{
  "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": 479.06511306762695,
      "epoch": 0.0010666666666666667,
      "grad_norm": 1.1869569851291564,
      "kl": 0.0,
      "learning_rate": 7.142857142857142e-08,
      "loss": -0.0,
      "reward": 0.3776041781529784,
      "reward_std": 0.4827471449971199,
      "rewards/equation_reward_func": 0.05989583441987634,
      "rewards/format_reward_func": 0.31770834140479565,
      "step": 2
    },
    {
      "completion_length": 504.68751525878906,
      "epoch": 0.0021333333333333334,
      "grad_norm": 1.0572288353765749,
      "kl": 0.00037407875061035156,
      "learning_rate": 1.4285714285714285e-07,
      "loss": 0.0,
      "reward": 0.26562500977888703,
      "reward_std": 0.4200526801869273,
      "rewards/equation_reward_func": 0.041666667675599456,
      "rewards/format_reward_func": 0.22395834024064243,
      "step": 4
    },
    {
      "completion_length": 484.2343940734863,
      "epoch": 0.0032,
      "grad_norm": 0.8713849573088774,
      "kl": 0.00042569637298583984,
      "learning_rate": 2.1428571428571426e-07,
      "loss": 0.0,
      "reward": 0.335937506519258,
      "reward_std": 0.4768502954393625,
      "rewards/equation_reward_func": 0.06250000186264515,
      "rewards/format_reward_func": 0.27343750884756446,
      "step": 6
    },
    {
      "completion_length": 504.5208511352539,
      "epoch": 0.004266666666666667,
      "grad_norm": 0.9415822021899375,
      "kl": 0.0004322528839111328,
      "learning_rate": 2.857142857142857e-07,
      "loss": 0.0,
      "reward": 0.2838541753590107,
      "reward_std": 0.4456023368984461,
      "rewards/equation_reward_func": 0.02343750069849193,
      "rewards/format_reward_func": 0.2604166753590107,
      "step": 8
    },
    {
      "completion_length": 501.39324378967285,
      "epoch": 0.005333333333333333,
      "grad_norm": 0.9228145520213509,
      "kl": 0.0004162788391113281,
      "learning_rate": 3.5714285714285716e-07,
      "loss": 0.0,
      "reward": 0.29687500884756446,
      "reward_std": 0.44185679126530886,
      "rewards/equation_reward_func": 0.03385416744276881,
      "rewards/format_reward_func": 0.26302084093913436,
      "step": 10
    },
    {
      "completion_length": 494.34636878967285,
      "epoch": 0.0064,
      "grad_norm": 1.058429498755132,
      "kl": 0.0004417896270751953,
      "learning_rate": 4.285714285714285e-07,
      "loss": 0.0,
      "reward": 0.3203125111758709,
      "reward_std": 0.4622091678902507,
      "rewards/equation_reward_func": 0.04947916814126074,
      "rewards/format_reward_func": 0.27083334047347307,
      "step": 12
    },
    {
      "completion_length": 448.00001525878906,
      "epoch": 0.007466666666666667,
      "grad_norm": 1.1373377904866069,
      "kl": 0.0008156299591064453,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 0.48958335258066654,
      "reward_std": 0.5410491544753313,
      "rewards/equation_reward_func": 0.057291667675599456,
      "rewards/format_reward_func": 0.43229167722165585,
      "step": 14
    },
    {
      "completion_length": 469.2656364440918,
      "epoch": 0.008533333333333334,
      "grad_norm": 1.0961526951904466,
      "kl": 0.0011763572692871094,
      "learning_rate": 4.999740409224932e-07,
      "loss": 0.0,
      "reward": 0.5260416828095913,
      "reward_std": 0.5119463540613651,
      "rewards/equation_reward_func": 0.06510416720993817,
      "rewards/format_reward_func": 0.46093751303851604,
      "step": 16
    },
    {
      "completion_length": 444.05730628967285,
      "epoch": 0.0096,
      "grad_norm": 0.8490475472929186,
      "kl": 0.00301361083984375,
      "learning_rate": 4.998961690809627e-07,
      "loss": 0.0,
      "reward": 0.598958358168602,
      "reward_std": 0.5392562765628099,
      "rewards/equation_reward_func": 0.05208333465270698,
      "rewards/format_reward_func": 0.5468750149011612,
      "step": 18
    },
    {
      "completion_length": 433.02865409851074,
      "epoch": 0.010666666666666666,
      "grad_norm": 0.8530996393331266,
      "kl": 0.0049877166748046875,
      "learning_rate": 4.997664006472578e-07,
      "loss": 0.0,
      "reward": 0.8125000186264515,
      "reward_std": 0.4681414752267301,
      "rewards/equation_reward_func": 0.0703125016298145,
      "rewards/format_reward_func": 0.7421875223517418,
      "step": 20
    },
    {
      "completion_length": 456.56772232055664,
      "epoch": 0.011733333333333333,
      "grad_norm": 0.7697292886020256,
      "kl": 0.00555419921875,
      "learning_rate": 4.995847625707292e-07,
      "loss": 0.0,
      "reward": 0.8281250186264515,
      "reward_std": 0.4001491153612733,
      "rewards/equation_reward_func": 0.0416666679084301,
      "rewards/format_reward_func": 0.7864583507180214,
      "step": 22
    },
    {
      "completion_length": 460.73438453674316,
      "epoch": 0.0128,
      "grad_norm": 0.7848337477423635,
      "kl": 0.0071258544921875,
      "learning_rate": 4.993512925726318e-07,
      "loss": 0.0,
      "reward": 0.8854166865348816,
      "reward_std": 0.3724366007372737,
      "rewards/equation_reward_func": 0.049479167675599456,
      "rewards/format_reward_func": 0.8359375186264515,
      "step": 24
    },
    {
      "completion_length": 434.8932399749756,
      "epoch": 0.013866666666666666,
      "grad_norm": 0.7039638744422072,
      "kl": 0.007732391357421875,
      "learning_rate": 4.990660391382923e-07,
      "loss": 0.0,
      "reward": 0.9375000298023224,
      "reward_std": 0.4019690901041031,
      "rewards/equation_reward_func": 0.07812500209547579,
      "rewards/format_reward_func": 0.8593750223517418,
      "step": 26
    },
    {
      "completion_length": 480.76824378967285,
      "epoch": 0.014933333333333333,
      "grad_norm": 0.705507612255758,
      "kl": 0.010210037231445312,
      "learning_rate": 4.987290615070384e-07,
      "loss": 0.0,
      "reward": 0.958333358168602,
      "reward_std": 0.3932732567191124,
      "rewards/equation_reward_func": 0.09114583558402956,
      "rewards/format_reward_func": 0.8671875223517418,
      "step": 28
    },
    {
      "completion_length": 432.20053482055664,
      "epoch": 0.016,
      "grad_norm": 0.6874108412133968,
      "kl": 0.012973785400390625,
      "learning_rate": 4.983404296598978e-07,
      "loss": 0.0,
      "reward": 1.0364583730697632,
      "reward_std": 0.31663169525563717,
      "rewards/equation_reward_func": 0.10156250349245965,
      "rewards/format_reward_func": 0.9348958507180214,
      "step": 30
    },
    {
      "completion_length": 407.95313835144043,
      "epoch": 0.017066666666666667,
      "grad_norm": 0.6793132961839309,
      "kl": 0.0137176513671875,
      "learning_rate": 4.979002243050646e-07,
      "loss": 0.0,
      "reward": 1.0989583767950535,
      "reward_std": 0.3416873565874994,
      "rewards/equation_reward_func": 0.14843750512227416,
      "rewards/format_reward_func": 0.9505208618938923,
      "step": 32
    },
    {
      "completion_length": 434.5573043823242,
      "epoch": 0.018133333333333335,
      "grad_norm": 0.520698934589799,
      "kl": 0.01392364501953125,
      "learning_rate": 4.974085368611381e-07,
      "loss": 0.0,
      "reward": 1.0052083618938923,
