{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.24,
  "eval_steps": 500,
  "global_step": 450,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 282.35938358306885,
      "epoch": 0.0010666666666666667,
      "grad_norm": 0.1740607843550731,
      "kl": 0.0,
      "learning_rate": 7.142857142857142e-08,
      "loss": -0.0,
      "reward": 0.026041667442768812,
      "reward_std": 0.06856952514499426,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.026041667442768812,
      "step": 2
    },
    {
      "completion_length": 255.33334159851074,
      "epoch": 0.0021333333333333334,
      "grad_norm": 0.1305423021351322,
      "kl": 0.00047016143798828125,
      "learning_rate": 1.4285714285714285e-07,
      "loss": 0.0,
      "reward": 0.02083333395421505,
      "reward_std": 0.053838133811950684,
      "rewards/equation_reward_func": 0.0026041667442768812,
      "rewards/format_reward_func": 0.01822916720993817,
      "step": 4
    },
    {
      "completion_length": 281.42969703674316,
      "epoch": 0.0032,
      "grad_norm": 0.19750475097577522,
      "kl": 0.0004911422729492188,
      "learning_rate": 2.1428571428571426e-07,
      "loss": 0.0,
      "reward": 0.033854167675599456,
      "reward_std": 0.085579180624336,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.033854167675599456,
      "step": 6
    },
    {
      "completion_length": 268.5130271911621,
      "epoch": 0.004266666666666667,
      "grad_norm": 0.194796000336386,
      "kl": 0.0005019903182983398,
      "learning_rate": 2.857142857142857e-07,
      "loss": 0.0,
      "reward": 0.03385416744276881,
      "reward_std": 0.08557918202131987,
      "rewards/equation_reward_func": 0.0052083334885537624,
      "rewards/format_reward_func": 0.02864583395421505,
      "step": 8
    },
    {
      "completion_length": 257.10938262939453,
      "epoch": 0.005333333333333333,
      "grad_norm": 0.292373371011067,
      "kl": 0.0005332231521606445,
      "learning_rate": 3.5714285714285716e-07,
      "loss": 0.0,
      "reward": 0.041666667675599456,
      "reward_std": 0.11276369681581855,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.041666667675599456,
      "step": 10
    },
    {
      "completion_length": 254.213547706604,
      "epoch": 0.0064,
      "grad_norm": 0.24542351307046964,
      "kl": 0.0005249977111816406,
      "learning_rate": 4.285714285714285e-07,
      "loss": 0.0,
      "reward": 0.033854167675599456,
      "reward_std": 0.09066661400720477,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.033854167675599456,
      "step": 12
    },
    {
      "completion_length": 240.59115505218506,
      "epoch": 0.007466666666666667,
      "grad_norm": 0.1629126525505696,
      "kl": 0.000682830810546875,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 0.03125000069849193,
      "reward_std": 0.08330091601237655,
      "rewards/equation_reward_func": 0.0026041667442768812,
      "rewards/format_reward_func": 0.02864583395421505,
      "step": 14
    },
    {
      "completion_length": 240.75261116027832,
      "epoch": 0.008533333333333334,
      "grad_norm": 0.3538782190019136,
      "kl": 0.0009417533874511719,
      "learning_rate": 4.999740409224932e-07,
      "loss": 0.0,
      "reward": 0.04947916814126074,
      "reward_std": 0.12977335462346673,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.04947916814126074,
      "step": 16
    },
    {
      "completion_length": 235.65886306762695,
      "epoch": 0.0096,
      "grad_norm": 0.37483939189307014,
      "kl": 0.0021331310272216797,
      "learning_rate": 4.998961690809627e-07,
      "loss": 0.0,
      "reward": 0.08333333558402956,
      "reward_std": 0.19386286428198218,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.08333333558402956,
      "step": 18
    },
    {
      "completion_length": 213.838547706604,
      "epoch": 0.010666666666666666,
      "grad_norm": 0.4353626183813759,
      "kl": 0.004715919494628906,
      "learning_rate": 4.997664006472578e-07,
      "loss": 0.0,
      "reward": 0.13281250465661287,
      "reward_std": 0.2736447872593999,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.13281250465661287,
      "step": 20
    },
    {
      "completion_length": 199.9505271911621,
      "epoch": 0.011733333333333333,
      "grad_norm": 0.35408470392901203,
      "kl": 0.007928848266601562,
      "learning_rate": 4.995847625707292e-07,
      "loss": 0.0,
      "reward": 0.21093750512227416,
      "reward_std": 0.37474397476762533,
      "rewards/equation_reward_func": 0.0052083334885537624,
      "rewards/format_reward_func": 0.2057291711680591,
      "step": 22
    },
    {
      "completion_length": 200.46094417572021,
      "epoch": 0.0128,
      "grad_norm": 0.39276482904098864,
      "kl": 0.01102447509765625,
      "learning_rate": 4.993512925726318e-07,
      "loss": 0.0,
      "reward": 0.2760416744276881,
      "reward_std": 0.40983634255826473,
      "rewards/equation_reward_func": 0.0052083334885537624,
      "rewards/format_reward_func": 0.27083334140479565,
      "step": 24
    },
    {
      "completion_length": 148.19792079925537,
      "epoch": 0.013866666666666666,
      "grad_norm": 0.48262457004852183,
      "kl": 0.025241851806640625,
      "learning_rate": 4.990660391382923e-07,
      "loss": 0.0,
      "reward": 0.4947916865348816,
      "reward_std": 0.48596240021288395,
      "rewards/equation_reward_func": 0.0,
      "rewards/format_reward_func": 0.4947916865348816,
      "step": 26
    },
    {
      "completion_length": 114.90885829925537,
      "epoch": 0.014933333333333333,
      "grad_norm": 0.38461913898848776,
      "kl": 0.0433349609375,
      "learning_rate": 4.987290615070384e-07,
      "loss": 0.0,
      "reward": 0.7291666828095913,
      "reward_std": 0.42544535594061017,
      "rewards/equation_reward_func": 0.0078125,
      "rewards/format_reward_func": 0.7213541828095913,
      "step": 28
    },
    {
      "completion_length": 105.90885734558105,
      "epoch": 0.016,
      "grad_norm": 0.4217166681599687,
      "kl": 0.0572509765625,
      "learning_rate": 4.983404296598978e-07,
      "loss": 0.0001,
      "reward": 0.7682291828095913,
      "reward_std": 0.399359080940485,
      "rewards/equation_reward_func": 0.0026041667442768812,
      "rewards/format_reward_func": 0.7656250149011612,
      "step": 30
    },
    {
      "completion_length": 85.84635543823242,
      "epoch": 0.017066666666666667,
      "grad_norm": 0.43581276711650174,
      "kl": 0.05517578125,
      "learning_rate": 4.979002243050646e-07,
      "loss": 0.0001,
      "reward": 0.8697916828095913,
      "reward_std": 0.2932830383069813,
      "rewards/equation_reward_func": 0.0052083334885537624,
      "rewards/format_reward_func": 0.8645833544433117,
      "step": 32
    },
    {
      "completion_length": 70.50521039962769,
      "epoch": 0.018133333333333335,
      "grad_norm": 0.27643278807711624,
      "kl": 0.0776519775390625,
      "learning_rate": 4.974085368611381e-07,
      "loss": 0.0001,
      "reward": 0.9192708507180214,
      "reward_std": 0.20361881935968995,
      "rewards/equation_reward_func": 0.0052083334885537624,
      "rewards/format_reward_func": 0.9140625186264515,
      "step": 34
    },
    {
      "completion_length": 75.24218940734863,
      "epoch": 0.0192,
      "grad_norm": 0.2824240988052863,
      "kl": 0.064453125,
      "learning_rate": 4.968654694381379e-07,
      "loss": 0.0001,
      "reward": 0.9140625223517418,
      "reward_std": 0.20272048702463508,
      "rewards/equation_reward_func": 0.0026041667442768812,
      "rewards/format_reward_func": 0.9114583544433117,
      "step": 36
    },
    {
      "completion_length": 63.29948019981384,
      "epoch": 0.020266666666666665,
      "grad_norm": 0.2805616280859793,
      "kl": 0.0658111572265625,
      "learning_rate": 4.962711348162987e-07,
      "loss": 0.0001,
      "reward": 0.9479166828095913,
      "reward_std": 0.17720681754872203,
      "rewards/equation_reward_func": 0.010416666977107525,
      "rewards/format_reward_func": 0.9375000149011612,
      "step": 38
    },
    {
      "completion_length": 54.588543176651,
      "epoch": 0.021333333333333333,
      "grad_norm": 0.23145153521903244,
      "kl": 0.0833587646484375,
      "learning_rate": 4.956256564226487e-07,
      "loss": 0.0001,
      "reward": 0.9791666902601719,
      "reward_std": 0.09592872625216842,
      "rewards/equation_reward_func": 0.007812500232830644,
      "rewards/format_reward_func": 0.9713541865348816,
      "step": 40
    },
    {
      "completion_length": 55.24739694595337,
      "epoch": 0.0224,
      "grad_norm": 0.2338979387975897,
      "kl": 0.09307861328125,
      "learning_rate": 4.949291683053768e-07,
      "loss": 0.0001,
      "reward": 0.9791666902601719,
      "reward_std": 0.10311973933130503,
      "rewards/equation_reward_func": 0.007812500232830644,
      "rewards/format_reward_func": 0.9713541902601719,
      "step": 42
    },
    {
      "completion_length": 54.289063453674316,
      "epoch": 0.023466666666666667,
      "grad_norm": 0.1329419349418815,
      "kl": 0.076385498046875,
      "learning_rate": 4.941818151059955e-07,
      "loss": 0.0001,
      "reward": 1.005208358168602,
      "reward_std": 0.0589255653321743,
      "rewards/equation_reward_func": 0.013020833721384406,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 44
    },
    {
      "completion_length": 57.59895944595337,
      "epoch": 0.024533333333333334,
      "grad_norm": 1.2311340623228515,
      "kl": 0.147796630859375,
      "learning_rate": 4.933837520293017e-07,
      "loss": 0.0001,
      "reward": 0.9765625298023224,
      "reward_std": 0.09066661214455962,
      "rewards/equation_reward_func": 0.0052083334885537624,
      "rewards/format_reward_func": 0.9713541865348816,
      "step": 46
    },
    {
      "completion_length": 51.78385543823242,
      "epoch": 0.0256,
      "grad_norm": 0.6985959466521252,
      "kl": 0.1004638671875,
      "learning_rate": 4.925351448111454e-07,
      "loss": 0.0001,
      "reward": 1.0234375298023224,
      "reward_std": 0.08557918202131987,
      "rewards/equation_reward_func": 0.02864583395421505,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 48
    },
    {
      "completion_length": 51.58333420753479,
      "epoch": 0.02666666666666667,
      "grad_norm": 0.12594191102876573,
      "kl": 0.10333251953125,
      "learning_rate": 4.91636169684011e-07,
      "loss": 0.0001,
      "reward": 1.0156250223517418,
      "reward_std": 0.07461797958239913,
      "rewards/equation_reward_func": 0.023437500232830644,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 50
    },
    {
      "completion_length": 51.73177242279053,
      "epoch": 0.027733333333333332,
      "grad_norm": 0.08753934735556794,
      "kl": 0.10040283203125,
      "learning_rate": 4.906870133404186e-07,
      "loss": 0.0001,
      "reward": 1.0182291865348816,
      "reward_std": 0.0549764484167099,
      "rewards/equation_reward_func": 0.020833333721384406,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 52
    },
    {
      "completion_length": 51.546876668930054,
      "epoch": 0.0288,
      "grad_norm": 0.17866705350514983,
      "kl": 0.1055908203125,
      "learning_rate": 4.896878728941531e-07,
      "loss": 0.0001,
      "reward": 1.0078125149011612,
      "reward_std": 0.046472438145428896,
      "rewards/equation_reward_func": 0.013020833721384406,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 54
    },
    {
      "completion_length": 51.22395944595337,
      "epoch": 0.029866666666666666,
      "grad_norm": 0.1730610808638548,
      "kl": 0.141632080078125,
      "learning_rate": 4.886389558393284e-07,
      "loss": 0.0001,
      "reward": 1.002604179084301,
      "reward_std": 0.051559869665652514,
      "rewards/equation_reward_func": 0.010416666977107525,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 56
