{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9965010496850945,
  "eval_steps": 100,
  "global_step": 89,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 869.9017639160156,
      "epoch": 0.01119664100769769,
      "grad_norm": 0.1226201057434082,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3118622424080968,
      "reward_std": 0.21511575765907764,
      "rewards/accuracy_reward": 0.3118622424080968,
      "step": 1
    },
    {
      "completion_length": 898.8724212646484,
      "epoch": 0.02239328201539538,
      "grad_norm": 0.06424866616725922,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.27614795323461294,
      "reward_std": 0.20953080570325255,
      "rewards/accuracy_reward": 0.27614795323461294,
      "step": 2
    },
    {
      "completion_length": 908.4138870239258,
      "epoch": 0.03358992302309307,
      "grad_norm": 0.16260072588920593,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.33673468697816133,
      "reward_std": 0.19594408059492707,
      "rewards/accuracy_reward": 0.33673468697816133,
      "step": 3
    },
    {
      "completion_length": 861.184928894043,
      "epoch": 0.04478656403079076,
      "grad_norm": 0.1080532968044281,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.32525509409606457,
      "reward_std": 0.19159379415214062,
      "rewards/accuracy_reward": 0.32525509409606457,
      "step": 4
    },
    {
      "completion_length": 837.5133781433105,
      "epoch": 0.05598320503848846,
      "grad_norm": 0.05878443643450737,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3373724417760968,
      "reward_std": 0.1757088042795658,
      "rewards/accuracy_reward": 0.3373724417760968,
      "step": 5
    },
    {
      "completion_length": 870.5720481872559,
      "epoch": 0.06717984604618614,
      "grad_norm": 0.045949794352054596,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.30293366871774197,
      "reward_std": 0.2117346664890647,
      "rewards/accuracy_reward": 0.30293366871774197,
      "step": 6
    },
    {
      "completion_length": 846.4999847412109,
      "epoch": 0.07837648705388384,
      "grad_norm": 0.07094219326972961,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.34948979038745165,
      "reward_std": 0.20344350347295403,
      "rewards/accuracy_reward": 0.34948979038745165,
      "step": 7
    },
    {
      "completion_length": 865.3762550354004,
      "epoch": 0.08957312806158152,
      "grad_norm": 0.03545041009783745,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.295918358489871,
      "reward_std": 0.16994785657152534,
      "rewards/accuracy_reward": 0.295918358489871,
      "step": 8
    },
    {
      "completion_length": 852.7289352416992,
      "epoch": 0.10076976906927922,
      "grad_norm": 0.045342933386564255,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.34438775200396776,
      "reward_std": 0.1971532292664051,
      "rewards/accuracy_reward": 0.34438775200396776,
      "step": 9
    },
    {
      "completion_length": 843.637092590332,
      "epoch": 0.11196641007697691,
      "grad_norm": 0.026218140497803688,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3373724464327097,
      "reward_std": 0.19200961757451296,
      "rewards/accuracy_reward": 0.3373724464327097,
      "step": 10
    },
    {
      "completion_length": 887.6957740783691,
      "epoch": 0.1231630510846746,
      "grad_norm": 0.04722030088305473,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3252550936304033,
      "reward_std": 0.1759854848496616,
      "rewards/accuracy_reward": 0.3252550936304033,
      "step": 11
    },
    {
      "completion_length": 812.6587829589844,
      "epoch": 0.13435969209237228,
      "grad_norm": 0.03267057612538338,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.37244897335767746,
      "reward_std": 0.19372099824249744,
      "rewards/accuracy_reward": 0.37244897335767746,
      "step": 12
    },
    {
      "completion_length": 833.1377372741699,
      "epoch": 0.14555633310007,
      "grad_norm": 0.028331954032182693,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3769132560119033,
      "reward_std": 0.18134982651099563,
      "rewards/accuracy_reward": 0.3769132560119033,
      "step": 13
    },
    {
      "completion_length": 818.9668273925781,
      "epoch": 0.15675297410776767,
      "grad_norm": 0.04085038602352142,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3348214216530323,
      "reward_std": 0.1621781517751515,
      "rewards/accuracy_reward": 0.3348214216530323,
      "step": 14
    },
    {
      "completion_length": 840.0886306762695,
      "epoch": 0.16794961511546536,
      "grad_norm": 0.028016554191708565,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.33673468697816133,