      "reward_std": 0.2466061031445861,
      "rewards/equation_reward_func": 0.06250000093132257,
      "rewards/format_reward_func": 0.9427083507180214,
      "step": 34
    },
    {
      "completion_length": 402.1171989440918,
      "epoch": 0.0192,
      "grad_norm": 0.5349273835140175,
      "kl": 0.013996124267578125,
      "learning_rate": 4.968654694381379e-07,
      "loss": 0.0,
      "reward": 1.0625000149011612,
      "reward_std": 0.20629451051354408,
      "rewards/equation_reward_func": 0.0859375016298145,
      "rewards/format_reward_func": 0.9765625186264515,
      "step": 36
    },
    {
      "completion_length": 412.4765739440918,
      "epoch": 0.020266666666666665,
      "grad_norm": 0.6378526041216871,
      "kl": 0.01590728759765625,
      "learning_rate": 4.962711348162987e-07,
      "loss": 0.0,
      "reward": 1.1067708656191826,
      "reward_std": 0.28954573161900043,
      "rewards/equation_reward_func": 0.13541666930541396,
      "rewards/format_reward_func": 0.9713541865348816,
      "step": 38
    },
    {
      "completion_length": 398.5989685058594,
      "epoch": 0.021333333333333333,
      "grad_norm": 0.656016728201151,
      "kl": 0.01578521728515625,
      "learning_rate": 4.956256564226487e-07,
      "loss": 0.0,
      "reward": 1.070312537252903,
      "reward_std": 0.2771536544896662,
      "rewards/equation_reward_func": 0.10677083628252149,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 40
    },
    {
      "completion_length": 400.770845413208,
      "epoch": 0.0224,
      "grad_norm": 0.8289361768943612,
      "kl": 0.01721954345703125,
      "learning_rate": 4.949291683053768e-07,
      "loss": 0.0,
      "reward": 1.1223958656191826,
      "reward_std": 0.3327407846227288,
      "rewards/equation_reward_func": 0.1484375037252903,
      "rewards/format_reward_func": 0.9739583544433117,
      "step": 42
    },
    {
      "completion_length": 398.7526111602783,
      "epoch": 0.023466666666666667,
      "grad_norm": 0.6768780771715479,
      "kl": 0.017475128173828125,
      "learning_rate": 4.941818151059955e-07,
      "loss": 0.0,
      "reward": 1.0859375335276127,
      "reward_std": 0.24852651357650757,
      "rewards/equation_reward_func": 0.11718750325962901,
      "rewards/format_reward_func": 0.9687500111758709,
      "step": 44
    },
    {
      "completion_length": 390.9244899749756,
      "epoch": 0.024533333333333334,
      "grad_norm": 0.6089623829481902,
      "kl": 0.04027557373046875,
      "learning_rate": 4.933837520293017e-07,
      "loss": 0.0,
      "reward": 1.0729167014360428,
      "reward_std": 0.25113596161827445,
      "rewards/equation_reward_func": 0.10416666814126074,
      "rewards/format_reward_func": 0.9687500223517418,
      "step": 46
    },
    {
      "completion_length": 369.0937614440918,
      "epoch": 0.0256,
      "grad_norm": 0.5754646954335966,
      "kl": 0.0209197998046875,
      "learning_rate": 4.925351448111454e-07,
      "loss": 0.0,
      "reward": 1.1067708730697632,
      "reward_std": 0.24725896958261728,
      "rewards/equation_reward_func": 0.13020833930931985,
      "rewards/format_reward_func": 0.9765625149011612,
      "step": 48
    },
    {
      "completion_length": 388.3880310058594,
      "epoch": 0.02666666666666667,
      "grad_norm": 0.599342426740638,
      "kl": 0.0214385986328125,
      "learning_rate": 4.91636169684011e-07,
      "loss": 0.0,
      "reward": 1.0807292088866234,
      "reward_std": 0.22517748409882188,
      "rewards/equation_reward_func": 0.10416666883975267,
      "rewards/format_reward_func": 0.9765625111758709,
      "step": 50
    },
    {
      "completion_length": 370.52345085144043,
      "epoch": 0.027733333333333332,
      "grad_norm": 0.6909841937162223,
      "kl": 0.0216064453125,
      "learning_rate": 4.906870133404186e-07,
      "loss": 0.0,
      "reward": 1.0833333693444729,
      "reward_std": 0.262856621760875,
      "rewards/equation_reward_func": 0.11458333674818277,
      "rewards/format_reward_func": 0.9687500186264515,
      "step": 52
    },
    {
      "completion_length": 374.856782913208,
      "epoch": 0.0288,
      "grad_norm": 0.6283094034758113,
      "kl": 0.0230560302734375,
      "learning_rate": 4.896878728941531e-07,
      "loss": 0.0,
      "reward": 1.1015625223517418,
      "reward_std": 0.2443550513125956,
      "rewards/equation_reward_func": 0.1276041695382446,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 54
    },
    {
      "completion_length": 364.94011306762695,
      "epoch": 0.029866666666666666,
      "grad_norm": 0.7799936650160872,
      "kl": 0.0264739990234375,
      "learning_rate": 4.886389558393284e-07,
      "loss": 0.0,
      "reward": 1.1666667088866234,
      "reward_std": 0.33124217577278614,
      "rewards/equation_reward_func": 0.19010417396202683,
      "rewards/format_reward_func": 0.9765625074505806,
      "step": 56
    },
    {
      "completion_length": 352.15365409851074,
      "epoch": 0.030933333333333334,
      "grad_norm": 0.6990776726434789,
      "kl": 0.0289306640625,
      "learning_rate": 4.875404800072976e-07,
      "loss": 0.0,
      "reward": 1.1093750447034836,
      "reward_std": 0.22618821123614907,
      "rewards/equation_reward_func": 0.12760416977107525,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 58
    },
    {
      "completion_length": 347.57553482055664,
      "epoch": 0.032,
      "grad_norm": 0.7882548556903084,
      "kl": 0.02877044677734375,
      "learning_rate": 4.86392673521415e-07,
      "loss": 0.0,
      "reward": 1.2187500447034836,
      "reward_std": 0.32150182826444507,
      "rewards/equation_reward_func": 0.2291666737291962,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 60
    },
    {
      "completion_length": 364.71875953674316,
      "epoch": 0.03306666666666667,
      "grad_norm": 0.6290974292701452,
      "kl": 0.02837371826171875,
      "learning_rate": 4.851957747496606e-07,
      "loss": 0.0,
      "reward": 1.143229190260172,
      "reward_std": 0.2681962251663208,
      "rewards/equation_reward_func": 0.1718750037252903,
      "rewards/format_reward_func": 0.971354179084301,
      "step": 62
    },
    {
      "completion_length": 351.825532913208,
      "epoch": 0.034133333333333335,
      "grad_norm": 0.7524850668668323,
      "kl": 0.03217315673828125,
      "learning_rate": 4.839500322551386e-07,
      "loss": 0.0,
      "reward": 1.1588542088866234,
      "reward_std": 0.2972170915454626,
      "rewards/equation_reward_func": 0.17968750488944352,
      "rewards/format_reward_func": 0.9791666753590107,
      "step": 64
    },
    {
      "completion_length": 360.0833435058594,
      "epoch": 0.0352,
      "grad_norm": 0.6380613373024979,
      "kl": 0.03417205810546875,
      "learning_rate": 4.826557047444563e-07,
      "loss": 0.0,
      "reward": 1.1302083805203438,
      "reward_std": 0.2470923252403736,
      "rewards/equation_reward_func": 0.15625000465661287,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 66
    },
    {
      "completion_length": 344.16928482055664,
      "epoch": 0.03626666666666667,
      "grad_norm": 0.8763117130489186,
      "kl": 0.03891754150390625,
      "learning_rate": 4.813130610139993e-07,
      "loss": 0.0,