    },
    {
      "completion_length": 51.179688930511475,
      "epoch": 0.030933333333333334,
      "grad_norm": 0.3129794370215895,
      "kl": 0.11151123046875,
      "learning_rate": 4.875404800072976e-07,
      "loss": 0.0001,
      "reward": 1.0364583656191826,
      "reward_std": 0.09294487768784165,
      "rewards/equation_reward_func": 0.03645833418704569,
      "rewards/format_reward_func": 1.0,
      "step": 58
    },
    {
      "completion_length": 51.335938930511475,
      "epoch": 0.032,
      "grad_norm": 0.25488432278587914,
      "kl": 0.124847412109375,
      "learning_rate": 4.86392673521415e-07,
      "loss": 0.0001,
      "reward": 1.0442708730697632,
      "reward_std": 0.11144645884633064,
      "rewards/equation_reward_func": 0.04687500116415322,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 60
    },
    {
      "completion_length": 49.33854293823242,
      "epoch": 0.03306666666666667,
      "grad_norm": 0.1521903024555833,
      "kl": 0.124847412109375,
      "learning_rate": 4.851957747496606e-07,
      "loss": 0.0001,
      "reward": 1.0338541939854622,
      "reward_std": 0.05953297670930624,
      "rewards/equation_reward_func": 0.0338541679084301,
      "rewards/format_reward_func": 1.0,
      "step": 62
    },
    {
      "completion_length": 48.43229293823242,
      "epoch": 0.034133333333333335,
      "grad_norm": 0.2524941713562646,
      "kl": 0.16082763671875,
      "learning_rate": 4.839500322551386e-07,
      "loss": 0.0002,
      "reward": 1.0442708507180214,
      "reward_std": 0.09732084395363927,
      "rewards/equation_reward_func": 0.04947916720993817,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 64
    },
    {
      "completion_length": 49.06250190734863,
      "epoch": 0.0352,
      "grad_norm": 0.14405133529333697,
      "kl": 0.1566162109375,
      "learning_rate": 4.826557047444563e-07,
      "loss": 0.0002,
      "reward": 1.028645858168602,
      "reward_std": 0.07426436804234982,
      "rewards/equation_reward_func": 0.03125000069849193,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 66
    },
    {
      "completion_length": 46.49739718437195,
      "epoch": 0.03626666666666667,
      "grad_norm": 0.21138777084113075,
      "kl": 0.166259765625,
      "learning_rate": 4.813130610139993e-07,
      "loss": 0.0002,
      "reward": 1.0729167088866234,
      "reward_std": 0.12511440878733993,
      "rewards/equation_reward_func": 0.07291666860692203,
      "rewards/format_reward_func": 1.0,
      "step": 68
    },
    {
      "completion_length": 45.867188692092896,
      "epoch": 0.037333333333333336,
      "grad_norm": 0.21057979945519745,
      "kl": 0.18072509765625,
      "learning_rate": 4.799223798941089e-07,
      "loss": 0.0002,
      "reward": 1.0729167088866234,
      "reward_std": 0.1658920031040907,
      "rewards/equation_reward_func": 0.07552083511836827,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 70
    },
    {
      "completion_length": 46.00260519981384,
      "epoch": 0.0384,
      "grad_norm": 0.22213959393069932,
      "kl": 0.2841796875,
      "learning_rate": 4.78483950191177e-07,
      "loss": 0.0003,
      "reward": 1.062500037252903,
      "reward_std": 0.12756996927782893,
      "rewards/equation_reward_func": 0.06250000116415322,
      "rewards/format_reward_func": 1.0,
      "step": 72
    },
    {
      "completion_length": 45.48177218437195,
      "epoch": 0.039466666666666664,
      "grad_norm": 0.25433299791887853,
      "kl": 0.2039794921875,
      "learning_rate": 4.769980706276687e-07,
      "loss": 0.0002,
      "reward": 1.0729167014360428,
      "reward_std": 0.13283371645957232,
      "rewards/equation_reward_func": 0.07291666837409139,
      "rewards/format_reward_func": 1.0,
      "step": 74
    },
    {
      "completion_length": 45.94270920753479,
      "epoch": 0.04053333333333333,
      "grad_norm": 0.197324342240113,
      "kl": 0.20538330078125,
      "learning_rate": 4.7546504978008595e-07,
      "loss": 0.0002,
      "reward": 1.083333358168602,
      "reward_std": 0.14913451066240668,
      "rewards/equation_reward_func": 0.08333333558402956,
      "rewards/format_reward_func": 1.0,
      "step": 76
    },
    {
      "completion_length": 44.643230676651,
      "epoch": 0.0416,
      "grad_norm": 0.19766456669835683,
      "kl": 0.210693359375,
      "learning_rate": 4.738852060148848e-07,
      "loss": 0.0002,
      "reward": 1.0807292014360428,
      "reward_std": 0.13949218904599547,
      "rewards/equation_reward_func": 0.08854166930541396,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 78
    },
    {
      "completion_length": 44.45052194595337,
      "epoch": 0.042666666666666665,
      "grad_norm": 0.18013367923974538,
      "kl": 0.2098388671875,
      "learning_rate": 4.722588674223593e-07,
      "loss": 0.0002,
      "reward": 1.1041667014360428,
      "reward_std": 0.1602703994140029,
      "rewards/equation_reward_func": 0.10677083651535213,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 80
    },
    {
      "completion_length": 44.59114694595337,
      "epoch": 0.04373333333333333,
      "grad_norm": 0.2089568877219373,
      "kl": 0.23651123046875,
      "learning_rate": 4.70586371748506e-07,
      "loss": 0.0002,
      "reward": 1.0729166939854622,
      "reward_std": 0.11073498986661434,
      "rewards/equation_reward_func": 0.07291667023673654,
      "rewards/format_reward_func": 1.0,
      "step": 82
    },
    {
      "completion_length": 44.25520968437195,
      "epoch": 0.0448,
      "grad_norm": 0.26774021355534683,
      "kl": 0.2432861328125,
      "learning_rate": 4.6886806632488363e-07,
      "loss": 0.0002,
      "reward": 1.0703125223517418,
      "reward_std": 0.14457962149754167,
      "rewards/equation_reward_func": 0.0729166679084301,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 84
    },
    {
      "completion_length": 45.085938453674316,
      "epoch": 0.04586666666666667,
      "grad_norm": 0.20779914302686117,
      "kl": 0.20538330078125,
      "learning_rate": 4.6710430799648143e-07,
      "loss": 0.0002,
      "reward": 1.0859375298023224,
      "reward_std": 0.13949218997731805,
      "rewards/equation_reward_func": 0.08593750209547579,
      "rewards/format_reward_func": 1.0,
      "step": 86
    },
    {
      "completion_length": 44.90885543823242,
      "epoch": 0.046933333333333334,
      "grad_norm": 0.20740948760165967,
      "kl": 0.24371337890625,
      "learning_rate": 4.652954630476127e-07,
      "loss": 0.0002,
      "reward": 1.0833333507180214,
      "reward_std": 0.13188487756997347,
      "rewards/equation_reward_func": 0.08854166837409139,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 88
    },
    {
      "completion_length": 44.250001192092896,
      "epoch": 0.048,
      "grad_norm": 0.20250885853430015,
      "kl": 0.20672607421875,
      "learning_rate": 4.6344190712584713e-07,
      "loss": 0.0002,
      "reward": 1.1015625521540642,
      "reward_std": 0.16360786417499185,
      "rewards/equation_reward_func": 0.10156250349245965,
      "rewards/format_reward_func": 1.0,
      "step": 90
    },
    {
      "completion_length": 44.132813930511475,
      "epoch": 0.04906666666666667,
      "grad_norm": 0.19451909809114717,
      "kl": 0.25396728515625,
      "learning_rate": 4.615440251639995e-07,
      "loss": 0.0003,
      "reward": 1.0703125447034836,
      "reward_std": 0.11827961495146155,
      "rewards/equation_reward_func": 0.07031250256113708,
      "rewards/format_reward_func": 1.0,
      "step": 92
    },
    {
      "completion_length": 43.86198043823242,
      "epoch": 0.050133333333333335,
      "grad_norm": 0.2114464718519318,
      "kl": 0.23681640625,
      "learning_rate": 4.596022113001894e-07,
      "loss": 0.0002,
      "reward": 1.109375037252903,
      "reward_std": 0.17087537981569767,
      "rewards/equation_reward_func": 0.10937500279396772,
      "rewards/format_reward_func": 1.0,
      "step": 94
    },
    {
      "completion_length": 43.35677218437195,
      "epoch": 0.0512,
      "grad_norm": 0.26777820236527494,
      "kl": 0.21124267578125,
      "learning_rate": 4.576168687959895e-07,
      "loss": 0.0002,
      "reward": 1.1197916939854622,
      "reward_std": 0.17219426156952977,
      "rewards/equation_reward_func": 0.11979167000390589,
      "rewards/format_reward_func": 1.0,
      "step": 96
    },
    {
      "completion_length": 43.971355676651,
      "epoch": 0.05226666666666667,
      "grad_norm": 0.22497833404680617,
      "kl": 0.21343994140625,
      "learning_rate": 4.555884099526793e-07,
      "loss": 0.0002,
      "reward": 1.0755208805203438,
      "reward_std": 0.12792357616126537,
      "rewards/equation_reward_func": 0.07552083535119891,
      "rewards/format_reward_func": 1.0,
      "step": 98
    },
    {
      "completion_length": 43.773438930511475,
      "epoch": 0.05333333333333334,
      "grad_norm": 0.20979703618125947,
      "kl": 0.2442626953125,
      "learning_rate": 4.5351725602562174e-07,
      "loss": 0.0002,
      "reward": 1.1145833656191826,
      "reward_std": 0.1417704545892775,
      "rewards/equation_reward_func": 0.11458333558402956,
      "rewards/format_reward_func": 1.0,
      "step": 100
    },
    {
      "completion_length": 44.796876668930054,
      "epoch": 0.0544,
      "grad_norm": 0.30411990281094664,
      "kl": 0.2236328125,
      "learning_rate": 4.514038371367791e-07,
      "loss": 0.0002,
      "reward": 1.085937537252903,
      "reward_std": 0.1542235817760229,
      "rewards/equation_reward_func": 0.08593750093132257,
      "rewards/format_reward_func": 1.0,
      "step": 102
    },
    {
      "completion_length": 44.39583444595337,
      "epoch": 0.055466666666666664,
      "grad_norm": 0.21533898443624652,
      "kl": 0.2440185546875,
      "learning_rate": 4.4924859218538936e-07,
      "loss": 0.0002,
      "reward": 1.1536458507180214,
      "reward_std": 0.15141277946531773,
      "rewards/equation_reward_func": 0.15364583488553762,
      "rewards/format_reward_func": 1.0,
      "step": 104
    },
    {
      "completion_length": 43.97135519981384,
      "epoch": 0.05653333333333333,
      "grad_norm": 0.15993730837299722,
      "kl": 0.21624755859375,
      "learning_rate": 4.470519687568185e-07,
      "loss": 0.0002,
      "reward": 1.130208358168602,
      "reward_std": 0.1776336650364101,
      "rewards/equation_reward_func": 0.13020833721384406,
      "rewards/format_reward_func": 1.0,
      "step": 106
    },
    {
      "completion_length": 44.664063692092896,
      "epoch": 0.0576,
      "grad_norm": 0.15852258859589255,
      "kl": 0.24371337890625,
      "learning_rate": 4.4481442302960923e-07,
      "loss": 0.0002,
      "reward": 1.0859375298023224,
      "reward_std": 0.11372047895565629,
      "rewards/equation_reward_func": 0.0885416679084301,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 108
    },
    {
      "completion_length": 45.56770896911621,
      "epoch": 0.058666666666666666,
      "grad_norm": 0.2312788417421762,
      "kl": 0.59844970703125,
      "learning_rate": 4.4253641968074505e-07,
      "loss": 0.0006,
      "reward": 1.1145833656191826,
      "reward_std": 0.19218612927943468,
      "rewards/equation_reward_func": 0.11718750442378223,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 110
    },
    {
      "completion_length": 45.84895992279053,
      "epoch": 0.05973333333333333,
      "grad_norm": 0.19052556511731025,
      "kl": 0.196533203125,
      "learning_rate": 4.402184317891501e-07,
      "loss": 0.0002,
      "reward": 1.0859375149011612,
      "reward_std": 0.10775277810171247,
      "rewards/equation_reward_func": 0.0859375016298145,
      "rewards/format_reward_func": 1.0,
      "step": 112
    },
    {
      "completion_length": 45.804689168930054,