      "reward_std": 0.13258170057088137,
      "rewards/accuracy_reward": 0.33673468697816133,
      "step": 15
    },
    {
      "completion_length": 826.784423828125,
      "epoch": 0.17914625612316304,
      "grad_norm": 0.015145727433264256,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3596938708797097,
      "reward_std": 0.18337781634181738,
      "rewards/accuracy_reward": 0.3596938708797097,
      "step": 16
    },
    {
      "completion_length": 803.470645904541,
      "epoch": 0.19034289713086075,
      "grad_norm": 0.026870042085647583,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.34502550307661295,
      "reward_std": 0.14298884477466345,
      "rewards/accuracy_reward": 0.34502550307661295,
      "step": 17
    },
    {
      "completion_length": 805.1179656982422,
      "epoch": 0.20153953813855843,
      "grad_norm": 0.01876048557460308,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.39732141699641943,
      "reward_std": 0.182099889498204,
      "rewards/accuracy_reward": 0.39732141699641943,
      "step": 18
    },
    {
      "completion_length": 805.6084022521973,
      "epoch": 0.21273617914625612,
      "grad_norm": 0.028706129640340805,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.40114795230329037,
      "reward_std": 0.1437693452462554,
      "rewards/accuracy_reward": 0.40114795230329037,
      "step": 19
    },
    {
      "completion_length": 813.2353134155273,
      "epoch": 0.22393282015395383,
      "grad_norm": 0.01931103877723217,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4183673420920968,
      "reward_std": 0.172919531352818,
      "rewards/accuracy_reward": 0.4183673420920968,
      "step": 20
    },
    {
      "completion_length": 822.071403503418,
      "epoch": 0.2351294611616515,
      "grad_norm": 0.013188039883971214,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.38839284982532263,
      "reward_std": 0.1737575875595212,
      "rewards/accuracy_reward": 0.38839284982532263,
      "step": 21
    },
    {
      "completion_length": 801.5975608825684,
      "epoch": 0.2463261021693492,
      "grad_norm": 0.01899808831512928,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.42028060369193554,
      "reward_std": 0.1760622572619468,
      "rewards/accuracy_reward": 0.42028060369193554,
      "step": 22
    },
    {
      "completion_length": 766.1243476867676,
      "epoch": 0.2575227431770469,
      "grad_norm": 0.015976959839463234,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.42155611142516136,
      "reward_std": 0.1581620997749269,
      "rewards/accuracy_reward": 0.42155611142516136,
      "step": 23
    },
    {
      "completion_length": 797.7608299255371,
      "epoch": 0.26871938418474456,
      "grad_norm": 0.016489814966917038,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3514030557125807,
      "reward_std": 0.14521192712709308,
      "rewards/accuracy_reward": 0.3514030557125807,
      "step": 24
    },
    {
      "completion_length": 812.9444999694824,
      "epoch": 0.27991602519244224,
      "grad_norm": 0.016817867755889893,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3150510140694678,
      "reward_std": 0.12147563928738236,
      "rewards/accuracy_reward": 0.3150510140694678,
      "step": 25
    },
    {
      "completion_length": 764.3099327087402,
      "epoch": 0.29111266620014,
      "grad_norm": 0.03271699696779251,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.39668366592377424,
      "reward_std": 0.1408104975707829,
      "rewards/accuracy_reward": 0.39668366592377424,
      "step": 26
    },
    {
      "completion_length": 781.2104415893555,
      "epoch": 0.30230930720783766,
      "grad_norm": 0.039215218275785446,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3903061170130968,
      "reward_std": 0.1559837511740625,
      "rewards/accuracy_reward": 0.3903061170130968,
      "step": 27
    },
    {
      "completion_length": 769.5580215454102,
      "epoch": 0.31350594821553535,
      "grad_norm": 0.036487579345703125,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3373724427074194,
      "reward_std": 0.12534773931838572,
      "rewards/accuracy_reward": 0.3373724427074194,
      "step": 28
    },
    {
      "completion_length": 733.8928413391113,
      "epoch": 0.32470258922323303,
      "grad_norm": 0.05340224504470825,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.45408162474632263,
      "reward_std": 0.15175338089466095,
      "rewards/accuracy_reward": 0.45408162474632263,
      "step": 29
    },
    {
      "completion_length": 784.5631217956543,
      "epoch": 0.3358992302309307,
      "grad_norm": 0.018003573641180992,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3507652999833226,
      "reward_std": 0.15918243979103863,