      "reward": 1.1614583879709244,
      "reward_std": 0.31256247218698263,
      "rewards/equation_reward_func": 0.17968750605359674,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 68
    },
    {
      "completion_length": 374.7916774749756,
      "epoch": 0.037333333333333336,
      "grad_norm": 0.6775447687819961,
      "kl": 0.0399627685546875,
      "learning_rate": 4.799223798941089e-07,
      "loss": 0.0,
      "reward": 1.1223958507180214,
      "reward_std": 0.2880659820511937,
      "rewards/equation_reward_func": 0.14843750279396772,
      "rewards/format_reward_func": 0.9739583507180214,
      "step": 70
    },
    {
      "completion_length": 341.1588668823242,
      "epoch": 0.0384,
      "grad_norm": 0.6315374119974878,
      "kl": 0.045684814453125,
      "learning_rate": 4.78483950191177e-07,
      "loss": 0.0,
      "reward": 1.171875037252903,
      "reward_std": 0.2657899674959481,
      "rewards/equation_reward_func": 0.1953125069849193,
      "rewards/format_reward_func": 0.9765625074505806,
      "step": 72
    },
    {
      "completion_length": 347.85938453674316,
      "epoch": 0.039466666666666664,
      "grad_norm": 0.6797093361910038,
      "kl": 0.0632171630859375,
      "learning_rate": 4.769980706276687e-07,
      "loss": 0.0001,
      "reward": 1.166666716337204,
      "reward_std": 0.28392915101721883,
      "rewards/equation_reward_func": 0.18489583884365857,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 74
    },
    {
      "completion_length": 358.37500953674316,
      "epoch": 0.04053333333333333,
      "grad_norm": 0.7347841080142689,
      "kl": 0.05291748046875,
      "learning_rate": 4.7546504978008595e-07,
      "loss": 0.0001,
      "reward": 1.226562537252903,
      "reward_std": 0.34029595321044326,
      "rewards/equation_reward_func": 0.24739584024064243,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 76
    },
    {
      "completion_length": 334.1198024749756,
      "epoch": 0.0416,
      "grad_norm": 0.6278097170931212,
      "kl": 0.0596466064453125,
      "learning_rate": 4.738852060148848e-07,
      "loss": 0.0001,
      "reward": 1.1432291939854622,
      "reward_std": 0.23812766559422016,
      "rewards/equation_reward_func": 0.15885417023673654,
      "rewards/format_reward_func": 0.9843750111758709,
      "step": 78
    },
    {
      "completion_length": 344.8177146911621,
      "epoch": 0.042666666666666665,
      "grad_norm": 0.6997187790917091,
      "kl": 0.061279296875,
      "learning_rate": 4.722588674223593e-07,
      "loss": 0.0001,
      "reward": 1.1666666939854622,
      "reward_std": 0.26324747782200575,
      "rewards/equation_reward_func": 0.18489583767950535,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 80
    },
    {
      "completion_length": 309.60938453674316,
      "epoch": 0.04373333333333333,
      "grad_norm": 0.6997044550717475,
      "kl": 0.070892333984375,
      "learning_rate": 4.70586371748506e-07,
      "loss": 0.0001,
      "reward": 1.2213542014360428,
      "reward_std": 0.2756146062165499,
      "rewards/equation_reward_func": 0.23177083861082792,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 82
    },
    {
      "completion_length": 335.4453239440918,
      "epoch": 0.0448,
      "grad_norm": 0.683238903553104,
      "kl": 0.0697021484375,
      "learning_rate": 4.6886806632488363e-07,
      "loss": 0.0001,
      "reward": 1.1614583656191826,
      "reward_std": 0.27705384651198983,
      "rewards/equation_reward_func": 0.18229167093522847,
      "rewards/format_reward_func": 0.979166679084301,
      "step": 84
    },
    {
      "completion_length": 305.8229236602783,
      "epoch": 0.04586666666666667,
      "grad_norm": 0.5785488696587523,
      "kl": 0.0760498046875,
      "learning_rate": 4.6710430799648143e-07,
      "loss": 0.0001,
      "reward": 1.1822917014360428,
      "reward_std": 0.21207982301712036,
      "rewards/equation_reward_func": 0.1875000058207661,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 86
    },
    {
      "completion_length": 313.07552909851074,
      "epoch": 0.046933333333333334,
      "grad_norm": 0.8039995179667074,
      "kl": 0.0819091796875,
      "learning_rate": 4.652954630476127e-07,
      "loss": 0.0001,
      "reward": 1.2161458805203438,
      "reward_std": 0.3164127543568611,
      "rewards/equation_reward_func": 0.22916667512618005,
      "rewards/format_reward_func": 0.986979179084301,
      "step": 88
    },
    {
      "completion_length": 321.54948902130127,
      "epoch": 0.048,
      "grad_norm": 0.7157243173905158,
      "kl": 0.093475341796875,
      "learning_rate": 4.6344190712584713e-07,
      "loss": 0.0001,
      "reward": 1.1927083805203438,
      "reward_std": 0.21211858373135328,
      "rewards/equation_reward_func": 0.20312500628642738,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 90
    },
    {
      "completion_length": 289.03386402130127,
      "epoch": 0.04906666666666667,
      "grad_norm": 0.9028411325076038,
      "kl": 0.0883941650390625,
      "learning_rate": 4.615440251639995e-07,
      "loss": 0.0001,
      "reward": 1.330729216337204,
      "reward_std": 0.28467333829030395,
      "rewards/equation_reward_func": 0.33333334140479565,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 92
    },
    {
      "completion_length": 304.7942819595337,
      "epoch": 0.050133333333333335,
      "grad_norm": 0.7882629426423002,
      "kl": 0.09014892578125,
      "learning_rate": 4.596022113001894e-07,
      "loss": 0.0001,
      "reward": 1.2890625521540642,
      "reward_std": 0.29563618497923017,
      "rewards/equation_reward_func": 0.2968750095460564,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 94
    },
    {
      "completion_length": 336.986985206604,
      "epoch": 0.0512,
      "grad_norm": 0.7426512420493854,
      "kl": 0.086090087890625,
      "learning_rate": 4.576168687959895e-07,
      "loss": 0.0001,
      "reward": 1.2552083656191826,
      "reward_std": 0.348067122977227,
      "rewards/equation_reward_func": 0.276041675824672,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 96
    },
    {
      "completion_length": 318.2812614440918,
      "epoch": 0.05226666666666667,
      "grad_norm": 0.8480603451919824,
      "kl": 0.09051513671875,
      "learning_rate": 4.555884099526793e-07,
      "loss": 0.0001,
      "reward": 1.2890625335276127,
      "reward_std": 0.3150825258344412,
      "rewards/equation_reward_func": 0.3072916748933494,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 98
    },
    {
      "completion_length": 352.70313453674316,
      "epoch": 0.05333333333333334,
      "grad_norm": 0.6067660652220382,
      "kl": 0.089141845703125,
      "learning_rate": 4.5351725602562174e-07,
      "loss": 0.0001,
      "reward": 1.2916667014360428,
      "reward_std": 0.3160873227752745,
      "rewards/equation_reward_func": 0.32031250768341124,
      "rewards/format_reward_func": 0.9713541828095913,
      "step": 100
    },
    {
      "completion_length": 316.4739646911621,
      "epoch": 0.0544,
      "grad_norm": 0.7599757297215466,
      "kl": 0.08282470703125,
      "learning_rate": 4.514038371367791e-07,
      "loss": 0.0001,
      "reward": 1.3177083879709244,
      "reward_std": 0.316761318128556,