      "epoch": 0.0608,
      "grad_norm": 0.16234373693343537,
      "kl": 0.22698974609375,
      "learning_rate": 4.37860940737443e-07,
      "loss": 0.0002,
      "reward": 1.1223958656191826,
      "reward_std": 0.10468817735090852,
      "rewards/equation_reward_func": 0.12239583698101342,
      "rewards/format_reward_func": 1.0,
      "step": 114
    },
    {
      "completion_length": 45.35416793823242,
      "epoch": 0.06186666666666667,
      "grad_norm": 0.22634476300945777,
      "kl": 0.2115478515625,
      "learning_rate": 4.354644361119671e-07,
      "loss": 0.0002,
      "reward": 1.0885416865348816,
      "reward_std": 0.12379553122445941,
      "rewards/equation_reward_func": 0.08854166767559946,
      "rewards/format_reward_func": 1.0,
      "step": 116
    },
    {
      "completion_length": 45.61198019981384,
      "epoch": 0.06293333333333333,
      "grad_norm": 0.21820113853204484,
      "kl": 0.3489990234375,
      "learning_rate": 4.3302941560111716e-07,
      "loss": 0.0003,
      "reward": 1.1119791939854622,
      "reward_std": 0.13397203059867024,
      "rewards/equation_reward_func": 0.11197916860692203,
      "rewards/format_reward_func": 1.0,
      "step": 118
    },
    {
      "completion_length": 44.20833468437195,
      "epoch": 0.064,
      "grad_norm": 0.2387693659050214,
      "kl": 0.21429443359375,
      "learning_rate": 4.3055638489198236e-07,
      "loss": 0.0002,
      "reward": 1.0833333656191826,
      "reward_std": 0.11888702632859349,
      "rewards/equation_reward_func": 0.08333333511836827,
      "rewards/format_reward_func": 1.0,
      "step": 120
    },
    {
      "completion_length": 44.02083468437195,
      "epoch": 0.06506666666666666,
      "grad_norm": 0.21362033365285313,
      "kl": 0.21478271484375,
      "learning_rate": 4.280458575653296e-07,
      "loss": 0.0002,
      "reward": 1.0937500223517418,
      "reward_std": 0.09985290700569749,
      "rewards/equation_reward_func": 0.09375000302679837,
      "rewards/format_reward_func": 1.0,
      "step": 122
    },
    {
      "completion_length": 45.31770944595337,
      "epoch": 0.06613333333333334,
      "grad_norm": 0.14881831765766107,
      "kl": 0.22314453125,
      "learning_rate": 4.2549835498894665e-07,
      "loss": 0.0002,
      "reward": 1.0703125149011612,
      "reward_std": 0.0941580655053258,
      "rewards/equation_reward_func": 0.07031250093132257,
      "rewards/format_reward_func": 1.0,
      "step": 124
    },
    {
      "completion_length": 45.281251430511475,
      "epoch": 0.0672,
      "grad_norm": 0.17735570373907916,
      "kl": 0.1910400390625,
      "learning_rate": 4.229144062093679e-07,
      "loss": 0.0002,
      "reward": 1.1510417014360428,
      "reward_std": 0.1525510959327221,
      "rewards/equation_reward_func": 0.15104167046956718,
      "rewards/format_reward_func": 1.0,
      "step": 126
    },
    {
      "completion_length": 46.502605676651,
      "epoch": 0.06826666666666667,
      "grad_norm": 0.21334686547504042,
      "kl": 0.2138671875,
      "learning_rate": 4.2029454784200675e-07,
      "loss": 0.0002,
      "reward": 1.091145858168602,
      "reward_std": 0.14115716284140944,
      "rewards/equation_reward_func": 0.09635416767559946,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 128
    },
    {
      "completion_length": 46.296876430511475,
      "epoch": 0.06933333333333333,
      "grad_norm": 0.25099329091191924,
      "kl": 0.2186279296875,
      "learning_rate": 4.1763932395971433e-07,
      "loss": 0.0002,
      "reward": 1.125000037252903,
      "reward_std": 0.1483498103916645,
      "rewards/equation_reward_func": 0.12760417093522847,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 130
    },
    {
      "completion_length": 44.74739694595337,
      "epoch": 0.0704,
      "grad_norm": 0.1981837661591498,
      "kl": 0.19793701171875,
      "learning_rate": 4.1494928597979117e-07,
      "loss": 0.0002,
      "reward": 1.1145833507180214,
      "reward_std": 0.1336184199899435,
      "rewards/equation_reward_func": 0.1171875016298145,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 132
    },
    {
      "completion_length": 47.390626430511475,
      "epoch": 0.07146666666666666,
      "grad_norm": 0.1835328837015611,
      "kl": 0.20135498046875,
      "learning_rate": 4.122249925494726e-07,
      "loss": 0.0002,
      "reward": 1.0911458730697632,
      "reward_std": 0.1366065163165331,
      "rewards/equation_reward_func": 0.09375000279396772,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 134
    },
    {
      "completion_length": 44.554689168930054,
      "epoch": 0.07253333333333334,
      "grad_norm": 0.20159012545509752,
      "kl": 0.22216796875,
      "learning_rate": 4.094670094299131e-07,
      "loss": 0.0002,
      "reward": 1.156250037252903,
      "reward_std": 0.14493323443457484,
      "rewards/equation_reward_func": 0.1588541732635349,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 136
    },
    {
      "completion_length": 44.72656440734863,
      "epoch": 0.0736,
      "grad_norm": 0.11786284329618611,
      "kl": 0.2099609375,
      "learning_rate": 4.066759093786931e-07,
      "loss": 0.0002,
      "reward": 1.177083358168602,
      "reward_std": 0.1451870333403349,
      "rewards/equation_reward_func": 0.17708333814516664,
      "rewards/format_reward_func": 1.0,
      "step": 138
    },
    {
      "completion_length": 45.02864742279053,
      "epoch": 0.07466666666666667,
      "grad_norm": 0.22367961788022572,
      "kl": 0.2369384765625,
      "learning_rate": 4.038522720308732e-07,
      "loss": 0.0002,
      "reward": 1.119791716337204,
      "reward_std": 0.13678119610995054,
      "rewards/equation_reward_func": 0.11979166977107525,
      "rewards/format_reward_func": 1.0,
      "step": 140
    },
    {
      "completion_length": 45.30989670753479,
      "epoch": 0.07573333333333333,
      "grad_norm": 0.14269442344109934,
      "kl": 0.2017822265625,
      "learning_rate": 4.009966837786194e-07,
      "loss": 0.0002,
      "reward": 1.0677083507180214,
      "reward_std": 0.09301975090056658,
      "rewards/equation_reward_func": 0.06770833441987634,
      "rewards/format_reward_func": 1.0,
      "step": 142
    },
    {
      "completion_length": 45.94270968437195,
      "epoch": 0.0768,
      "grad_norm": 0.24577923545533975,
      "kl": 0.7618408203125,
      "learning_rate": 3.981097376494259e-07,
      "loss": 0.0008,
      "reward": 1.1380208432674408,
      "reward_std": 0.1372155612334609,
      "rewards/equation_reward_func": 0.14062500069849193,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 144
    },
    {
      "completion_length": 47.07812547683716,
      "epoch": 0.07786666666666667,
      "grad_norm": 0.19129102899241263,
      "kl": 0.21478271484375,
      "learning_rate": 3.951920331829592e-07,
      "loss": 0.0002,
      "reward": 1.0885417014360428,
      "reward_std": 0.11959685944020748,
      "rewards/equation_reward_func": 0.09114583535119891,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 146
    },
    {
      "completion_length": 45.51302218437195,
      "epoch": 0.07893333333333333,
      "grad_norm": 0.21814326010845567,
      "kl": 0.2347412109375,
      "learning_rate": 3.922441763065506e-07,
      "loss": 0.0002,
      "reward": 1.1093750298023224,
      "reward_std": 0.1479919538833201,
      "rewards/equation_reward_func": 0.10937500116415322,
      "rewards/format_reward_func": 1.0,
      "step": 148
    },
    {
      "completion_length": 45.59635543823242,
      "epoch": 0.08,
      "grad_norm": 0.18274820151952206,
      "kl": 0.2232666015625,
      "learning_rate": 3.8926677920936093e-07,
      "loss": 0.0002,
      "reward": 1.0807292014360428,
      "reward_std": 0.1173159833997488,
      "rewards/equation_reward_func": 0.08072916837409139,
      "rewards/format_reward_func": 1.0,
      "step": 150
    },
    {
      "completion_length": 45.554688930511475,
      "epoch": 0.08106666666666666,
      "grad_norm": 0.15633525103555487,
      "kl": 0.2222900390625,
      "learning_rate": 3.862604602152464e-07,
      "loss": 0.0002,
      "reward": 1.1171875223517418,
      "reward_std": 0.11152132926508784,
      "rewards/equation_reward_func": 0.11979166977107525,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 152
    },
    {
      "completion_length": 46.007813692092896,
      "epoch": 0.08213333333333334,
      "grad_norm": 0.19415601668961927,
      "kl": 0.3138427734375,
      "learning_rate": 3.8322584365434934e-07,
      "loss": 0.0003,
      "reward": 1.1145833656191826,
      "reward_std": 0.1543982638977468,
      "rewards/equation_reward_func": 0.11718750325962901,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 154
    },
    {
      "completion_length": 46.99739694595337,
      "epoch": 0.0832,
      "grad_norm": 0.14577695713107947,
      "kl": 0.24224853515625,
      "learning_rate": 3.8016355973344173e-07,
      "loss": 0.0002,
      "reward": 1.0598958656191826,
      "reward_std": 0.09180068224668503,
      "rewards/equation_reward_func": 0.06250000209547579,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 156
    },
    {
      "completion_length": 46.22135543823242,
      "epoch": 0.08426666666666667,
      "grad_norm": 0.21784331239552168,
      "kl": 0.22943115234375,
      "learning_rate": 3.7707424440504863e-07,
      "loss": 0.0002,
      "reward": 1.1484375223517418,
      "reward_std": 0.11511683976277709,
      "rewards/equation_reward_func": 0.14843750465661287,
      "rewards/format_reward_func": 1.0,
      "step": 158
    },
    {
      "completion_length": 46.17708444595337,
      "epoch": 0.08533333333333333,
      "grad_norm": 0.24049435923447107,
      "kl": 0.27606201171875,
      "learning_rate": 3.739585392353787e-07,
      "loss": 0.0003,
      "reward": 1.132812537252903,
      "reward_std": 0.15746123250573874,
      "rewards/equation_reward_func": 0.132812503259629,
      "rewards/format_reward_func": 1.0,
      "step": 160
    },
    {
      "completion_length": 46.359375953674316,
      "epoch": 0.0864,
      "grad_norm": 0.2193554833555193,
      "kl": 0.2012939453125,
      "learning_rate": 3.7081709127108767e-07,
      "loss": 0.0002,
      "reward": 1.1067708805203438,
      "reward_std": 0.16360786044970155,
      "rewards/equation_reward_func": 0.10937500325962901,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 162
    },
    {
      "completion_length": 45.882813930511475,
      "epoch": 0.08746666666666666,
      "grad_norm": 0.18492158245762613,
      "kl": 0.2122802734375,
      "learning_rate": 3.6765055290490513e-07,
      "loss": 0.0002,
      "reward": 1.1041667088866234,
      "reward_std": 0.14379328675568104,
      "rewards/equation_reward_func": 0.10416667093522847,
      "rewards/format_reward_func": 1.0,
      "step": 164
    },
    {
      "completion_length": 47.39583468437195,
      "epoch": 0.08853333333333334,
      "grad_norm": 0.2878383465739314,
      "kl": 0.24993896484375,
      "learning_rate": 3.644595817401501e-07,
      "loss": 0.0003,
      "reward": 1.101562537252903,
      "reward_std": 0.1667499477043748,
      "rewards/equation_reward_func": 0.10416667000390589,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 166
    },
    {
      "completion_length": 46.401043176651,
      "epoch": 0.0896,
      "grad_norm": 0.22390708828311653,
      "kl": 0.22808837890625,
      "learning_rate": 3.6124484045416483e-07,
      "loss": 0.0002,
      "reward": 1.1171875298023224,
      "reward_std": 0.17219262197613716,
      "rewards/equation_reward_func": 0.11979166977107525,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 168
    },
    {
      "completion_length": 48.34895944595337,
      "epoch": 0.09066666666666667,
      "grad_norm": 0.20272961863237604,
      "kl": 0.23046875,
      "learning_rate": 3.580069966606949e-07,