      "rewards/accuracy_reward": 0.3507652999833226,
      "step": 30
    },
    {
      "completion_length": 784.8794441223145,
      "epoch": 0.3470958712386284,
      "grad_norm": 0.0389319472014904,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3896683603525162,
      "reward_std": 0.14439309062436223,
      "rewards/accuracy_reward": 0.3896683603525162,
      "step": 31
    },
    {
      "completion_length": 768.0031700134277,
      "epoch": 0.3582925122463261,
      "grad_norm": 0.06784125417470932,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3903061170130968,
      "reward_std": 0.13286479515954852,
      "rewards/accuracy_reward": 0.3903061170130968,
      "step": 32
    },
    {
      "completion_length": 767.3890075683594,
      "epoch": 0.3694891532540238,
      "grad_norm": 0.05318552255630493,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4355867328122258,
      "reward_std": 0.13299752678722143,
      "rewards/accuracy_reward": 0.4355867328122258,
      "step": 33
    },
    {
      "completion_length": 762.5069999694824,
      "epoch": 0.3806857942617215,
      "grad_norm": 0.06330139189958572,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3941326458007097,
      "reward_std": 0.1394878369756043,
      "rewards/accuracy_reward": 0.3941326458007097,
      "step": 34
    },
    {
      "completion_length": 812.4967880249023,
      "epoch": 0.3918824352694192,
      "grad_norm": 0.10220023989677429,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.35395407397300005,
      "reward_std": 0.14999566785991192,
      "rewards/accuracy_reward": 0.35395407397300005,
      "step": 35
    },
    {
      "completion_length": 776.1479415893555,
      "epoch": 0.40307907627711687,
      "grad_norm": 0.11629489064216614,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4107142761349678,
      "reward_std": 0.1379204117693007,
      "rewards/accuracy_reward": 0.4107142761349678,
      "step": 36
    },
    {
      "completion_length": 740.758918762207,
      "epoch": 0.41427571728481455,
      "grad_norm": 0.17305724322795868,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4400510098785162,
      "reward_std": 0.10666065919212997,
      "rewards/accuracy_reward": 0.4400510098785162,
      "step": 37
    },
    {
      "completion_length": 779.7040634155273,
      "epoch": 0.42547235829251223,
      "grad_norm": 0.2567689120769501,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.39732142351567745,
      "reward_std": 0.13629061914980412,
      "rewards/accuracy_reward": 0.39732142351567745,
      "step": 38
    },
    {
      "completion_length": 757.4642677307129,
      "epoch": 0.4366689993002099,
      "grad_norm": 0.11120310425758362,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.38903060741722584,
      "reward_std": 0.1286280113272369,
      "rewards/accuracy_reward": 0.38903060741722584,
      "step": 39
    },
    {
      "completion_length": 741.5631141662598,
      "epoch": 0.44786564030790765,
      "grad_norm": 0.32911476492881775,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.42155611142516136,
      "reward_std": 0.13757976656779647,
      "rewards/accuracy_reward": 0.42155611142516136,
      "step": 40
    },
    {
      "completion_length": 751.7557182312012,
      "epoch": 0.45906228131560534,
      "grad_norm": 0.35593709349632263,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.45535713247954845,
      "reward_std": 0.14861705573275685,
      "rewards/accuracy_reward": 0.45535713247954845,
      "step": 41
    },
    {
      "completion_length": 687.6504974365234,
      "epoch": 0.470258922323303,
      "grad_norm": 0.13686875998973846,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4253826420754194,
      "reward_std": 0.1371575309894979,
      "rewards/accuracy_reward": 0.4253826420754194,
      "step": 42
    },
    {
      "completion_length": 789.5031661987305,
      "epoch": 0.4814555633310007,
      "grad_norm": 0.1229076161980629,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3954081544652581,
      "reward_std": 0.11565714655444026,
      "rewards/accuracy_reward": 0.3954081544652581,
      "step": 43
    },
    {
      "completion_length": 765.0101890563965,
      "epoch": 0.4926522043386984,
      "grad_norm": 0.17409658432006836,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.42665815632790327,
      "reward_std": 0.1478876848705113,
      "rewards/accuracy_reward": 0.42665815632790327,
      "step": 44
    },
    {
      "completion_length": 741.691951751709,
      "epoch": 0.5038488453463961,
      "grad_norm": 0.19942258298397064,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4323979513719678,
      "reward_std": 0.13503032480366528,