      "rewards/equation_reward_func": 0.32552084513008595,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 102
    },
    {
      "completion_length": 346.94793033599854,
      "epoch": 0.055466666666666664,
      "grad_norm": 0.7139942682254907,
      "kl": 0.082855224609375,
      "learning_rate": 4.4924859218538936e-07,
      "loss": 0.0001,
      "reward": 1.2343750298023224,
      "reward_std": 0.28453153278678656,
      "rewards/equation_reward_func": 0.247395837912336,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 104
    },
    {
      "completion_length": 333.86459159851074,
      "epoch": 0.05653333333333333,
      "grad_norm": 0.7262973137337115,
      "kl": 0.081298828125,
      "learning_rate": 4.470519687568185e-07,
      "loss": 0.0001,
      "reward": 1.216145858168602,
      "reward_std": 0.2771544805727899,
      "rewards/equation_reward_func": 0.23958334210328758,
      "rewards/format_reward_func": 0.9765625149011612,
      "step": 106
    },
    {
      "completion_length": 347.1354236602783,
      "epoch": 0.0576,
      "grad_norm": 0.8289662610034518,
      "kl": 0.0821533203125,
      "learning_rate": 4.4481442302960923e-07,
      "loss": 0.0001,
      "reward": 1.322916716337204,
      "reward_std": 0.3112520845606923,
      "rewards/equation_reward_func": 0.34375001257285476,
      "rewards/format_reward_func": 0.979166679084301,
      "step": 108
    },
    {
      "completion_length": 358.14323806762695,
      "epoch": 0.058666666666666666,
      "grad_norm": 0.7194510877107703,
      "kl": 0.09747314453125,
      "learning_rate": 4.4253641968074505e-07,
      "loss": 0.0001,
      "reward": 1.2656250447034836,
      "reward_std": 0.27732502575963736,
      "rewards/equation_reward_func": 0.29166667233221233,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 110
    },
    {
      "completion_length": 353.95573806762695,
      "epoch": 0.05973333333333333,
      "grad_norm": 0.8309996583293423,
      "kl": 0.08343505859375,
      "learning_rate": 4.402184317891501e-07,
      "loss": 0.0001,
      "reward": 1.3046875596046448,
      "reward_std": 0.3259678762406111,
      "rewards/equation_reward_func": 0.32552084466442466,
      "rewards/format_reward_func": 0.9791666753590107,
      "step": 112
    },
    {
      "completion_length": 340.52345085144043,
      "epoch": 0.0608,
      "grad_norm": 0.860260137408963,
      "kl": 0.085205078125,
      "learning_rate": 4.37860940737443e-07,
      "loss": 0.0001,
      "reward": 1.3645833656191826,
      "reward_std": 0.28312036022543907,
      "rewards/equation_reward_func": 0.3854166744276881,
      "rewards/format_reward_func": 0.979166679084301,
      "step": 114
    },
    {
      "completion_length": 369.455735206604,
      "epoch": 0.06186666666666667,
      "grad_norm": 0.8461609359358174,
      "kl": 0.092681884765625,
      "learning_rate": 4.354644361119671e-07,
      "loss": 0.0001,
      "reward": 1.3489583656191826,
      "reward_std": 0.3290282329544425,
      "rewards/equation_reward_func": 0.38541667675599456,
      "rewards/format_reward_func": 0.963541679084301,
      "step": 116
    },
    {
      "completion_length": 392.994797706604,
      "epoch": 0.06293333333333333,
      "grad_norm": 0.7075500368751798,
      "kl": 0.0854949951171875,
      "learning_rate": 4.3302941560111716e-07,
      "loss": 0.0001,
      "reward": 1.3020833767950535,
      "reward_std": 0.38001349521800876,
      "rewards/equation_reward_func": 0.3515625090803951,
      "rewards/format_reward_func": 0.9505208544433117,
      "step": 118
    },
    {
      "completion_length": 386.47136306762695,
      "epoch": 0.064,
      "grad_norm": 0.7448272303035507,
      "kl": 0.095428466796875,
      "learning_rate": 4.3055638489198236e-07,
      "loss": 0.0001,
      "reward": 1.3281250447034836,
      "reward_std": 0.2941426238976419,
      "rewards/equation_reward_func": 0.3645833453629166,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 120
    },
    {
      "completion_length": 390.67970085144043,
      "epoch": 0.06506666666666666,
      "grad_norm": 0.6028452890831756,
      "kl": 0.102203369140625,
      "learning_rate": 4.280458575653296e-07,
      "loss": 0.0001,
      "reward": 1.3151042088866234,
      "reward_std": 0.3121520522981882,
      "rewards/equation_reward_func": 0.3593750149011612,
      "rewards/format_reward_func": 0.9557291828095913,
      "step": 122
    },
    {
      "completion_length": 411.3906307220459,
      "epoch": 0.06613333333333334,
      "grad_norm": 0.6972432278932138,
      "kl": 0.1053466796875,
      "learning_rate": 4.2549835498894665e-07,
      "loss": 0.0001,
      "reward": 1.3046875298023224,
      "reward_std": 0.3263821485452354,
      "rewards/equation_reward_func": 0.3541666744276881,
      "rewards/format_reward_func": 0.9505208469927311,
      "step": 124
    },
    {
      "completion_length": 398.1380367279053,
      "epoch": 0.0672,
      "grad_norm": 0.5908309999266357,
      "kl": 0.11505126953125,
      "learning_rate": 4.229144062093679e-07,
      "loss": 0.0001,
      "reward": 1.3645833656191826,
      "reward_std": 0.30290936632081866,
      "rewards/equation_reward_func": 0.414062513737008,
      "rewards/format_reward_func": 0.9505208544433117,
      "step": 126
    },
    {
      "completion_length": 407.2057456970215,
      "epoch": 0.06826666666666667,
      "grad_norm": 0.696587638578745,
      "kl": 0.113983154296875,
      "learning_rate": 4.2029454784200675e-07,
      "loss": 0.0001,
      "reward": 1.3723958730697632,
      "reward_std": 0.32597592612728477,
      "rewards/equation_reward_func": 0.41927084675990045,
      "rewards/format_reward_func": 0.9531250111758709,
      "step": 128
    },
    {
      "completion_length": 401.0286531448364,
      "epoch": 0.06933333333333333,
      "grad_norm": 0.6464987173940265,
      "kl": 0.109283447265625,
      "learning_rate": 4.1763932395971433e-07,
      "loss": 0.0001,
      "reward": 1.3593750223517418,
      "reward_std": 0.283989277202636,
      "rewards/equation_reward_func": 0.3984375111758709,
      "rewards/format_reward_func": 0.9609375149011612,
      "step": 130
    },
    {
      "completion_length": 443.7916851043701,
      "epoch": 0.0704,
      "grad_norm": 0.7004776947090245,
      "kl": 0.105438232421875,
      "learning_rate": 4.1494928597979117e-07,
      "loss": 0.0001,
      "reward": 1.289062537252903,
      "reward_std": 0.3336244923993945,
      "rewards/equation_reward_func": 0.34114584419876337,
      "rewards/format_reward_func": 0.9479166902601719,
      "step": 132
    },
    {
      "completion_length": 415.70313262939453,
      "epoch": 0.07146666666666666,
      "grad_norm": 0.6650285479645688,
      "kl": 0.1195068359375,
      "learning_rate": 4.122249925494726e-07,
      "loss": 0.0001,
      "reward": 1.3333333656191826,
      "reward_std": 0.3276231326162815,
      "rewards/equation_reward_func": 0.38020834140479565,
      "rewards/format_reward_func": 0.9531250260770321,
      "step": 134
    },
    {
      "completion_length": 397.13542652130127,
      "epoch": 0.07253333333333334,
      "grad_norm": 0.6650549480411382,
      "kl": 0.12451171875,
      "learning_rate": 4.094670094299131e-07,