      "loss": 0.0002,
      "reward": 1.1145833656191826,
      "reward_std": 0.11888703191652894,
      "rewards/equation_reward_func": 0.11458333721384406,
      "rewards/format_reward_func": 1.0,
      "step": 170
    },
    {
      "completion_length": 48.83333396911621,
      "epoch": 0.09173333333333333,
      "grad_norm": 0.19966493587754258,
      "kl": 0.19439697265625,
      "learning_rate": 3.547467227712444e-07,
      "loss": 0.0002,
      "reward": 1.1119792014360428,
      "reward_std": 0.15475187450647354,
      "rewards/equation_reward_func": 0.11197917000390589,
      "rewards/format_reward_func": 1.0,
      "step": 172
    },
    {
      "completion_length": 49.828126192092896,
      "epoch": 0.0928,
      "grad_norm": 0.2062659396560917,
      "kl": 0.189208984375,
      "learning_rate": 3.5146469585543386e-07,
      "loss": 0.0002,
      "reward": 1.1119792088866234,
      "reward_std": 0.14019352989271283,
      "rewards/equation_reward_func": 0.11197917093522847,
      "rewards/format_reward_func": 1.0,
      "step": 174
    },
    {
      "completion_length": 50.174480676651,
      "epoch": 0.09386666666666667,
      "grad_norm": 0.18081037831493624,
      "kl": 0.1900634765625,
      "learning_rate": 3.481615975003922e-07,
      "loss": 0.0002,
      "reward": 1.1380208656191826,
      "reward_std": 0.19034059438854456,
      "rewards/equation_reward_func": 0.1380208362825215,
      "rewards/format_reward_func": 1.0,
      "step": 176
    },
    {
      "completion_length": 51.210939168930054,
      "epoch": 0.09493333333333333,
      "grad_norm": 0.17579198600080073,
      "kl": 0.20147705078125,
      "learning_rate": 3.448381136692089e-07,
      "loss": 0.0002,
      "reward": 1.1145833730697632,
      "reward_std": 0.13870585104450583,
      "rewards/equation_reward_func": 0.11458333651535213,
      "rewards/format_reward_func": 1.0,
      "step": 178
    },
    {
      "completion_length": 51.36979293823242,
      "epoch": 0.096,
      "grad_norm": 0.18254205979883245,
      "kl": 0.221923828125,
      "learning_rate": 3.4149493455847897e-07,
      "loss": 0.0002,
      "reward": 1.0989583656191826,
      "reward_std": 0.10924306651577353,
      "rewards/equation_reward_func": 0.09895833674818277,
      "rewards/format_reward_func": 1.0,
      "step": 180
    },
    {
      "completion_length": 51.070313692092896,
      "epoch": 0.09706666666666666,
      "grad_norm": 0.2087922795870936,
      "kl": 0.18853759765625,
      "learning_rate": 3.3813275445496766e-07,
      "loss": 0.0002,
      "reward": 1.1067708656191826,
      "reward_std": 0.15386572387069464,
      "rewards/equation_reward_func": 0.10677083721384406,
      "rewards/format_reward_func": 1.0,
      "step": 182
    },
    {
      "completion_length": 52.156251430511475,
      "epoch": 0.09813333333333334,
      "grad_norm": 0.15534013320671317,
      "kl": 0.19378662109375,
      "learning_rate": 3.347522715914262e-07,
      "loss": 0.0002,
      "reward": 1.0937500298023224,
      "reward_std": 0.11073498986661434,
      "rewards/equation_reward_func": 0.09375000279396772,
      "rewards/format_reward_func": 1.0,
      "step": 184
    },
    {
      "completion_length": 51.320313930511475,
      "epoch": 0.0992,
      "grad_norm": 0.1262123444893106,
      "kl": 0.228515625,
      "learning_rate": 3.313541880015877e-07,
      "loss": 0.0002,
      "reward": 1.1822917014360428,
      "reward_std": 0.11888702679425478,
      "rewards/equation_reward_func": 0.1822916737291962,
      "rewards/format_reward_func": 1.0,
      "step": 186
    },
    {
      "completion_length": 52.42448019981384,
      "epoch": 0.10026666666666667,
      "grad_norm": 0.1430986027890787,
      "kl": 0.237060546875,
      "learning_rate": 3.279392093743747e-07,
      "loss": 0.0002,
      "reward": 1.1145833730697632,
      "reward_std": 0.14404708426445723,
      "rewards/equation_reward_func": 0.11458333814516664,
      "rewards/format_reward_func": 1.0,
      "step": 188
    },
    {
      "completion_length": 51.20052170753479,
      "epoch": 0.10133333333333333,
      "grad_norm": 0.20439810248884324,
      "kl": 0.21112060546875,
      "learning_rate": 3.245080449073459e-07,
      "loss": 0.0002,
      "reward": 1.132812537252903,
      "reward_std": 0.17279839795082808,
      "rewards/equation_reward_func": 0.13281250605359674,
      "rewards/format_reward_func": 1.0,
      "step": 190
    },
    {
      "completion_length": 52.40885543823242,
      "epoch": 0.1024,
      "grad_norm": 0.1965605802630599,
      "kl": 0.17791748046875,
      "learning_rate": 3.210614071594162e-07,
      "loss": 0.0002,
      "reward": 1.1458333656191826,
      "reward_std": 0.1817574673332274,
      "rewards/equation_reward_func": 0.14583333604969084,
      "rewards/format_reward_func": 1.0,
      "step": 192
    },
    {
      "completion_length": 53.968751668930054,
      "epoch": 0.10346666666666667,
      "grad_norm": 0.23763193671091185,
      "kl": 0.20513916015625,
      "learning_rate": 3.1760001190287695e-07,
      "loss": 0.0002,
      "reward": 1.1171875298023224,
      "reward_std": 0.15158909326419234,
      "rewards/equation_reward_func": 0.1171875037252903,
      "rewards/format_reward_func": 1.0,
      "step": 194
    },
    {
      "completion_length": 52.812501430511475,
      "epoch": 0.10453333333333334,
      "grad_norm": 0.17800547807143347,
      "kl": 0.1702880859375,
      "learning_rate": 3.141245779747502e-07,
      "loss": 0.0002,
      "reward": 1.161458358168602,
      "reward_std": 0.21051041642203927,
      "rewards/equation_reward_func": 0.16145833651535213,
      "rewards/format_reward_func": 1.0,
      "step": 196
    },
    {
      "completion_length": 51.90364742279053,
      "epoch": 0.1056,
      "grad_norm": 0.2064774084034785,
      "kl": 0.19232177734375,
      "learning_rate": 3.106358271275056e-07,
      "loss": 0.0002,
      "reward": 1.171875037252903,
      "reward_std": 0.20094394078478217,
      "rewards/equation_reward_func": 0.17187500349245965,
      "rewards/format_reward_func": 1.0,
      "step": 198
    },
    {
      "completion_length": 52.158855676651,
      "epoch": 0.10666666666666667,
      "grad_norm": 0.184608288760905,
      "kl": 0.187255859375,
      "learning_rate": 3.0713448387917227e-07,
      "loss": 0.0002,
      "reward": 1.1276042014360428,
      "reward_std": 0.1629039072431624,
      "rewards/equation_reward_func": 0.1276041711680591,
      "rewards/format_reward_func": 1.0,
      "step": 200
    },
    {
      "completion_length": 51.776043176651,
      "epoch": 0.10773333333333333,
      "grad_norm": 0.21803004520447924,
      "kl": 0.20123291015625,
      "learning_rate": 3.0362127536287636e-07,
      "loss": 0.0002,
      "reward": 1.1822917088866234,
      "reward_std": 0.16254702908918262,
      "rewards/equation_reward_func": 0.18229167419485748,
      "rewards/format_reward_func": 1.0,
      "step": 202
    },
    {
      "completion_length": 51.91666793823242,
      "epoch": 0.1088,
      "grad_norm": 0.18447509633488482,
      "kl": 0.20062255859375,
      "learning_rate": 3.0009693117583523e-07,
      "loss": 0.0002,
      "reward": 1.080729179084301,
      "reward_std": 0.14440069487318397,
      "rewards/equation_reward_func": 0.08333333441987634,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 204
    },
    {
      "completion_length": 52.398438930511475,
      "epoch": 0.10986666666666667,
      "grad_norm": 0.2229206997835387,
      "kl": 0.23052978515625,
      "learning_rate": 2.965621832278401e-07,
      "loss": 0.0002,
      "reward": 1.1380208805203438,
      "reward_std": 0.16973706847056746,
      "rewards/equation_reward_func": 0.13802083698101342,
      "rewards/format_reward_func": 1.0,
      "step": 206
    },
    {
      "completion_length": 52.710938692092896,
      "epoch": 0.11093333333333333,
      "grad_norm": 0.14734135218225045,
      "kl": 0.19921875,
      "learning_rate": 2.9301776558925875e-07,
      "loss": 0.0002,
      "reward": 1.0989583507180214,
      "reward_std": 0.12650489015504718,
      "rewards/equation_reward_func": 0.09895833511836827,
      "rewards/format_reward_func": 1.0,
      "step": 208
    },
    {
      "completion_length": 52.140626430511475,
      "epoch": 0.112,
      "grad_norm": 0.1687353946337615,
      "kl": 0.21435546875,
      "learning_rate": 2.894644143385885e-07,
      "loss": 0.0002,
      "reward": 1.1015625298023224,
      "reward_std": 0.14098411658778787,
      "rewards/equation_reward_func": 0.10156250349245965,
      "rewards/format_reward_func": 1.0,
      "step": 210
    },
    {
      "completion_length": 53.45833468437195,
      "epoch": 0.11306666666666666,
      "grad_norm": 0.15808793530127088,
      "kl": 0.19915771484375,
      "learning_rate": 2.859028674095937e-07,
      "loss": 0.0002,
      "reward": 1.0572916865348816,
      "reward_std": 0.08828429412096739,
      "rewards/equation_reward_func": 0.05729166744276881,
      "rewards/format_reward_func": 1.0,
      "step": 212
    },
    {
      "completion_length": 52.619793176651,
      "epoch": 0.11413333333333334,
      "grad_norm": 0.19253308864646929,
      "kl": 0.188232421875,
      "learning_rate": 2.823338644380566e-07,
      "loss": 0.0002,
      "reward": 1.1588541939854622,
      "reward_std": 0.1635096836835146,
      "rewards/equation_reward_func": 0.15885417140088975,
      "rewards/format_reward_func": 1.0,
      "step": 214
    },
    {
      "completion_length": 52.255210161209106,
      "epoch": 0.1152,
      "grad_norm": 0.24083809626870048,
      "kl": 0.20391845703125,
      "learning_rate": 2.7875814660817504e-07,
      "loss": 0.0002,
      "reward": 1.091145858168602,
      "reward_std": 0.1561423558741808,
      "rewards/equation_reward_func": 0.09375000093132257,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 216
    },
    {
      "completion_length": 52.28385519981384,
      "epoch": 0.11626666666666667,
      "grad_norm": 0.20187979910979093,
      "kl": 0.38525390625,
      "learning_rate": 2.751764564986396e-07,
      "loss": 0.0004,
      "reward": 1.101562537252903,
      "reward_std": 0.16439583525061607,
      "rewards/equation_reward_func": 0.10677083558402956,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 218
    },
    {
      "completion_length": 53.070313692092896,
      "epoch": 0.11733333333333333,
      "grad_norm": 0.17926301782298698,
      "kl": 0.21136474609375,
      "learning_rate": 2.715895379284194e-07,
      "loss": 0.0002,
      "reward": 1.0911458656191826,
      "reward_std": 0.1521200006827712,
      "rewards/equation_reward_func": 0.09114583488553762,
      "rewards/format_reward_func": 1.0,
      "step": 220
    },
    {
      "completion_length": 52.690105676651,
      "epoch": 0.1184,
      "grad_norm": 0.15533228660785858,
      "kl": 0.19342041015625,
      "learning_rate": 2.6799813580229174e-07,
      "loss": 0.0002,
      "reward": 1.1614583730697632,
      "reward_std": 0.1649999739602208,
      "rewards/equation_reward_func": 0.16406250721774995,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 222
    },
    {
      "completion_length": 52.828126192092896,
      "epoch": 0.11946666666666667,
      "grad_norm": 0.19751493038075355,
      "kl": 0.30401611328125,
      "learning_rate": 2.6440299595614606e-07,
      "loss": 0.0003,
      "reward": 1.1276042088866234,
      "reward_std": 0.15816845558583736,
      "rewards/equation_reward_func": 0.12760417209938169,
      "rewards/format_reward_func": 1.0,
      "step": 224
    },
    {
      "completion_length": 53.20573043823242,
      "epoch": 0.12053333333333334,
      "grad_norm": 0.21915930241414958,
      "kl": 0.22528076171875,
      "learning_rate": 2.6080486500209347e-07,
      "loss": 0.0002,
      "reward": 1.1041666865348816,
      "reward_std": 0.11888702772557735,
