804316180093,
      "step": 89
    },
    {
      "clip_ratio": 0.017403810085462674,
      "epoch": 0.12572340850129715,
      "grad_norm": 2.5053609302596263,
      "learning_rate": 1e-06,
      "loss": 0.0153,
      "step": 90
    },
    {
      "clip_ratio": 0.018204916534679278,
      "epoch": 0.12712033526242267,
      "grad_norm": 1.3673142623484746,
      "learning_rate": 1e-06,
      "loss": 0.0147,
      "step": 91
    },
    {
      "clip_ratio": 0.019155576026865413,
      "epoch": 0.12851726202354818,
      "grad_norm": 0.285245341653243,
      "learning_rate": 1e-06,
      "loss": 0.0145,
      "step": 92
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 4607.285940987723,
      "epoch": 0.12991418878467373,
      "grad_norm": 0.5382478014891808,
      "learning_rate": 1e-06,
      "loss": 0.0251,
      "num_tokens": 29256889.0,
      "reward": 0.4975069910287857,
      "reward_std": 0.39376414673668997,
      "rewards/reward_func": 0.4975069612264633,
      "step": 93
    },
    {
      "clip_ratio": 0.017131159414670298,
      "epoch": 0.13131111554579925,
      "grad_norm": 277.06893594236533,
      "learning_rate": 1e-06,
      "loss": 0.041,
      "step": 94
    },
    {
      "clip_ratio": 0.017470890456544503,
      "epoch": 0.13270804230692476,
      "grad_norm": 0.595127984757126,
      "learning_rate": 1e-06,
      "loss": 0.0264,
      "step": 95
    },
    {
      "clip_ratio": 0.017918821956430162,
      "epoch": 0.13410496906805028,
      "grad_norm": 0.3727932311400719,
      "learning_rate": 1e-06,
      "loss": 0.026,
      "step": 96
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 5105.030866350447,
      "epoch": 0.13550189582917582,
      "grad_norm": 0.5204808748332284,
      "learning_rate": 1e-06,
      "loss": 0.0166,
      "num_tokens": 30346587.0,
      "reward": 0.40901294563497814,
      "reward_std": 0.3415069899388722,
      "rewards/reward_func": 0.4090129222188677,
      "step": 97
    },
    {
      "clip_ratio": 0.0174449899766062,
      "epoch": 0.13689882259030134,
      "grad_norm": 1.1528181627310667,
      "learning_rate": 1e-06,
      "loss": 0.0183,
      "step": 98
    },
    {
      "clip_ratio": 0.01785454393497535,
      "epoch": 0.13829574935142686,
      "grad_norm": 0.6193098655808096,
      "learning_rate": 1e-06,
      "loss": 0.0177,
      "step": 99
    },
    {
      "clip_ratio": 0.018185677073363746,
      "epoch": 0.13969267611255237,
      "grad_norm": 0.29035255969562146,
      "learning_rate": 1e-06,
      "loss": 0.0174,
      "step": 100
    }
  ],
  "logging_steps": 1,
  "max_steps": 715,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 2,
  "trial_name": null,
  "trial_params": null
}