      "loss": 0.0001,
      "reward": 1.4140625335276127,
      "reward_std": 0.33390075946226716,
      "rewards/equation_reward_func": 0.4557291781529784,
      "rewards/format_reward_func": 0.958333358168602,
      "step": 136
    },
    {
      "completion_length": 427.82293033599854,
      "epoch": 0.0736,
      "grad_norm": 0.58867517657451,
      "kl": 0.1121826171875,
      "learning_rate": 4.066759093786931e-07,
      "loss": 0.0001,
      "reward": 1.3046875521540642,
      "reward_std": 0.2838647039607167,
      "rewards/equation_reward_func": 0.3567708474583924,
      "rewards/format_reward_func": 0.9479166828095913,
      "step": 138
    },
    {
      "completion_length": 435.25261878967285,
      "epoch": 0.07466666666666667,
      "grad_norm": 0.7308572976998943,
      "kl": 0.1190185546875,
      "learning_rate": 4.038522720308732e-07,
      "loss": 0.0001,
      "reward": 1.2994792088866234,
      "reward_std": 0.32467597210779786,
      "rewards/equation_reward_func": 0.3723958465270698,
      "rewards/format_reward_func": 0.9270833469927311,
      "step": 140
    },
    {
      "completion_length": 399.5807399749756,
      "epoch": 0.07573333333333333,
      "grad_norm": 0.7010882655857836,
      "kl": 0.128326416015625,
      "learning_rate": 4.009966837786194e-07,
      "loss": 0.0001,
      "reward": 1.4166667014360428,
      "reward_std": 0.34676419477909803,
      "rewards/equation_reward_func": 0.48437501303851604,
      "rewards/format_reward_func": 0.932291679084301,
      "step": 142
    },
    {
      "completion_length": 385.9557418823242,
      "epoch": 0.0768,
      "grad_norm": 0.7470074693923137,
      "kl": 0.162322998046875,
      "learning_rate": 3.981097376494259e-07,
      "loss": 0.0002,
      "reward": 1.4010417014360428,
      "reward_std": 0.26487945253029466,
      "rewards/equation_reward_func": 0.43750001466833055,
      "rewards/format_reward_func": 0.963541679084301,
      "step": 144
    },
    {
      "completion_length": 408.0442810058594,
      "epoch": 0.07786666666666667,
      "grad_norm": 0.6999375294187344,
      "kl": 0.137908935546875,
      "learning_rate": 3.951920331829592e-07,
      "loss": 0.0001,
      "reward": 1.3828125484287739,
      "reward_std": 0.29149100510403514,
      "rewards/equation_reward_func": 0.4375000128056854,
      "rewards/format_reward_func": 0.9453125186264515,
      "step": 146
    },
    {
      "completion_length": 403.9765729904175,
      "epoch": 0.07893333333333333,
      "grad_norm": 0.5287881916983996,
      "kl": 0.151641845703125,
      "learning_rate": 3.922441763065506e-07,
      "loss": 0.0002,
      "reward": 1.416666705161333,
      "reward_std": 0.28373547829687595,
      "rewards/equation_reward_func": 0.4817708465270698,
      "rewards/format_reward_func": 0.9348958544433117,
      "step": 148
    },
    {
      "completion_length": 410.20573902130127,
      "epoch": 0.08,
      "grad_norm": 0.5024042458612308,
      "kl": 0.14007568359375,
      "learning_rate": 3.8926677920936093e-07,
      "loss": 0.0001,
      "reward": 1.3619792088866234,
      "reward_std": 0.28874032804742455,
      "rewards/equation_reward_func": 0.4218750086147338,
      "rewards/format_reward_func": 0.9401041865348816,
      "step": 150
    },
    {
      "completion_length": 404.48959732055664,
      "epoch": 0.08106666666666666,
      "grad_norm": 0.4963601555912834,
      "kl": 0.1383056640625,
      "learning_rate": 3.862604602152464e-07,
      "loss": 0.0001,
      "reward": 1.4088541939854622,
      "reward_std": 0.2753666895441711,
      "rewards/equation_reward_func": 0.4687500107102096,
      "rewards/format_reward_func": 0.9401041865348816,
      "step": 152
    },
    {
      "completion_length": 413.2682399749756,
      "epoch": 0.08213333333333334,
      "grad_norm": 0.8055471860932037,
      "kl": 0.279388427734375,
      "learning_rate": 3.8322584365434934e-07,
      "loss": 0.0003,
      "reward": 1.3802083656191826,
      "reward_std": 0.26039192313328385,
      "rewards/equation_reward_func": 0.42968751210719347,
      "rewards/format_reward_func": 0.950520858168602,
      "step": 154
    },
    {
      "completion_length": 402.37240505218506,
      "epoch": 0.0832,
      "grad_norm": 0.42808798597812875,
      "kl": 0.150421142578125,
      "learning_rate": 3.8016355973344173e-07,
      "loss": 0.0002,
      "reward": 1.3593750447034836,
      "reward_std": 0.1753570344299078,
      "rewards/equation_reward_func": 0.4062500090803951,
      "rewards/format_reward_func": 0.9531250149011612,
      "step": 156
    },
    {
      "completion_length": 362.34376335144043,
      "epoch": 0.08426666666666667,
      "grad_norm": 0.7017085808063098,
      "kl": 0.1455078125,
      "learning_rate": 3.7707424440504863e-07,
      "loss": 0.0001,
      "reward": 1.416666705161333,
      "reward_std": 0.2758899559266865,
      "rewards/equation_reward_func": 0.4635416786186397,
      "rewards/format_reward_func": 0.9531250111758709,
      "step": 158
    },
    {
      "completion_length": 402.34636402130127,
      "epoch": 0.08533333333333333,
      "grad_norm": 0.7033098011768847,
      "kl": 0.13641357421875,
      "learning_rate": 3.739585392353787e-07,
      "loss": 0.0001,
      "reward": 1.398437537252903,
      "reward_std": 0.2615419952198863,
      "rewards/equation_reward_func": 0.4375000111758709,
      "rewards/format_reward_func": 0.9609375111758709,
      "step": 160
    },
    {
      "completion_length": 349.5468854904175,
      "epoch": 0.0864,
      "grad_norm": 0.8470177529007129,
      "kl": 0.17486572265625,
      "learning_rate": 3.7081709127108767e-07,
      "loss": 0.0002,
      "reward": 1.4609375447034836,
      "reward_std": 0.21957123186439276,
      "rewards/equation_reward_func": 0.4791666835080832,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 162
    },
    {
      "completion_length": 408.7864656448364,
      "epoch": 0.08746666666666666,
      "grad_norm": 0.6697773440506227,
      "kl": 0.13555908203125,
      "learning_rate": 3.6765055290490513e-07,
      "loss": 0.0001,
      "reward": 1.3645833767950535,
      "reward_std": 0.24416533997282386,
      "rewards/equation_reward_func": 0.40885417559184134,
      "rewards/format_reward_func": 0.9557291828095913,
      "step": 164
    },
    {
      "completion_length": 377.4479274749756,
      "epoch": 0.08853333333333334,
      "grad_norm": 0.5585350437896919,
      "kl": 0.15234375,
      "learning_rate": 3.644595817401501e-07,
      "loss": 0.0002,
      "reward": 1.3958333767950535,
      "reward_std": 0.19870867347344756,
      "rewards/equation_reward_func": 0.4348958439659327,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 166
    },
    {
      "completion_length": 375.86198902130127,
      "epoch": 0.0896,
      "grad_norm": 0.5191461899813266,
      "kl": 0.14447021484375,
      "learning_rate": 3.6124484045416483e-07,
      "loss": 0.0001,
      "reward": 1.4244792126119137,
      "reward_std": 0.21074969647452235,
      "rewards/equation_reward_func": 0.4479166853707284,
      "rewards/format_reward_func": 0.9765625074505806,
      "step": 168
    },
    {
      "completion_length": 367.8724069595337,
      "epoch": 0.09066666666666667,
      "grad_norm": 0.634261109914142,