      "rewards/equation_reward_func": 0.1067708358168602,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 226
    },
    {
      "completion_length": 52.81770920753479,
      "epoch": 0.1216,
      "grad_norm": 0.14977686548839603,
      "kl": 0.1834716796875,
      "learning_rate": 2.572044901734166e-07,
      "loss": 0.0002,
      "reward": 1.1380208656191826,
      "reward_std": 0.1404499444179237,
      "rewards/equation_reward_func": 0.138020837912336,
      "rewards/format_reward_func": 1.0,
      "step": 228
    },
    {
      "completion_length": 51.85677218437195,
      "epoch": 0.12266666666666666,
      "grad_norm": 0.17821214679465763,
      "kl": 0.18194580078125,
      "learning_rate": 2.536026191693893e-07,
      "loss": 0.0002,
      "reward": 1.1510416939854622,
      "reward_std": 0.17009067768231034,
      "rewards/equation_reward_func": 0.15104167046956718,
      "rewards/format_reward_func": 1.0,
      "step": 230
    },
    {
      "completion_length": 51.55729269981384,
      "epoch": 0.12373333333333333,
      "grad_norm": 0.23137257100096653,
      "kl": 0.24261474609375,
      "learning_rate": 2.5e-07,
      "loss": 0.0002,
      "reward": 1.1666667014360428,
      "reward_std": 0.18595874728634953,
      "rewards/equation_reward_func": 0.1666666737291962,
      "rewards/format_reward_func": 1.0,
      "step": 232
    },
    {
      "completion_length": 52.67708468437195,
      "epoch": 0.1248,
      "grad_norm": 0.19279391734084148,
      "kl": 0.177001953125,
      "learning_rate": 2.4639738083061073e-07,
      "loss": 0.0002,
      "reward": 1.1223958730697632,
      "reward_std": 0.1664935341104865,
      "rewards/equation_reward_func": 0.12239583721384406,
      "rewards/format_reward_func": 1.0,
      "step": 234
    },
    {
      "completion_length": 52.98958468437195,
      "epoch": 0.12586666666666665,
      "grad_norm": 0.15143571679180334,
      "kl": 0.1856689453125,
      "learning_rate": 2.4279550982658345e-07,
      "loss": 0.0002,
      "reward": 1.138020858168602,
      "reward_std": 0.1319492026232183,
      "rewards/equation_reward_func": 0.1380208374466747,
      "rewards/format_reward_func": 1.0,
      "step": 236
    },
    {
      "completion_length": 50.99739670753479,
      "epoch": 0.12693333333333334,
      "grad_norm": 0.22787357517371357,
      "kl": 0.1998291015625,
      "learning_rate": 2.3919513499790646e-07,
      "loss": 0.0002,
      "reward": 1.1197917014360428,
      "reward_std": 0.14380546379834414,
      "rewards/equation_reward_func": 0.12239583698101342,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 238
    },
    {
      "completion_length": 52.351563930511475,
      "epoch": 0.128,
      "grad_norm": 0.14580546119741225,
      "kl": 0.22467041015625,
      "learning_rate": 2.3559700404385394e-07,
      "loss": 0.0002,
      "reward": 1.0598958656191826,
      "reward_std": 0.10504014790058136,
      "rewards/equation_reward_func": 0.059895834885537624,
      "rewards/format_reward_func": 1.0,
      "step": 240
    },
    {
      "completion_length": 51.82291793823242,
      "epoch": 0.12906666666666666,
      "grad_norm": 0.14158913115406083,
      "kl": 0.18646240234375,
      "learning_rate": 2.3200186419770823e-07,
      "loss": 0.0002,
      "reward": 1.1380208730697632,
      "reward_std": 0.1534329978749156,
      "rewards/equation_reward_func": 0.13802083721384406,
      "rewards/format_reward_func": 1.0,
      "step": 242
    },
    {
      "completion_length": 51.421876192092896,
      "epoch": 0.13013333333333332,
      "grad_norm": 0.22087661886865304,
      "kl": 0.2193603515625,
      "learning_rate": 2.284104620715807e-07,
      "loss": 0.0002,
      "reward": 1.0833333507180214,
      "reward_std": 0.125720186624676,
      "rewards/equation_reward_func": 0.08333333511836827,
      "rewards/format_reward_func": 1.0,
      "step": 244
    },
    {
      "completion_length": 52.179688453674316,
      "epoch": 0.1312,
      "grad_norm": 0.16870268080255363,
      "kl": 0.3653564453125,
      "learning_rate": 2.2482354350136043e-07,
      "loss": 0.0004,
      "reward": 1.0625000447034836,
      "reward_std": 0.12713723676279187,
      "rewards/equation_reward_func": 0.06510416930541396,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 246
    },
    {
      "completion_length": 52.390626192092896,
      "epoch": 0.13226666666666667,
      "grad_norm": 0.14530852488953735,
      "kl": 0.2147216796875,
      "learning_rate": 2.2124185339182496e-07,
      "loss": 0.0002,
      "reward": 1.1484375447034836,
      "reward_std": 0.16176396189257503,
      "rewards/equation_reward_func": 0.14843750442378223,
      "rewards/format_reward_func": 1.0,
      "step": 248
    },
    {
      "completion_length": 50.76041841506958,
      "epoch": 0.13333333333333333,
      "grad_norm": 0.22855594919537586,
      "kl": 0.20050048828125,
      "learning_rate": 2.1766613556194344e-07,
      "loss": 0.0002,
      "reward": 1.2005208730697632,
      "reward_std": 0.20069177821278572,
      "rewards/equation_reward_func": 0.2005208416376263,
      "rewards/format_reward_func": 1.0,
      "step": 250
    },
    {
      "completion_length": 52.320313692092896,
      "epoch": 0.1344,
      "grad_norm": 0.2415517772564965,
      "kl": 0.2119140625,
      "learning_rate": 2.1409713259040628e-07,
      "loss": 0.0002,
      "reward": 1.1432291939854622,
      "reward_std": 0.1831495789811015,
      "rewards/equation_reward_func": 0.1432291683740914,
      "rewards/format_reward_func": 1.0,
      "step": 252
    },
    {
      "completion_length": 53.713543176651,
      "epoch": 0.13546666666666668,
      "grad_norm": 0.15914212143213688,
      "kl": 0.21624755859375,
      "learning_rate": 2.105355856614115e-07,
      "loss": 0.0002,
      "reward": 1.101562537252903,
      "reward_std": 0.12432807218283415,
      "rewards/equation_reward_func": 0.10156250232830644,
      "rewards/format_reward_func": 1.0,
      "step": 254
    },
    {
      "completion_length": 52.90364694595337,
      "epoch": 0.13653333333333334,
      "grad_norm": 0.19034242360043133,
      "kl": 0.2158203125,
      "learning_rate": 2.069822344107413e-07,
      "loss": 0.0002,
      "reward": 1.1536458730697632,
      "reward_std": 0.17639390658587217,
      "rewards/equation_reward_func": 0.15625000419095159,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 256
    },
    {
      "completion_length": 53.86198091506958,
      "epoch": 0.1376,
      "grad_norm": 0.24513606248660633,
      "kl": 0.18865966796875,
      "learning_rate": 2.034378167721599e-07,
      "loss": 0.0002,
      "reward": 1.0859375298023224,
      "reward_std": 0.12966930214315653,
      "rewards/equation_reward_func": 0.08593750139698386,
      "rewards/format_reward_func": 1.0,
      "step": 258
    },
    {
      "completion_length": 52.54427218437195,
      "epoch": 0.13866666666666666,
      "grad_norm": 0.24848551325322174,
      "kl": 0.185546875,
      "learning_rate": 1.9990306882416485e-07,
      "loss": 0.0002,
      "reward": 1.1744791939854622,
      "reward_std": 0.1572823068127036,
      "rewards/equation_reward_func": 0.17447917070239782,
      "rewards/format_reward_func": 1.0,
      "step": 260
    },
    {
      "completion_length": 53.015625953674316,
      "epoch": 0.13973333333333332,
      "grad_norm": 0.22262610083917836,
      "kl": 0.20172119140625,
      "learning_rate": 1.9637872463712362e-07,
      "loss": 0.0002,
      "reward": 1.1432292088866234,
      "reward_std": 0.19831370282918215,
      "rewards/equation_reward_func": 0.14322917209938169,
      "rewards/format_reward_func": 1.0,
      "step": 262
    },
    {
      "completion_length": 54.210938453674316,
      "epoch": 0.1408,
      "grad_norm": 0.1794027751210474,
      "kl": 0.23321533203125,
      "learning_rate": 1.9286551612082773e-07,
      "loss": 0.0002,
      "reward": 1.0963541939854622,
      "reward_std": 0.13248010585084558,
      "rewards/equation_reward_func": 0.09635416977107525,
      "rewards/format_reward_func": 1.0,
      "step": 264
    },
    {
      "completion_length": 53.828126192092896,
      "epoch": 0.14186666666666667,
      "grad_norm": 0.18259812892353872,
      "kl": 0.20245361328125,
      "learning_rate": 1.8936417287249446e-07,
      "loss": 0.0002,
      "reward": 1.1250000447034836,
      "reward_std": 0.1474636597558856,
      "rewards/equation_reward_func": 0.12500000395812094,
      "rewards/format_reward_func": 1.0,
      "step": 266
    },
    {
      "completion_length": 52.562501430511475,
      "epoch": 0.14293333333333333,
      "grad_norm": 0.24921676446631377,
      "kl": 0.1982421875,
      "learning_rate": 1.8587542202524985e-07,
      "loss": 0.0002,
      "reward": 1.200520858168602,
      "reward_std": 0.18297653505578637,
      "rewards/equation_reward_func": 0.200520837912336,
      "rewards/format_reward_func": 1.0,
      "step": 268
    },
    {
      "completion_length": 53.94010519981384,
      "epoch": 0.144,
      "grad_norm": 0.19714793742591244,
      "kl": 0.18914794921875,
      "learning_rate": 1.82399988097123e-07,
      "loss": 0.0002,
      "reward": 1.125000037252903,
      "reward_std": 0.1472072503529489,
      "rewards/equation_reward_func": 0.12500000488944352,
      "rewards/format_reward_func": 1.0,
      "step": 270
    },
    {
      "completion_length": 54.406251192092896,
      "epoch": 0.14506666666666668,
      "grad_norm": 0.16973791015526338,
      "kl": 0.1890869140625,
      "learning_rate": 1.7893859284058378e-07,
      "loss": 0.0002,
      "reward": 1.1380208656191826,
      "reward_std": 0.11205223575234413,
      "rewards/equation_reward_func": 0.13802083698101342,
      "rewards/format_reward_func": 1.0,
      "step": 272
    },
    {
      "completion_length": 53.26302194595337,
      "epoch": 0.14613333333333334,
      "grad_norm": 0.20327426265511855,
      "kl": 0.2113037109375,
      "learning_rate": 1.7549195509265407e-07,
      "loss": 0.0002,
      "reward": 1.109375037252903,
      "reward_std": 0.14528520731255412,
      "rewards/equation_reward_func": 0.11197917116805911,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 274
    },
    {
      "completion_length": 53.588543176651,
      "epoch": 0.1472,
      "grad_norm": 0.19180730581931288,
      "kl": 0.2054443359375,
      "learning_rate": 1.7206079062562536e-07,
      "loss": 0.0002,
      "reward": 1.1822917014360428,
      "reward_std": 0.1643941984511912,
      "rewards/equation_reward_func": 0.18229167046956718,
      "rewards/format_reward_func": 1.0,
      "step": 276
    },
    {
      "completion_length": 54.61979293823242,
      "epoch": 0.14826666666666666,
      "grad_norm": 0.2653834699516297,
      "kl": 0.1865234375,
      "learning_rate": 1.6864581199841226e-07,
      "loss": 0.0002,
      "reward": 1.1979167014360428,
      "reward_std": 0.18140549072995782,
      "rewards/equation_reward_func": 0.19791667140088975,
      "rewards/format_reward_func": 1.0,
      "step": 278
    },
    {
      "completion_length": 53.97395968437195,
      "epoch": 0.14933333333333335,
      "grad_norm": 0.26142387581207355,
      "kl": 0.19573974609375,
      "learning_rate": 1.6524772840857388e-07,
      "loss": 0.0002,
      "reward": 1.1484375298023224,
      "reward_std": 0.20208552666008472,
      "rewards/equation_reward_func": 0.148437503259629,
      "rewards/format_reward_func": 1.0,
      "step": 280
    },
    {
      "completion_length": 54.179688930511475,
      "epoch": 0.1504,
      "grad_norm": 0.1543914017100232,
      "kl": 0.20465087890625,
      "learning_rate": 1.6186724554503237e-07,
      "loss": 0.0002,
      "reward": 1.1588541939854622,
      "reward_std": 0.1389612853527069,
      "rewards/equation_reward_func": 0.15885417023673654,
      "rewards/format_reward_func": 1.0,
      "step": 282