      "rewards/accuracy_reward": 0.4323979513719678,
      "step": 45
    },
    {
      "completion_length": 761.8099365234375,
      "epoch": 0.5150454863540938,
      "grad_norm": 0.13838064670562744,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.39413264486938715,
      "reward_std": 0.11990180658176541,
      "rewards/accuracy_reward": 0.39413264486938715,
      "step": 46
    },
    {
      "completion_length": 760.2117118835449,
      "epoch": 0.5262421273617914,
      "grad_norm": 0.1158590167760849,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.40369896963238716,
      "reward_std": 0.139884436968714,
      "rewards/accuracy_reward": 0.40369896963238716,
      "step": 47
    },
    {
      "completion_length": 765.3348045349121,
      "epoch": 0.5374387683694891,
      "grad_norm": 0.17314597964286804,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.36607141979038715,
      "reward_std": 0.12787795555777848,
      "rewards/accuracy_reward": 0.36607141979038715,
      "step": 48
    },
    {
      "completion_length": 778.6645202636719,
      "epoch": 0.5486354093771868,
      "grad_norm": 0.14133551716804504,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3813775386661291,
      "reward_std": 0.12611543014645576,
      "rewards/accuracy_reward": 0.3813775386661291,
      "step": 49
    },
    {
      "completion_length": 739.5631217956543,
      "epoch": 0.5598320503848845,
      "grad_norm": 0.10777953267097473,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4381377473473549,
      "reward_std": 0.12034807563759387,
      "rewards/accuracy_reward": 0.4381377473473549,
      "step": 50
    },
    {
      "completion_length": 777.4795761108398,
      "epoch": 0.5710286913925823,
      "grad_norm": 0.349727988243103,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3947704005986452,
      "reward_std": 0.1090469197370112,
      "rewards/accuracy_reward": 0.3947704005986452,
      "step": 51
    },
    {
      "completion_length": 749.7888832092285,
      "epoch": 0.58222533240028,
      "grad_norm": 0.45803913474082947,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3909438708797097,
      "reward_std": 0.13125263480469584,
      "rewards/accuracy_reward": 0.3909438708797097,
      "step": 52
    },
    {
      "completion_length": 731.6294479370117,
      "epoch": 0.5934219734079776,
      "grad_norm": 0.6970997452735901,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4923469312489033,
      "reward_std": 0.17931061703711748,
      "rewards/accuracy_reward": 0.4923469312489033,
      "step": 53
    },
    {
      "completion_length": 746.658145904541,
      "epoch": 0.6046186144156753,
      "grad_norm": 1.771281361579895,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3934948928654194,
      "reward_std": 0.1266880240291357,
      "rewards/accuracy_reward": 0.3934948928654194,
      "step": 54
    },
    {
      "completion_length": 707.3239631652832,
      "epoch": 0.615815255423373,
      "grad_norm": 3.0594873428344727,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.42729591205716133,
      "reward_std": 0.12295506754890084,
      "rewards/accuracy_reward": 0.42729591205716133,
      "step": 55
    },
    {
      "completion_length": 754.7155456542969,
      "epoch": 0.6270118964310707,
      "grad_norm": 0.8483421802520752,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.42665815632790327,
      "reward_std": 0.13817946589551866,
      "rewards/accuracy_reward": 0.42665815632790327,
      "step": 56
    },
    {
      "completion_length": 739.8411827087402,
      "epoch": 0.6382085374387684,
      "grad_norm": 5.133347988128662,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.43494897335767746,
      "reward_std": 0.1543779973872006,
      "rewards/accuracy_reward": 0.43494897335767746,
      "step": 57
    },
    {
      "completion_length": 747.3603248596191,
      "epoch": 0.6494051784464661,
      "grad_norm": 2.267815351486206,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.414540808647871,
      "reward_std": 0.14408436114899814,
      "rewards/accuracy_reward": 0.414540808647871,
      "step": 58
    },
    {
      "completion_length": 736.4936027526855,
      "epoch": 0.6606018194541637,
      "grad_norm": 1.127752661705017,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.400510192848742,
      "reward_std": 0.10821045027114451,
      "rewards/accuracy_reward": 0.400510192848742,
      "step": 59
    },
    {
      "completion_length": 752.1415672302246,
      "epoch": 0.6717984604618614,
      "grad_norm": 1.0189523696899414,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4126275419257581,
      "reward_std": 0.12289111129939556,
      "rewards/accuracy_reward": 0.4126275419257581,
      "step": 60
    },