      "kl": 0.1561279296875,
      "learning_rate": 3.580069966606949e-07,
      "loss": 0.0002,
      "reward": 1.4140625447034836,
      "reward_std": 0.23517010640352964,
      "rewards/equation_reward_func": 0.4375000095460564,
      "rewards/format_reward_func": 0.9765625111758709,
      "step": 170
    },
    {
      "completion_length": 420.29949378967285,
      "epoch": 0.09173333333333333,
      "grad_norm": 0.6666515479679946,
      "kl": 0.17108154296875,
      "learning_rate": 3.547467227712444e-07,
      "loss": 0.0002,
      "reward": 1.3671875335276127,
      "reward_std": 0.23187633976340294,
      "rewards/equation_reward_func": 0.411458347691223,
      "rewards/format_reward_func": 0.9557291902601719,
      "step": 172
    },
    {
      "completion_length": 380.9713659286499,
      "epoch": 0.0928,
      "grad_norm": 0.5233105345475487,
      "kl": 0.151123046875,
      "learning_rate": 3.5146469585543386e-07,
      "loss": 0.0002,
      "reward": 1.4635417088866234,
      "reward_std": 0.21822098083794117,
      "rewards/equation_reward_func": 0.4973958465270698,
      "rewards/format_reward_func": 0.9661458544433117,
      "step": 174
    },
    {
      "completion_length": 367.92969512939453,
      "epoch": 0.09386666666666667,
      "grad_norm": 0.5157976215776814,
      "kl": 0.148345947265625,
      "learning_rate": 3.481615975003922e-07,
      "loss": 0.0001,
      "reward": 1.4244792014360428,
      "reward_std": 0.21948427613824606,
      "rewards/equation_reward_func": 0.45572917768731713,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 176
    },
    {
      "completion_length": 376.94011878967285,
      "epoch": 0.09493333333333333,
      "grad_norm": 0.6881879649420064,
      "kl": 0.17608642578125,
      "learning_rate": 3.448381136692089e-07,
      "loss": 0.0002,
      "reward": 1.3984375484287739,
      "reward_std": 0.2576333871111274,
      "rewards/equation_reward_func": 0.4348958448972553,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 178
    },
    {
      "completion_length": 390.6302185058594,
      "epoch": 0.096,
      "grad_norm": 0.8531012725646003,
      "kl": 0.1448974609375,
      "learning_rate": 3.4149493455847897e-07,
      "loss": 0.0001,
      "reward": 1.369791716337204,
      "reward_std": 0.229800954926759,
      "rewards/equation_reward_func": 0.3906250102445483,
      "rewards/format_reward_func": 0.979166679084301,
      "step": 180
    },
    {
      "completion_length": 391.755220413208,
      "epoch": 0.09706666666666666,
      "grad_norm": 0.8499423182750618,
      "kl": 0.14874267578125,
      "learning_rate": 3.3813275445496766e-07,
      "loss": 0.0001,
      "reward": 1.4114583730697632,
      "reward_std": 0.24884725222364068,
      "rewards/equation_reward_func": 0.4348958469927311,
      "rewards/format_reward_func": 0.9765625149011612,
      "step": 182
    },
    {
      "completion_length": 432.4192810058594,
      "epoch": 0.09813333333333334,
      "grad_norm": 0.6105983090425946,
      "kl": 0.15069580078125,
      "learning_rate": 3.347522715914262e-07,
      "loss": 0.0002,
      "reward": 1.3125000409781933,
      "reward_std": 0.18763548182323575,
      "rewards/equation_reward_func": 0.34635417559184134,
      "rewards/format_reward_func": 0.9661458469927311,
      "step": 184
    },
    {
      "completion_length": 347.3046941757202,
      "epoch": 0.0992,
      "grad_norm": 0.5844753474786499,
      "kl": 0.1790771484375,
      "learning_rate": 3.313541880015877e-07,
      "loss": 0.0002,
      "reward": 1.4843750558793545,
      "reward_std": 0.15258984873071313,
      "rewards/equation_reward_func": 0.5104166809469461,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 186
    },
    {
      "completion_length": 347.9427185058594,
      "epoch": 0.10026666666666667,
      "grad_norm": 0.6723125365768322,
      "kl": 0.14996337890625,
      "learning_rate": 3.279392093743747e-07,
      "loss": 0.0002,
      "reward": 1.5052083730697632,
      "reward_std": 0.1946416893042624,
      "rewards/equation_reward_func": 0.5156250139698386,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 188
    },
    {
      "completion_length": 375.9453239440918,
      "epoch": 0.10133333333333333,
      "grad_norm": 0.9149096307542565,
      "kl": 0.17333984375,
      "learning_rate": 3.245080449073459e-07,
      "loss": 0.0002,
      "reward": 1.4062500447034836,
      "reward_std": 0.17988635739311576,
      "rewards/equation_reward_func": 0.42968751094304025,
      "rewards/format_reward_func": 0.9765625186264515,
      "step": 190
    },
    {
      "completion_length": 379.95834827423096,
      "epoch": 0.1024,
      "grad_norm": 0.8006457880594571,
      "kl": 0.171051025390625,
      "learning_rate": 3.210614071594162e-07,
      "loss": 0.0002,
      "reward": 1.346354205161333,
      "reward_std": 0.19684508023783565,
      "rewards/equation_reward_func": 0.3671875058207661,
      "rewards/format_reward_func": 0.9791666753590107,
      "step": 192
    },
    {
      "completion_length": 362.75261402130127,
      "epoch": 0.10346666666666667,
      "grad_norm": 1.2837479387779884,
      "kl": 0.1580810546875,
      "learning_rate": 3.1760001190287695e-07,
      "loss": 0.0002,
      "reward": 1.406250037252903,
      "reward_std": 0.19557411642745137,
      "rewards/equation_reward_func": 0.42447917722165585,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 194
    },
    {
      "completion_length": 362.6302146911621,
      "epoch": 0.10453333333333334,
      "grad_norm": 0.6465378835667693,
      "kl": 0.152679443359375,
      "learning_rate": 3.141245779747502e-07,
      "loss": 0.0002,
      "reward": 1.4192708767950535,
      "reward_std": 0.17333257384598255,
      "rewards/equation_reward_func": 0.4401041816454381,
      "rewards/format_reward_func": 0.9791666753590107,
      "step": 196
    },
    {
      "completion_length": 354.5338625907898,
      "epoch": 0.1056,
      "grad_norm": 0.5360554028298243,
      "kl": 0.178070068359375,
      "learning_rate": 3.106358271275056e-07,
      "loss": 0.0002,
      "reward": 1.4140625447034836,
      "reward_std": 0.16571144526824355,
      "rewards/equation_reward_func": 0.4322916786186397,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 198
    },
    {
      "completion_length": 385.5208435058594,
      "epoch": 0.10666666666666667,
      "grad_norm": 0.5592760034852065,
      "kl": 0.15142822265625,
      "learning_rate": 3.0713448387917227e-07,
      "loss": 0.0002,
      "reward": 1.3463541939854622,
      "reward_std": 0.1820155093446374,
      "rewards/equation_reward_func": 0.35677084093913436,
      "rewards/format_reward_func": 0.9895833358168602,
      "step": 200
    },
    {
      "completion_length": 377.02084827423096,
      "epoch": 0.10773333333333333,
      "grad_norm": 0.5785507707354157,
      "kl": 0.174224853515625,
      "learning_rate": 3.0362127536287636e-07,
      "loss": 0.0002,
      "reward": 1.3541667088866234,
      "reward_std": 0.13336461875587702,
      "rewards/equation_reward_func": 0.36197917349636555,
      "rewards/format_reward_func": 0.9921875074505806,
      "step": 202
    },
    {
      "completion_length": 339.1198043823242,