    },
    {
      "completion_length": 54.72135543823242,
      "epoch": 0.15146666666666667,
      "grad_norm": 0.20935901318611044,
      "kl": 0.2032470703125,
      "learning_rate": 1.5850506544152103e-07,
      "loss": 0.0002,
      "reward": 1.1380208730697632,
      "reward_std": 0.14668059069663286,
      "rewards/equation_reward_func": 0.13802083861082792,
      "rewards/format_reward_func": 1.0,
      "step": 284
    },
    {
      "completion_length": 56.87239670753479,
      "epoch": 0.15253333333333333,
      "grad_norm": 0.198714634239171,
      "kl": 0.20599365234375,
      "learning_rate": 1.5516188633079107e-07,
      "loss": 0.0002,
      "reward": 1.1276041865348816,
      "reward_std": 0.10511502251029015,
      "rewards/equation_reward_func": 0.1276041716337204,
      "rewards/format_reward_func": 1.0,
      "step": 286
    },
    {
      "completion_length": 55.796875953674316,
      "epoch": 0.1536,
      "grad_norm": 0.19363272496423903,
      "kl": 0.2156982421875,
      "learning_rate": 1.5183840249960784e-07,
      "loss": 0.0002,
      "reward": 1.171875037252903,
      "reward_std": 0.1659636083059013,
      "rewards/equation_reward_func": 0.17447917233221233,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 288
    },
    {
      "completion_length": 56.611980676651,
      "epoch": 0.15466666666666667,
      "grad_norm": 0.2579729420426016,
      "kl": 0.19622802734375,
      "learning_rate": 1.4853530414456612e-07,
      "loss": 0.0002,
      "reward": 1.192708358168602,
      "reward_std": 0.1792004620656371,
      "rewards/equation_reward_func": 0.19270833884365857,
      "rewards/format_reward_func": 1.0,
      "step": 290
    },
    {
      "completion_length": 55.84895992279053,
      "epoch": 0.15573333333333333,
      "grad_norm": 0.1982128281172869,
      "kl": 0.20867919921875,
      "learning_rate": 1.4525327722875568e-07,
      "loss": 0.0002,
      "reward": 1.1015625298023224,
      "reward_std": 0.1330083985812962,
      "rewards/equation_reward_func": 0.1015625016298145,
      "rewards/format_reward_func": 1.0,
      "step": 292
    },
    {
      "completion_length": 55.71614718437195,
      "epoch": 0.1568,
      "grad_norm": 0.23193236770472453,
      "kl": 0.21417236328125,
      "learning_rate": 1.4199300333930515e-07,
      "loss": 0.0002,
      "reward": 1.1588542088866234,
      "reward_std": 0.1723656700924039,
      "rewards/equation_reward_func": 0.1588541732635349,
      "rewards/format_reward_func": 1.0,
      "step": 294
    },
    {
      "completion_length": 56.80208468437195,
      "epoch": 0.15786666666666666,
      "grad_norm": 0.18074332848087488,
      "kl": 0.2083740234375,
      "learning_rate": 1.3875515954583523e-07,
      "loss": 0.0002,
      "reward": 1.1432292014360428,
      "reward_std": 0.1035482264123857,
      "rewards/equation_reward_func": 0.14322917209938169,
      "rewards/format_reward_func": 1.0,
      "step": 296
    },
    {
      "completion_length": 54.99479293823242,
      "epoch": 0.15893333333333334,
      "grad_norm": 0.23848033310887906,
      "kl": 0.22198486328125,
      "learning_rate": 1.3554041825985e-07,
      "loss": 0.0002,
      "reward": 1.1432292014360428,
      "reward_std": 0.13624865701422095,
      "rewards/equation_reward_func": 0.14322917233221233,
      "rewards/format_reward_func": 1.0,
      "step": 298
    },
    {
      "completion_length": 57.58333420753479,
      "epoch": 0.16,
      "grad_norm": 0.20740052313812263,
      "kl": 0.22088623046875,
      "learning_rate": 1.323494470950949e-07,
      "loss": 0.0002,
      "reward": 1.0937500149011612,
      "reward_std": 0.11923900293186307,
      "rewards/equation_reward_func": 0.0937500016298145,
      "rewards/format_reward_func": 1.0,
      "step": 300
    },
    {
      "completion_length": 56.468751430511475,
      "epoch": 0.16106666666666666,
      "grad_norm": 0.2225036708029395,
      "kl": 0.20849609375,
      "learning_rate": 1.2918290872891236e-07,
      "loss": 0.0002,
      "reward": 1.0885416939854622,
      "reward_std": 0.11415157234296203,
      "rewards/equation_reward_func": 0.09375000279396772,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 302
    },
    {
      "completion_length": 54.992188692092896,
      "epoch": 0.16213333333333332,
      "grad_norm": 0.20501829560165247,
      "kl": 0.1993408203125,
      "learning_rate": 1.260414607646213e-07,
      "loss": 0.0002,
      "reward": 1.1067708730697632,
      "reward_std": 0.12625272199511528,
      "rewards/equation_reward_func": 0.10677083698101342,
      "rewards/format_reward_func": 1.0,
      "step": 304
    },
    {
      "completion_length": 55.30729341506958,
      "epoch": 0.1632,
      "grad_norm": 0.1625796706131234,
      "kl": 0.21453857421875,
      "learning_rate": 1.2292575559495143e-07,
      "loss": 0.0002,
      "reward": 1.1197916939854622,
      "reward_std": 0.11205386929214001,
      "rewards/equation_reward_func": 0.12239583535119891,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 306
    },
    {
      "completion_length": 55.40364742279053,
      "epoch": 0.16426666666666667,
      "grad_norm": 0.17660631945613436,
      "kl": 0.19049072265625,
      "learning_rate": 1.1983644026655835e-07,
      "loss": 0.0002,
      "reward": 1.0781250223517418,
      "reward_std": 0.0990682034753263,
      "rewards/equation_reward_func": 0.07812500209547579,
      "rewards/format_reward_func": 1.0,
      "step": 308
    },
    {
      "completion_length": 56.37239670753479,
      "epoch": 0.16533333333333333,
      "grad_norm": 0.21887886658794795,
      "kl": 0.22833251953125,
      "learning_rate": 1.1677415634565066e-07,
      "loss": 0.0002,
      "reward": 1.1354166865348816,
      "reward_std": 0.1642960263416171,
      "rewards/equation_reward_func": 0.13541666977107525,
      "rewards/format_reward_func": 1.0,
      "step": 310
    },
    {
      "completion_length": 55.22395920753479,
      "epoch": 0.1664,
      "grad_norm": 0.30991099610944484,
      "kl": 0.19610595703125,
      "learning_rate": 1.1373953978475353e-07,
      "loss": 0.0002,
      "reward": 1.1406250298023224,
      "reward_std": 0.14536008005961776,
      "rewards/equation_reward_func": 0.14062500605359674,
      "rewards/format_reward_func": 1.0,
      "step": 312
    },
    {
      "completion_length": 55.109376192092896,
      "epoch": 0.16746666666666668,
      "grad_norm": 0.11748492375994124,
      "kl": 0.2703857421875,
      "learning_rate": 1.1073322079063913e-07,
      "loss": 0.0003,
      "reward": 1.1276041939854622,
      "reward_std": 0.11870646476745605,
      "rewards/equation_reward_func": 0.12760417046956718,
      "rewards/format_reward_func": 1.0,
      "step": 314
    },
    {
      "completion_length": 54.89323019981384,
      "epoch": 0.16853333333333334,
      "grad_norm": 0.2835236931065035,
      "kl": 0.23651123046875,
      "learning_rate": 1.0775582369344946e-07,
      "loss": 0.0002,
      "reward": 1.1562500223517418,
      "reward_std": 0.18165602022781968,
      "rewards/equation_reward_func": 0.1562500020954758,
      "rewards/format_reward_func": 1.0,
      "step": 316
    },
    {
      "completion_length": 54.95312571525574,
      "epoch": 0.1696,
      "grad_norm": 0.3033848386113242,
      "kl": 0.18084716796875,
      "learning_rate": 1.0480796681704077e-07,
      "loss": 0.0002,
      "reward": 1.171875037252903,
      "reward_std": 0.1689507276751101,
      "rewards/equation_reward_func": 0.17187500605359674,
      "rewards/format_reward_func": 1.0,
      "step": 318
    },
    {
      "completion_length": 55.18229293823242,
      "epoch": 0.17066666666666666,
      "grad_norm": 0.18904924808830093,
      "kl": 0.18798828125,
      "learning_rate": 1.018902623505741e-07,
      "loss": 0.0002,
      "reward": 1.1744792088866234,
      "reward_std": 0.14378903713077307,
      "rewards/equation_reward_func": 0.1744791716337204,
      "rewards/format_reward_func": 1.0,
      "step": 320
    },
    {
      "completion_length": 55.791668176651,
      "epoch": 0.17173333333333332,
      "grad_norm": 0.15824811484075788,
      "kl": 0.192626953125,
      "learning_rate": 9.900331622138063e-08,
      "loss": 0.0002,
      "reward": 1.1223958730697632,
      "reward_std": 0.12073256447911263,
      "rewards/equation_reward_func": 0.12239583767950535,
      "rewards/format_reward_func": 1.0,
      "step": 322
    },
    {
      "completion_length": 57.47135519981384,
      "epoch": 0.1728,
      "grad_norm": 0.24122866889709543,
      "kl": 0.273193359375,
      "learning_rate": 9.614772796912681e-08,
      "loss": 0.0003,
      "reward": 1.1536458656191826,
      "reward_std": 0.13116122921928763,
      "rewards/equation_reward_func": 0.15625000465661287,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 324
    },
    {
      "completion_length": 55.867188930511475,
      "epoch": 0.17386666666666667,
      "grad_norm": 0.2140944203573884,
      "kl": 0.19073486328125,
      "learning_rate": 9.332409062130686e-08,
      "loss": 0.0002,
      "reward": 1.1510416939854622,
      "reward_std": 0.17692383378744125,
      "rewards/equation_reward_func": 0.1510416711680591,
      "rewards/format_reward_func": 1.0,
      "step": 326
    },
    {
      "completion_length": 56.45573019981384,
      "epoch": 0.17493333333333333,
      "grad_norm": 0.25020509022355314,
      "kl": 0.20635986328125,
      "learning_rate": 9.053299057008699e-08,
      "loss": 0.0002,
      "reward": 1.1015625298023224,
      "reward_std": 0.1254680184647441,
      "rewards/equation_reward_func": 0.10416666860692203,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 328
    },
    {
      "completion_length": 55.67708444595337,
      "epoch": 0.176,
      "grad_norm": 0.23914885533951835,
      "kl": 0.19207763671875,
      "learning_rate": 8.777500745052743e-08,
      "loss": 0.0002,
      "reward": 1.1718750298023224,
      "reward_std": 0.15009553357958794,
      "rewards/equation_reward_func": 0.17187500465661287,
      "rewards/format_reward_func": 1.0,
      "step": 330
    },
    {
      "completion_length": 57.98958468437195,
      "epoch": 0.17706666666666668,
      "grad_norm": 0.1801495946567539,
      "kl": 0.20220947265625,
      "learning_rate": 8.505071402020892e-08,
      "loss": 0.0002,
      "reward": 1.1770833656191826,
      "reward_std": 0.11941793095320463,
      "rewards/equation_reward_func": 0.1796875053551048,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 332
    },
    {
      "completion_length": 57.226563692092896,
      "epoch": 0.17813333333333334,
      "grad_norm": 0.19524459206106992,
      "kl": 0.19586181640625,
      "learning_rate": 8.236067604028562e-08,
      "loss": 0.0002,
      "reward": 1.125000037252903,
      "reward_std": 0.10995029006153345,
      "rewards/equation_reward_func": 0.12760417093522847,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 334
    },
    {
      "completion_length": 55.335938930511475,
      "epoch": 0.1792,
      "grad_norm": 0.2559778596356376,
      "kl": 0.1851806640625,
      "learning_rate": 7.970545215799327e-08,
      "loss": 0.0002,
      "reward": 1.1614583730697632,
      "reward_std": 0.15176639100536704,
      "rewards/equation_reward_func": 0.1614583390764892,
      "rewards/format_reward_func": 1.0,
      "step": 336
    },
    {
      "completion_length": 57.507813692092896,
      "epoch": 0.18026666666666666,
      "grad_norm": 0.17799848022874795,
      "kl": 0.20086669921875,
      "learning_rate": 7.708559379063204e-08,
      "loss": 0.0002,
      "reward": 1.1093750298023224,
      "reward_std": 0.11755679547786713,
      "rewards/equation_reward_func": 0.11197917046956718,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 338
    },
    {
      "completion_length": 56.45833444595337,
      "epoch": 0.18133333333333335,