    {
      "completion_length": 785.941951751709,
      "epoch": 0.6829951014695591,
      "grad_norm": 1.3853994607925415,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.36096938140690327,
      "reward_std": 0.14290725137107074,
      "rewards/accuracy_reward": 0.36096938140690327,
      "step": 61
    },
    {
      "completion_length": 736.9368476867676,
      "epoch": 0.6941917424772568,
      "grad_norm": 0.6334269046783447,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.43686223588883877,
      "reward_std": 0.13453931966796517,
      "rewards/accuracy_reward": 0.43686223588883877,
      "step": 62
    },
    {
      "completion_length": 745.4457702636719,
      "epoch": 0.7053883834849545,
      "grad_norm": 1.0885730981826782,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.45408162102103233,
      "reward_std": 0.12036089389584959,
      "rewards/accuracy_reward": 0.45408162102103233,
      "step": 63
    },
    {
      "completion_length": 755.409423828125,
      "epoch": 0.7165850244926522,
      "grad_norm": 0.753265917301178,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.452168358489871,
      "reward_std": 0.1216771425679326,
      "rewards/accuracy_reward": 0.452168358489871,
      "step": 64
    },
    {
      "completion_length": 767.2990875244141,
      "epoch": 0.72778166550035,
      "grad_norm": 0.7473352551460266,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4011479467153549,
      "reward_std": 0.12710385350510478,
      "rewards/accuracy_reward": 0.4011479467153549,
      "step": 65
    },
    {
      "completion_length": 749.1913146972656,
      "epoch": 0.7389783065080476,
      "grad_norm": 0.7118561267852783,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.33418366499245167,
      "reward_std": 0.1169286584481597,
      "rewards/accuracy_reward": 0.33418366499245167,
      "step": 66
    },
    {
      "completion_length": 749.1740875244141,
      "epoch": 0.7501749475157453,
      "grad_norm": 1.282776951789856,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4317601975053549,
      "reward_std": 0.1398092587478459,
      "rewards/accuracy_reward": 0.4317601975053549,
      "step": 67
    },
    {
      "completion_length": 725.507640838623,
      "epoch": 0.761371588523443,
      "grad_norm": 1.5337843894958496,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.41262754052877426,
      "reward_std": 0.1240762178786099,
      "rewards/accuracy_reward": 0.41262754052877426,
      "step": 68
    },
    {
      "completion_length": 762.5490913391113,
      "epoch": 0.7725682295311407,
      "grad_norm": 0.6803719997406006,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3998724380508065,
      "reward_std": 0.14388926466926932,
      "rewards/accuracy_reward": 0.3998724380508065,
      "step": 69
    },
    {
      "completion_length": 736.7321319580078,
      "epoch": 0.7837648705388384,
      "grad_norm": 0.8348100781440735,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4400510163977742,
      "reward_std": 0.16068750014528632,
      "rewards/accuracy_reward": 0.4400510163977742,
      "step": 70
    },
    {
      "completion_length": 776.1109466552734,
      "epoch": 0.794961511546536,
      "grad_norm": 0.8664807677268982,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.37436223588883877,
      "reward_std": 0.1090405157301575,
      "rewards/accuracy_reward": 0.37436223588883877,
      "step": 71
    },
    {
      "completion_length": 740.8290710449219,
      "epoch": 0.8061581525542337,
      "grad_norm": 0.653354823589325,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.422193868085742,
      "reward_std": 0.10819763550534844,
      "rewards/accuracy_reward": 0.422193868085742,
      "step": 72
    },
    {
      "completion_length": 757.2959022521973,
      "epoch": 0.8173547935619314,
      "grad_norm": 3.8601436614990234,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4808673430234194,
      "reward_std": 0.1389840147458017,
      "rewards/accuracy_reward": 0.4808673430234194,
      "step": 73
    },
    {
      "completion_length": 764.6084060668945,
      "epoch": 0.8285514345696291,
      "grad_norm": 3.1642909049987793,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3864795845001936,
      "reward_std": 0.1433918485417962,
      "rewards/accuracy_reward": 0.3864795845001936,
      "step": 74
    },
    {
      "completion_length": 718.1115913391113,
      "epoch": 0.8397480755773268,
      "grad_norm": 0.8680428862571716,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.43303570430725813,
      "reward_std": 0.10496690985746682,
      "rewards/accuracy_reward": 0.43303570430725813,
      "step": 75
    },
    {
      "completion_length": 827.1135025024414,