      "epoch": 0.1088,
      "grad_norm": 0.49158548802495156,
      "kl": 0.180938720703125,
      "learning_rate": 3.0009693117583523e-07,
      "loss": 0.0002,
      "reward": 1.4505208693444729,
      "reward_std": 0.18924278765916824,
      "rewards/equation_reward_func": 0.47135418141260743,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 204
    },
    {
      "completion_length": 361.62500762939453,
      "epoch": 0.10986666666666667,
      "grad_norm": 0.6202583629431399,
      "kl": 0.19464111328125,
      "learning_rate": 2.965621832278401e-07,
      "loss": 0.0002,
      "reward": 1.4192708656191826,
      "reward_std": 0.186479932628572,
      "rewards/equation_reward_func": 0.4427083437331021,
      "rewards/format_reward_func": 0.9765625111758709,
      "step": 206
    },
    {
      "completion_length": 349.294282913208,
      "epoch": 0.11093333333333333,
      "grad_norm": 0.6453352495768684,
      "kl": 0.18829345703125,
      "learning_rate": 2.9301776558925875e-07,
      "loss": 0.0002,
      "reward": 1.4817708805203438,
      "reward_std": 0.15738636022433639,
      "rewards/equation_reward_func": 0.4973958469927311,
      "rewards/format_reward_func": 0.9843750149011612,
      "step": 208
    },
    {
      "completion_length": 373.44271659851074,
      "epoch": 0.112,
      "grad_norm": 0.5146343243630107,
      "kl": 0.180908203125,
      "learning_rate": 2.894644143385885e-07,
      "loss": 0.0002,
      "reward": 1.3984375521540642,
      "reward_std": 0.1677132830955088,
      "rewards/equation_reward_func": 0.42968750931322575,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 210
    },
    {
      "completion_length": 339.29948711395264,
      "epoch": 0.11306666666666666,
      "grad_norm": 0.6039103056491067,
      "kl": 0.2369384765625,
      "learning_rate": 2.859028674095937e-07,
      "loss": 0.0002,
      "reward": 1.4635417088866234,
      "reward_std": 0.11582242278382182,
      "rewards/equation_reward_func": 0.4661458439659327,
      "rewards/format_reward_func": 0.9973958358168602,
      "step": 212
    },
    {
      "completion_length": 348.96354961395264,
      "epoch": 0.11413333333333334,
      "grad_norm": 0.6551166906763881,
      "kl": 0.19390869140625,
      "learning_rate": 2.823338644380566e-07,
      "loss": 0.0002,
      "reward": 1.4270833656191826,
      "reward_std": 0.14563801139593124,
      "rewards/equation_reward_func": 0.450520847691223,
      "rewards/format_reward_func": 0.9765625111758709,
      "step": 214
    },
    {
      "completion_length": 346.88542652130127,
      "epoch": 0.1152,
      "grad_norm": 0.5877900931360449,
      "kl": 0.17950439453125,
      "learning_rate": 2.7875814660817504e-07,
      "loss": 0.0002,
      "reward": 1.4531250596046448,
      "reward_std": 0.2405162900686264,
      "rewards/equation_reward_func": 0.46875001583248377,
      "rewards/format_reward_func": 0.9843750149011612,
      "step": 216
    },
    {
      "completion_length": 359.08073711395264,
      "epoch": 0.11626666666666667,
      "grad_norm": 0.4948533563264289,
      "kl": 0.18475341796875,
      "learning_rate": 2.751764564986396e-07,
      "loss": 0.0002,
      "reward": 1.4088542088866234,
      "reward_std": 0.14888234762474895,
      "rewards/equation_reward_func": 0.42187500884756446,
      "rewards/format_reward_func": 0.986979179084301,
      "step": 218
    },
    {
      "completion_length": 319.13542795181274,
      "epoch": 0.11733333333333333,
      "grad_norm": 0.7214924505421153,
      "kl": 0.18463134765625,
      "learning_rate": 2.715895379284194e-07,
      "loss": 0.0002,
      "reward": 1.4869792014360428,
      "reward_std": 0.17930027330294251,
      "rewards/equation_reward_func": 0.5078125083819032,
      "rewards/format_reward_func": 0.979166679084301,
      "step": 220
    },
    {
      "completion_length": 347.49480152130127,
      "epoch": 0.1184,
      "grad_norm": 0.5439846803561813,
      "kl": 0.2037353515625,
      "learning_rate": 2.6799813580229174e-07,
      "loss": 0.0002,
      "reward": 1.4192708879709244,
      "reward_std": 0.1784167350269854,
      "rewards/equation_reward_func": 0.4401041753590107,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 222
    },
    {
      "completion_length": 317.0234432220459,
      "epoch": 0.11946666666666667,
      "grad_norm": 1.0377272313575363,
      "kl": 0.216064453125,
      "learning_rate": 2.6440299595614606e-07,
      "loss": 0.0002,
      "reward": 1.468750037252903,
      "reward_std": 0.2108682761900127,
      "rewards/equation_reward_func": 0.4791666774544865,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 224
    },
    {
      "completion_length": 304.04688358306885,
      "epoch": 0.12053333333333334,
      "grad_norm": 0.5374280295870404,
      "kl": 0.20159912109375,
      "learning_rate": 2.6080486500209347e-07,
      "loss": 0.0002,
      "reward": 1.546875037252903,
      "reward_std": 0.12545502791181207,
      "rewards/equation_reward_func": 0.5572916779201478,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 226
    },
    {
      "completion_length": 346.9557342529297,
      "epoch": 0.1216,
      "grad_norm": 0.6836413099426727,
      "kl": 0.1895751953125,
      "learning_rate": 2.572044901734166e-07,
      "loss": 0.0002,
      "reward": 1.3333333805203438,
      "reward_std": 0.1696605570614338,
      "rewards/equation_reward_func": 0.35156251420266926,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 228
    },
    {
      "completion_length": 322.09636402130127,
      "epoch": 0.12266666666666666,
      "grad_norm": 0.6445867346856486,
      "kl": 0.18853759765625,
      "learning_rate": 2.536026191693893e-07,
      "loss": 0.0002,
      "reward": 1.4661458693444729,
      "reward_std": 0.11827961774542928,
      "rewards/equation_reward_func": 0.47916668094694614,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 230
    },
    {
      "completion_length": 310.0989685058594,
      "epoch": 0.12373333333333333,
      "grad_norm": 0.543792761903507,
      "kl": 0.29705810546875,
      "learning_rate": 2.5e-07,
      "loss": 0.0003,
      "reward": 1.4505208656191826,
      "reward_std": 0.17755454638972878,
      "rewards/equation_reward_func": 0.46093750838190317,
      "rewards/format_reward_func": 0.9895833395421505,
      "step": 232
    },
    {
      "completion_length": 324.65625762939453,
      "epoch": 0.1248,
      "grad_norm": 0.6314276758218007,
      "kl": 0.2232666015625,
      "learning_rate": 2.4639738083061073e-07,
      "loss": 0.0002,
      "reward": 1.4557292237877846,
      "reward_std": 0.18806396471336484,
      "rewards/equation_reward_func": 0.4739583507180214,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 234
    },
    {
      "completion_length": 320.23178005218506,
      "epoch": 0.12586666666666665,
      "grad_norm": 0.7391960189787573,
      "kl": 0.20880126953125,
      "learning_rate": 2.4279550982658345e-07,
      "loss": 0.0002,
      "reward": 1.484375026077032,
      "reward_std": 0.2127929269336164,
      "rewards/equation_reward_func": 0.49739584140479565,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 236
    },
    {
      "completion_length": 305.5807361602783,
      "epoch": 0.12693333333333334,