      "grad_norm": 0.13176298149720098,
      "kl": 0.1859130859375,
      "learning_rate": 7.45016450110534e-08,
      "loss": 0.0002,
      "reward": 1.161458358168602,
      "reward_std": 0.14746366580948234,
      "rewards/equation_reward_func": 0.16406250395812094,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 340
    },
    {
      "completion_length": 56.984376430511475,
      "epoch": 0.1824,
      "grad_norm": 0.08420346762402851,
      "kl": 0.18780517578125,
      "learning_rate": 7.195414243467029e-08,
      "loss": 0.0002,
      "reward": 1.0572916865348816,
      "reward_std": 0.0555838611908257,
      "rewards/equation_reward_func": 0.05729166860692203,
      "rewards/format_reward_func": 1.0,
      "step": 342
    },
    {
      "completion_length": 56.281251192092896,
      "epoch": 0.18346666666666667,
      "grad_norm": 0.1339259257169896,
      "kl": 0.19757080078125,
      "learning_rate": 6.944361510801763e-08,
      "loss": 0.0002,
      "reward": 1.1145833507180214,
      "reward_std": 0.11265964666381478,
      "rewards/equation_reward_func": 0.11458333628252149,
      "rewards/format_reward_func": 1.0,
      "step": 344
    },
    {
      "completion_length": 56.515626192092896,
      "epoch": 0.18453333333333333,
      "grad_norm": 0.24369984772040754,
      "kl": 0.20953369140625,
      "learning_rate": 6.697058439888283e-08,
      "loss": 0.0002,
      "reward": 1.164062537252903,
      "reward_std": 0.16614156123250723,
      "rewards/equation_reward_func": 0.1640625037252903,
      "rewards/format_reward_func": 1.0,
      "step": 346
    },
    {
      "completion_length": 56.27604269981384,
      "epoch": 0.1856,
      "grad_norm": 0.20891042464235796,
      "kl": 0.1956787109375,
      "learning_rate": 6.453556388803288e-08,
      "loss": 0.0002,
      "reward": 1.1510416939854622,
      "reward_std": 0.14159152610227466,
      "rewards/equation_reward_func": 0.15104167186655104,
      "rewards/format_reward_func": 1.0,
      "step": 348
    },
    {
      "completion_length": 56.265626430511475,
      "epoch": 0.18666666666666668,
      "grad_norm": 0.2013483142099634,
      "kl": 0.19561767578125,
      "learning_rate": 6.213905926255697e-08,
      "loss": 0.0002,
      "reward": 1.114583358168602,
      "reward_std": 0.097322478890419,
      "rewards/equation_reward_func": 0.11458333651535213,
      "rewards/format_reward_func": 1.0,
      "step": 350
    },
    {
      "completion_length": 54.901043176651,
      "epoch": 0.18773333333333334,
      "grad_norm": 0.18736870135125286,
      "kl": 0.34832763671875,
      "learning_rate": 5.978156821084987e-08,
      "loss": 0.0003,
      "reward": 1.1718750298023224,
      "reward_std": 0.14368922589346766,
      "rewards/equation_reward_func": 0.17187500488944352,
      "rewards/format_reward_func": 1.0,
      "step": 352
    },
    {
      "completion_length": 56.213542461395264,
      "epoch": 0.1888,
      "grad_norm": 0.23588576373730524,
      "kl": 0.19842529296875,
      "learning_rate": 5.7463580319254853e-08,
      "loss": 0.0002,
      "reward": 1.177083358168602,
      "reward_std": 0.18525153072550893,
      "rewards/equation_reward_func": 0.1770833358168602,
      "rewards/format_reward_func": 1.0,
      "step": 354
    },
    {
      "completion_length": 56.401043176651,
      "epoch": 0.18986666666666666,
      "grad_norm": 0.1965308569670095,
      "kl": 0.193115234375,
      "learning_rate": 5.518557697039081e-08,
      "loss": 0.0002,
      "reward": 1.1276041939854622,
      "reward_std": 0.10607865452766418,
      "rewards/equation_reward_func": 0.12760417186655104,
      "rewards/format_reward_func": 1.0,
      "step": 356
    },
    {
      "completion_length": 56.31770944595337,
      "epoch": 0.19093333333333334,
      "grad_norm": 0.24544345942440962,
      "kl": 0.21868896484375,
      "learning_rate": 5.294803124318145e-08,
      "loss": 0.0002,
      "reward": 1.1250000298023224,
      "reward_std": 0.15678852004930377,
      "rewards/equation_reward_func": 0.1276041695382446,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 358
    },
    {
      "completion_length": 55.739585161209106,
      "epoch": 0.192,
      "grad_norm": 0.25844244454008874,
      "kl": 0.234619140625,
      "learning_rate": 5.07514078146106e-08,
      "loss": 0.0002,
      "reward": 1.1328125298023224,
      "reward_std": 0.13668138859793544,
      "rewards/equation_reward_func": 0.13281250465661287,
      "rewards/format_reward_func": 1.0,
      "step": 360
    },
    {
      "completion_length": 56.58854269981384,
      "epoch": 0.19306666666666666,
      "grad_norm": 0.175480373166462,
      "kl": 0.2298583984375,
      "learning_rate": 4.859616286322094e-08,
      "loss": 0.0002,
      "reward": 1.1302083432674408,
      "reward_std": 0.11230767332017422,
      "rewards/equation_reward_func": 0.13020833488553762,
      "rewards/format_reward_func": 1.0,
      "step": 362
    },
    {
      "completion_length": 56.60677218437195,
      "epoch": 0.19413333333333332,
      "grad_norm": 0.3271596777219154,
      "kl": 0.192626953125,
      "learning_rate": 4.648274397437829e-08,
      "loss": 0.0002,
      "reward": 1.1067708656191826,
      "reward_std": 0.11528988601639867,
      "rewards/equation_reward_func": 0.10677083558402956,
      "rewards/format_reward_func": 1.0,
      "step": 364
    },
    {
      "completion_length": 56.658854961395264,
      "epoch": 0.1952,
      "grad_norm": 0.2951506047350438,
      "kl": 0.288818359375,
      "learning_rate": 4.4411590047320617e-08,
      "loss": 0.0003,
      "reward": 1.1328125298023224,
      "reward_std": 0.1543966280296445,
      "rewards/equation_reward_func": 0.1354166716337204,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 366
    },
    {
      "completion_length": 56.86458468437195,
      "epoch": 0.19626666666666667,
      "grad_norm": 0.1704950407789878,
      "kl": 0.2303466796875,
      "learning_rate": 4.2383131204010494e-08,
      "loss": 0.0002,
      "reward": 1.1041666939854622,
      "reward_std": 0.08031118754297495,
      "rewards/equation_reward_func": 0.10416666907258332,
      "rewards/format_reward_func": 1.0,
      "step": 368
    },
    {
      "completion_length": 56.02604293823242,
      "epoch": 0.19733333333333333,
      "grad_norm": 0.17572637332497476,
      "kl": 0.20947265625,
      "learning_rate": 4.039778869981064e-08,
      "loss": 0.0002,
      "reward": 1.1354167014360428,
      "reward_std": 0.11696237418800592,
      "rewards/equation_reward_func": 0.13541667046956718,
      "rewards/format_reward_func": 1.0,
      "step": 370
    },
    {
      "completion_length": 58.16666841506958,
      "epoch": 0.1984,
      "grad_norm": 0.16929728780013165,
      "kl": 0.23858642578125,
      "learning_rate": 3.845597483600049e-08,
      "loss": 0.0002,
      "reward": 1.140625037252903,
      "reward_std": 0.12678526248782873,
      "rewards/equation_reward_func": 0.1406250058207661,
      "rewards/format_reward_func": 1.0,
      "step": 372
    },
    {
      "completion_length": 56.390625953674316,
      "epoch": 0.19946666666666665,
      "grad_norm": 0.23023184690495188,
      "kl": 0.19927978515625,
      "learning_rate": 3.655809287415284e-08,
      "loss": 0.0002,
      "reward": 1.1822916939854622,
      "reward_std": 0.17535115545615554,
      "rewards/equation_reward_func": 0.18229167023673654,
      "rewards/format_reward_func": 1.0,
      "step": 374
    },
    {
      "completion_length": 57.822918176651,
      "epoch": 0.20053333333333334,
      "grad_norm": 0.17921347505393254,
      "kl": 0.20977783203125,
      "learning_rate": 3.4704536952387285e-08,
      "loss": 0.0002,
      "reward": 1.1640625298023224,
      "reward_std": 0.13212224887683988,
      "rewards/equation_reward_func": 0.16406250395812094,
      "rewards/format_reward_func": 1.0,
      "step": 376
    },
    {
      "completion_length": 57.23437571525574,
      "epoch": 0.2016,
      "grad_norm": 0.2242393332943885,
      "kl": 0.20806884765625,
      "learning_rate": 3.2895692003518575e-08,
      "loss": 0.0002,
      "reward": 1.1953125298023224,
      "reward_std": 0.1640438586473465,
      "rewards/equation_reward_func": 0.19531250442378223,
      "rewards/format_reward_func": 1.0,
      "step": 378
    },
    {
      "completion_length": 57.539063453674316,
      "epoch": 0.20266666666666666,
      "grad_norm": 0.2234153005175412,
      "kl": 0.2264404296875,
      "learning_rate": 3.113193367511635e-08,
      "loss": 0.0002,
      "reward": 1.1276042014360428,
      "reward_std": 0.13984416658058763,
      "rewards/equation_reward_func": 0.13020833884365857,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 380
    },
    {
      "completion_length": 56.906251430511475,
      "epoch": 0.20373333333333332,
      "grad_norm": 0.19389399900613372,
      "kl": 0.201416015625,
      "learning_rate": 2.9413628251493934e-08,
      "loss": 0.0002,
      "reward": 1.0833333656191826,
      "reward_std": 0.08461391646414995,
      "rewards/equation_reward_func": 0.0833333358168602,
      "rewards/format_reward_func": 1.0,
      "step": 382
    },
    {
      "completion_length": 55.69010519981384,
      "epoch": 0.2048,
      "grad_norm": 0.16073109173125164,
      "kl": 0.21087646484375,
      "learning_rate": 2.774113257764066e-08,
      "loss": 0.0002,
      "reward": 1.122395858168602,
      "reward_std": 0.10468817176297307,
      "rewards/equation_reward_func": 0.12239583651535213,
      "rewards/format_reward_func": 1.0,
      "step": 384
    },
    {
      "completion_length": 58.12239694595337,
      "epoch": 0.20586666666666667,
      "grad_norm": 0.16315603157541186,
      "kl": 0.21099853515625,
      "learning_rate": 2.611479398511518e-08,
      "loss": 0.0002,
      "reward": 1.0989583507180214,
      "reward_std": 0.12098472937941551,
      "rewards/equation_reward_func": 0.10156250139698386,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 386
    },
    {
      "completion_length": 56.40364694595337,
      "epoch": 0.20693333333333333,
      "grad_norm": 0.18228098703805926,
      "kl": 0.22320556640625,
      "learning_rate": 2.4534950219914057e-08,
      "loss": 0.0002,
      "reward": 1.1458333656191826,
      "reward_std": 0.15667489264160395,
      "rewards/equation_reward_func": 0.14843750488944352,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 388
    },
    {
      "completion_length": 56.335939168930054,
      "epoch": 0.208,
      "grad_norm": 0.08793205228078631,
      "kl": 0.22027587890625,
      "learning_rate": 2.300192937233128e-08,
      "loss": 0.0002,
      "reward": 1.145833358168602,
      "reward_std": 0.09170087426900864,
      "rewards/equation_reward_func": 0.14583333814516664,
      "rewards/format_reward_func": 1.0,
      "step": 390
    },
    {
      "completion_length": 57.601563930511475,
      "epoch": 0.20906666666666668,
      "grad_norm": 0.13829741360617928,
      "kl": 0.207763671875,
      "learning_rate": 2.1516049808822935e-08,
      "loss": 0.0002,
      "reward": 1.0963541939854622,
      "reward_std": 0.08522132784128189,
      "rewards/equation_reward_func": 0.09635416930541396,
      "rewards/format_reward_func": 1.0,
      "step": 392
    },
    {
      "completion_length": 56.835938453674316,
      "epoch": 0.21013333333333334,
      "grad_norm": 0.20152462319314116,
      "kl": 0.20361328125,
      "learning_rate": 2.007762010589098e-08,
      "loss": 0.0002,
      "reward": 1.1223958656191826,
      "reward_std": 0.15199717972427607,
      "rewards/equation_reward_func": 0.12500000465661287,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 394
    },
    {
      "completion_length": 57.429688453674316,
      "epoch": 0.2112,
      "grad_norm": 0.15920228598590455,
      "kl": 0.193359375,
      "learning_rate": 1.8686938986000627e-08,