      "epoch": 0.8509447165850245,
      "grad_norm": 1.043720006942749,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.40178570710122585,
      "reward_std": 0.13079995079897344,
      "rewards/accuracy_reward": 0.40178570710122585,
      "step": 76
    },
    {
      "completion_length": 726.0165634155273,
      "epoch": 0.8621413575927221,
      "grad_norm": 0.9036036729812622,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4693877473473549,
      "reward_std": 0.1229550673160702,
      "rewards/accuracy_reward": 0.4693877473473549,
      "step": 77
    },
    {
      "completion_length": 758.820140838623,
      "epoch": 0.8733379986004198,
      "grad_norm": 1.107183575630188,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.389030606020242,
      "reward_std": 0.11657038610428572,
      "rewards/accuracy_reward": 0.389030606020242,
      "step": 78
    },
    {
      "completion_length": 751.1798362731934,
      "epoch": 0.8845346396081175,
      "grad_norm": 1.0681264400482178,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4068877473473549,
      "reward_std": 0.11915174382738769,
      "rewards/accuracy_reward": 0.4068877473473549,
      "step": 79
    },
    {
      "completion_length": 769.7729415893555,
      "epoch": 0.8957312806158153,
      "grad_norm": 1.9977368116378784,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.41900509502738714,
      "reward_std": 0.1269023446366191,
      "rewards/accuracy_reward": 0.41900509502738714,
      "step": 80
    },
    {
      "completion_length": 806.9208984375,
      "epoch": 0.906927921623513,
      "grad_norm": 2.3192858695983887,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.45408162474632263,
      "reward_std": 0.1139010286424309,
      "rewards/accuracy_reward": 0.45408162474632263,
      "step": 81
    },
    {
      "completion_length": 756.2856941223145,
      "epoch": 0.9181245626312107,
      "grad_norm": 1.9081544876098633,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.42091835755854845,
      "reward_std": 0.1216771425679326,
      "rewards/accuracy_reward": 0.42091835755854845,
      "step": 82
    },
    {
      "completion_length": 810.2716598510742,
      "epoch": 0.9293212036389084,
      "grad_norm": 3.4622349739074707,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3686224380508065,
      "reward_std": 0.1060064712073654,
      "rewards/accuracy_reward": 0.3686224380508065,
      "step": 83
    },
    {
      "completion_length": 720.8679637908936,
      "epoch": 0.940517844646606,
      "grad_norm": 2.7656006813049316,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4355867253616452,
      "reward_std": 0.13207787927240133,
      "rewards/accuracy_reward": 0.4355867253616452,
      "step": 84
    },
    {
      "completion_length": 760.388370513916,
      "epoch": 0.9517144856543037,
      "grad_norm": 5.173759937286377,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4317601937800646,
      "reward_std": 0.12493191100656986,
      "rewards/accuracy_reward": 0.4317601937800646,
      "step": 85
    },
    {
      "completion_length": 746.6460304260254,
      "epoch": 0.9629111266620014,
      "grad_norm": 3.111050844192505,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.44579080678522587,
      "reward_std": 0.09203596017323434,
      "rewards/accuracy_reward": 0.44579080678522587,
      "step": 86
    },
    {
      "completion_length": 719.7404174804688,
      "epoch": 0.9741077676696991,
      "grad_norm": 2.178539276123047,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4253826439380646,
      "reward_std": 0.1273181731812656,
      "rewards/accuracy_reward": 0.4253826439380646,
      "step": 87
    },
    {
      "completion_length": 769.220645904541,
      "epoch": 0.9853044086773968,
      "grad_norm": 1.5965837240219116,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.4075255058705807,
      "reward_std": 0.12709103501401842,
      "rewards/accuracy_reward": 0.4075255058705807,
      "step": 88
    },
    {
      "completion_length": 774.408145904541,
      "epoch": 0.9965010496850945,
      "grad_norm": 2.0273616313934326,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.3966836631298065,
      "reward_std": 0.11341002094559371,
      "rewards/accuracy_reward": 0.3966836631298065,
      "step": 89
    },
    {
      "epoch": 0.9965010496850945,
      "step": 89,
      "total_flos": 0.0,
      "train_loss": 0.0,
      "train_runtime": 2.2646,
      "train_samples_per_second": 8831.586,
      "train_steps_per_second": 39.301
    }
  ],
  "logging_steps": 1,
  "max_steps": 89,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 20,
  "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": 2,
  "trial_name": null,
  "trial_params": null
}