      "loss": 0.0002,
      "reward": 1.1354166939854622,
      "reward_std": 0.1378197013400495,
      "rewards/equation_reward_func": 0.13541667093522847,
      "rewards/format_reward_func": 1.0,
      "step": 396
    },
    {
      "completion_length": 56.05989718437195,
      "epoch": 0.21226666666666666,
      "grad_norm": 0.2321801917334021,
      "kl": 0.1983642578125,
      "learning_rate": 1.734429525554365e-08,
      "loss": 0.0002,
      "reward": 1.1093750149011612,
      "reward_std": 0.12607543170452118,
      "rewards/equation_reward_func": 0.10937500186264515,
      "rewards/format_reward_func": 1.0,
      "step": 398
    },
    {
      "completion_length": 57.57552242279053,
      "epoch": 0.21333333333333335,
      "grad_norm": 0.21983985221412547,
      "kl": 0.19842529296875,
      "learning_rate": 1.604996774486145e-08,
      "loss": 0.0002,
      "reward": 1.2161458656191826,
      "reward_std": 0.1926204962655902,
      "rewards/equation_reward_func": 0.21614583977498114,
      "rewards/format_reward_func": 1.0,
      "step": 400
    },
    {
      "completion_length": 55.960938692092896,
      "epoch": 0.2144,
      "grad_norm": 0.26196521736280276,
      "kl": 0.20556640625,
      "learning_rate": 1.4804225250339281e-08,
      "loss": 0.0002,
      "reward": 1.2500000447034836,
      "reward_std": 0.1827184883877635,
      "rewards/equation_reward_func": 0.25000001140870154,
      "rewards/format_reward_func": 1.0,
      "step": 402
    },
    {
      "completion_length": 56.73177218437195,
      "epoch": 0.21546666666666667,
      "grad_norm": 0.15469340979183918,
      "kl": 0.1986083984375,
      "learning_rate": 1.360732647858498e-08,
      "loss": 0.0002,
      "reward": 1.0677083656191826,
      "reward_std": 0.08618659200146794,
      "rewards/equation_reward_func": 0.0677083358168602,
      "rewards/format_reward_func": 1.0,
      "step": 404
    },
    {
      "completion_length": 56.89843940734863,
      "epoch": 0.21653333333333333,
      "grad_norm": 0.23801117725189178,
      "kl": 0.20068359375,
      "learning_rate": 1.2459519992702311e-08,
      "loss": 0.0002,
      "reward": 1.218750037252903,
      "reward_std": 0.21339609054848552,
      "rewards/equation_reward_func": 0.21875000675208867,
      "rewards/format_reward_func": 1.0,
      "step": 406
    },
    {
      "completion_length": 57.031251192092896,
      "epoch": 0.2176,
      "grad_norm": 0.2651467420541166,
      "kl": 0.20458984375,
      "learning_rate": 1.1361044160671629e-08,
      "loss": 0.0002,
      "reward": 1.1640625298023224,
      "reward_std": 0.12432807171717286,
      "rewards/equation_reward_func": 0.16406250419095159,
      "rewards/format_reward_func": 1.0,
      "step": 408
    },
    {
      "completion_length": 57.26041793823242,
      "epoch": 0.21866666666666668,
      "grad_norm": 0.23371932864049302,
      "kl": 0.29351806640625,
      "learning_rate": 1.0312127105846947e-08,
      "loss": 0.0003,
      "reward": 1.1119792014360428,
      "reward_std": 0.14308181405067444,
      "rewards/equation_reward_func": 0.11197916930541396,
      "rewards/format_reward_func": 1.0,
      "step": 410
    },
    {
      "completion_length": 57.369792461395264,
      "epoch": 0.21973333333333334,
      "grad_norm": 0.1736174390883632,
      "kl": 0.19873046875,
      "learning_rate": 9.312986659581301e-09,
      "loss": 0.0002,
      "reward": 1.1562500298023224,
      "reward_std": 0.12151890201494098,
      "rewards/equation_reward_func": 0.1562500090803951,
      "rewards/format_reward_func": 1.0,
      "step": 412
    },
    {
      "completion_length": 56.98958444595337,
      "epoch": 0.2208,
      "grad_norm": 0.15941088538197615,
      "kl": 0.1998291015625,
      "learning_rate": 8.363830315988945e-09,
      "loss": 0.0002,
      "reward": 1.1354166939854622,
      "reward_std": 0.08215835923328996,
      "rewards/equation_reward_func": 0.13541667349636555,
      "rewards/format_reward_func": 1.0,
      "step": 414
    },
    {
      "completion_length": 56.79427218437195,
      "epoch": 0.22186666666666666,
      "grad_norm": 0.2139924970111357,
      "kl": 0.199951171875,
      "learning_rate": 7.46485518885462e-09,
      "loss": 0.0002,
      "reward": 1.1484375223517418,
      "reward_std": 0.1481725163757801,
      "rewards/equation_reward_func": 0.14843750349245965,
      "rewards/format_reward_func": 1.0,
      "step": 416
    },
    {
      "completion_length": 56.44270920753479,
      "epoch": 0.22293333333333334,
      "grad_norm": 0.23307793608390492,
      "kl": 0.43865966796875,
      "learning_rate": 6.616247970698319e-09,
      "loss": 0.0004,
      "reward": 1.2239583507180214,
      "reward_std": 0.1484246808104217,
      "rewards/equation_reward_func": 0.22395833721384406,
      "rewards/format_reward_func": 1.0,
      "step": 418
    },
    {
      "completion_length": 58.179688692092896,
      "epoch": 0.224,
      "grad_norm": 0.21251783571984495,
      "kl": 0.19537353515625,
      "learning_rate": 5.8181848940044855e-09,
      "loss": 0.0002,
      "reward": 1.1458333730697632,
      "reward_std": 0.13906109519302845,
      "rewards/equation_reward_func": 0.14583333767950535,
      "rewards/format_reward_func": 1.0,
      "step": 420
    },
    {
      "completion_length": 56.026043176651,
      "epoch": 0.22506666666666666,
      "grad_norm": 0.1951023187826417,
      "kl": 0.3018798828125,
      "learning_rate": 5.070831694623135e-09,
      "loss": 0.0003,
      "reward": 1.1692708730697632,
      "reward_std": 0.16852224431931973,
      "rewards/equation_reward_func": 0.16927084093913436,
      "rewards/format_reward_func": 1.0,
      "step": 422
    },
    {
      "completion_length": 57.281251430511475,
      "epoch": 0.22613333333333333,
      "grad_norm": 0.16558815637953483,
      "kl": 0.205078125,
      "learning_rate": 4.374343577351336e-09,
      "loss": 0.0002,
      "reward": 1.1276041865348816,
      "reward_std": 0.09337336057797074,
      "rewards/equation_reward_func": 0.12760416883975267,
      "rewards/format_reward_func": 1.0,
      "step": 424
    },
    {
      "completion_length": 57.273438453674316,
      "epoch": 0.2272,
      "grad_norm": 0.2828010199505921,
      "kl": 0.1973876953125,
      "learning_rate": 3.7288651837012745e-09,
      "loss": 0.0002,
      "reward": 1.1770833730697632,
      "reward_std": 0.15816681599244475,
      "rewards/equation_reward_func": 0.17708333930931985,
      "rewards/format_reward_func": 1.0,
      "step": 426
    },
    {
      "completion_length": 57.86458468437195,
      "epoch": 0.22826666666666667,
      "grad_norm": 0.16310761392861836,
      "kl": 0.20111083984375,
      "learning_rate": 3.134530561862081e-09,
      "loss": 0.0002,
      "reward": 1.140625037252903,
      "reward_std": 0.14361174451187253,
      "rewards/equation_reward_func": 0.14062500419095159,
      "rewards/format_reward_func": 1.0,
      "step": 428
    },
    {
      "completion_length": 56.289063453674316,
      "epoch": 0.22933333333333333,
      "grad_norm": 0.2579970264668867,
      "kl": 0.2320556640625,
      "learning_rate": 2.5914631388619103e-09,
      "loss": 0.0002,
      "reward": 1.1406250298023224,
      "reward_std": 0.10959831206128001,
      "rewards/equation_reward_func": 0.1406250037252903,
      "rewards/format_reward_func": 1.0,
      "step": 430
    },
    {
      "completion_length": 56.97135543823242,
      "epoch": 0.2304,
      "grad_norm": 0.2717261797447828,
      "kl": 0.19573974609375,
      "learning_rate": 2.0997756949353297e-09,
      "loss": 0.0002,
      "reward": 1.1718750298023224,
      "reward_std": 0.13361841812729836,
      "rewards/equation_reward_func": 0.17187500465661287,
      "rewards/format_reward_func": 1.0,
      "step": 432
    },
    {
      "completion_length": 56.51302146911621,
      "epoch": 0.23146666666666665,
      "grad_norm": 0.18185669781914815,
      "kl": 0.2061767578125,
      "learning_rate": 1.6595703401020844e-09,
      "loss": 0.0002,
      "reward": 1.1328125447034836,
      "reward_std": 0.10959667665883899,
      "rewards/equation_reward_func": 0.13281250512227416,
      "rewards/format_reward_func": 1.0,
      "step": 434
    },
    {
      "completion_length": 57.86198043823242,
      "epoch": 0.23253333333333334,
      "grad_norm": 0.2738736381853624,
      "kl": 0.21160888671875,
      "learning_rate": 1.2709384929615596e-09,
      "loss": 0.0002,
      "reward": 1.1848958730697632,
      "reward_std": 0.188415945507586,
      "rewards/equation_reward_func": 0.18489583861082792,
      "rewards/format_reward_func": 1.0,
      "step": 436
    },
    {
      "completion_length": 59.369793176651,
      "epoch": 0.2336,
      "grad_norm": 0.17477850475532822,
      "kl": 0.20587158203125,
      "learning_rate": 9.339608617077165e-10,
      "loss": 0.0002,
      "reward": 1.1223958656191826,
      "reward_std": 0.13273390335962176,
      "rewards/equation_reward_func": 0.12239583721384406,
      "rewards/format_reward_func": 1.0,
      "step": 438
    },
    {
      "completion_length": 56.815105676651,
      "epoch": 0.23466666666666666,
      "grad_norm": 0.3085306578565163,
      "kl": 0.23553466796875,
      "learning_rate": 6.487074273681114e-10,
      "loss": 0.0002,
      "reward": 1.1510417014360428,
      "reward_std": 0.17482286458835006,
      "rewards/equation_reward_func": 0.1510416695382446,
      "rewards/format_reward_func": 1.0,
      "step": 440
    },
    {
      "completion_length": 56.54166769981384,
      "epoch": 0.23573333333333332,
      "grad_norm": 0.2011170990973669,
      "kl": 0.2030029296875,
      "learning_rate": 4.152374292708538e-10,
      "loss": 0.0002,
      "reward": 1.1302083656191826,
      "reward_std": 0.12204980757087469,
      "rewards/equation_reward_func": 0.13020833721384406,
      "rewards/format_reward_func": 1.0,
      "step": 442
    },
    {
      "completion_length": 56.26041793823242,
      "epoch": 0.2368,
      "grad_norm": 0.2688813373112446,
      "kl": 0.21539306640625,
      "learning_rate": 2.3359935274214204e-10,
      "loss": 0.0002,
      "reward": 1.2239583656191826,
      "reward_std": 0.16631885152310133,
      "rewards/equation_reward_func": 0.2265625037252903,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 444
    },
    {
      "completion_length": 56.65885519981384,
      "epoch": 0.23786666666666667,
      "grad_norm": 0.197419015036114,
      "kl": 0.2901611328125,
      "learning_rate": 1.0383091903720665e-10,
      "loss": 0.0003,
      "reward": 1.0703125223517418,
      "reward_std": 0.07724821660667658,
      "rewards/equation_reward_func": 0.07291666814126074,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 446
    },
    {
      "completion_length": 57.031251668930054,
      "epoch": 0.23893333333333333,
      "grad_norm": 0.25032841927694754,
      "kl": 0.21551513671875,
      "learning_rate": 2.595907750671533e-11,
      "loss": 0.0002,
      "reward": 1.1927083730697632,
      "reward_std": 0.1517663886770606,
      "rewards/equation_reward_func": 0.1927083390764892,
      "rewards/format_reward_func": 1.0,
      "step": 448
    },
    {
      "completion_length": 57.27083444595337,
      "epoch": 0.24,
      "grad_norm": 0.22159697962120828,
      "kl": 0.21142578125,
      "learning_rate": 0.0,
      "loss": 0.0002,
      "reward": 1.161458358168602,
      "reward_std": 0.15973949525505304,
      "rewards/equation_reward_func": 0.1614583358168602,
      "rewards/format_reward_func": 1.0,
      "step": 450
    },
    {
      "epoch": 0.24,
      "step": 450,
      "total_flos": 0.0,
      "train_loss": 0.0001983640876704208,
      "train_runtime": 5270.959,
      "train_samples_per_second": 2.049,
      "train_steps_per_second": 0.085
    }
  ],
  "logging_steps": 2,
  "max_steps": 450,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}