      "grad_norm": 0.5628054512286292,
      "kl": 0.20721435546875,
      "learning_rate": 2.3919513499790646e-07,
      "loss": 0.0002,
      "reward": 1.4791667014360428,
      "reward_std": 0.1500206645578146,
      "rewards/equation_reward_func": 0.4947916797827929,
      "rewards/format_reward_func": 0.9843750149011612,
      "step": 238
    },
    {
      "completion_length": 361.0677194595337,
      "epoch": 0.128,
      "grad_norm": 0.6408903693134582,
      "kl": 0.22772216796875,
      "learning_rate": 2.3559700404385394e-07,
      "loss": 0.0002,
      "reward": 1.3489583693444729,
      "reward_std": 0.23990731267258525,
      "rewards/equation_reward_func": 0.38802084675990045,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 240
    },
    {
      "completion_length": 293.1484432220459,
      "epoch": 0.12906666666666666,
      "grad_norm": 0.5407067428590937,
      "kl": 0.22113037109375,
      "learning_rate": 2.3200186419770823e-07,
      "loss": 0.0002,
      "reward": 1.4843750521540642,
      "reward_std": 0.1898468122817576,
      "rewards/equation_reward_func": 0.5026041865348816,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 242
    },
    {
      "completion_length": 313.075532913208,
      "epoch": 0.13013333333333332,
      "grad_norm": 0.7079833699404031,
      "kl": 0.21197509765625,
      "learning_rate": 2.284104620715807e-07,
      "loss": 0.0002,
      "reward": 1.4192708730697632,
      "reward_std": 0.2272069714963436,
      "rewards/equation_reward_func": 0.43229167675599456,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 244
    },
    {
      "completion_length": 347.36980056762695,
      "epoch": 0.1312,
      "grad_norm": 0.8054821131616522,
      "kl": 0.2183837890625,
      "learning_rate": 2.2482354350136043e-07,
      "loss": 0.0002,
      "reward": 1.3359375409781933,
      "reward_std": 0.1773815001361072,
      "rewards/equation_reward_func": 0.36718751094304025,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 246
    },
    {
      "completion_length": 332.950532913208,
      "epoch": 0.13226666666666667,
      "grad_norm": 0.5333317411402013,
      "kl": 0.22491455078125,
      "learning_rate": 2.2124185339182496e-07,
      "loss": 0.0002,
      "reward": 1.3906250447034836,
      "reward_std": 0.23194079333916306,
      "rewards/equation_reward_func": 0.42447917722165585,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 248
    },
    {
      "completion_length": 342.0364685058594,
      "epoch": 0.13333333333333333,
      "grad_norm": 0.5916476351623432,
      "kl": 0.21429443359375,
      "learning_rate": 2.1766613556194344e-07,
      "loss": 0.0002,
      "reward": 1.3463542088866234,
      "reward_std": 0.20954444538801908,
      "rewards/equation_reward_func": 0.3906250074505806,
      "rewards/format_reward_func": 0.9557291902601719,
      "step": 250
    },
    {
      "completion_length": 333.47396659851074,
      "epoch": 0.1344,
      "grad_norm": 0.8337535734326338,
      "kl": 0.23358154296875,
      "learning_rate": 2.1409713259040628e-07,
      "loss": 0.0002,
      "reward": 1.3281250447034836,
      "reward_std": 0.23853017063811421,
      "rewards/equation_reward_func": 0.36979167931713164,
      "rewards/format_reward_func": 0.9583333507180214,
      "step": 252
    },
    {
      "completion_length": 302.9713649749756,
      "epoch": 0.13546666666666668,
      "grad_norm": 0.5975375572638757,
      "kl": 0.2205810546875,
      "learning_rate": 2.105355856614115e-07,
      "loss": 0.0002,
      "reward": 1.455729216337204,
      "reward_std": 0.19006694853305817,
      "rewards/equation_reward_func": 0.4817708460614085,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 254
    },
    {
      "completion_length": 295.7500066757202,
      "epoch": 0.13653333333333334,
      "grad_norm": 0.7360600957552589,
      "kl": 0.23883056640625,
      "learning_rate": 2.069822344107413e-07,
      "loss": 0.0002,
      "reward": 1.4895833730697632,
      "reward_std": 0.2519920407794416,
      "rewards/equation_reward_func": 0.5156250165309757,
      "rewards/format_reward_func": 0.9739583507180214,
      "step": 256
    },
    {
      "completion_length": 298.57552909851074,
      "epoch": 0.1376,
      "grad_norm": 0.6139982646016184,
      "kl": 0.2281494140625,
      "learning_rate": 2.034378167721599e-07,
      "loss": 0.0002,
      "reward": 1.4687500298023224,
      "reward_std": 0.20331065729260445,
      "rewards/equation_reward_func": 0.5000000093132257,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 258
    },
    {
      "completion_length": 267.87500762939453,
      "epoch": 0.13866666666666666,
      "grad_norm": 0.6087434332809726,
      "kl": 0.23504638671875,
      "learning_rate": 1.9990306882416485e-07,
      "loss": 0.0002,
      "reward": 1.5546875335276127,
      "reward_std": 0.19739143131300807,
      "rewards/equation_reward_func": 0.585937513737008,
      "rewards/format_reward_func": 0.9687500111758709,
      "step": 260
    },
    {
      "completion_length": 276.205735206604,
      "epoch": 0.13973333333333332,
      "grad_norm": 0.4844743573865835,
      "kl": 0.23095703125,
      "learning_rate": 1.9637872463712362e-07,
      "loss": 0.0002,
      "reward": 1.5026042088866234,
      "reward_std": 0.20278947753831744,
      "rewards/equation_reward_func": 0.5156250186264515,
      "rewards/format_reward_func": 0.986979179084301,
      "step": 262
    },
    {
      "completion_length": 317.5833406448364,
      "epoch": 0.1408,
      "grad_norm": 4.225936192922191,
      "kl": 0.6976318359375,
      "learning_rate": 1.9286551612082773e-07,
      "loss": 0.0007,
      "reward": 1.4218750409781933,
      "reward_std": 0.25739257177338004,
      "rewards/equation_reward_func": 0.4635416828095913,
      "rewards/format_reward_func": 0.9583333544433117,
      "step": 264
    },
    {
      "completion_length": 289.7031316757202,
      "epoch": 0.14186666666666667,
      "grad_norm": 0.7733885366987531,
      "kl": 0.23797607421875,
      "learning_rate": 1.8936417287249446e-07,
      "loss": 0.0002,
      "reward": 1.4427083730697632,
      "reward_std": 0.1840976132079959,
      "rewards/equation_reward_func": 0.4609375123400241,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 266
    },
    {
      "completion_length": 279.06511211395264,
      "epoch": 0.14293333333333333,
      "grad_norm": 0.7812632663890211,
      "kl": 0.24505615234375,
      "learning_rate": 1.8587542202524985e-07,
      "loss": 0.0002,
      "reward": 1.4140625521540642,
      "reward_std": 0.20719708874821663,
      "rewards/equation_reward_func": 0.4375000107102096,
      "rewards/format_reward_func": 0.9765625149011612,
      "step": 268
    },
    {
      "completion_length": 279.8958435058594,
      "epoch": 0.144,
      "grad_norm": 0.9200641519700825,
      "kl": 0.25006103515625,
      "learning_rate": 1.82399988097123e-07,
      "loss": 0.0003,
      "reward": 1.4479167126119137,
      "reward_std": 0.194319240283221,
      "rewards/equation_reward_func": 0.497395847691223,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 270
    },
    {
      "completion_length": 308.63542556762695,
      "epoch": 0.14506666666666668,
      "grad_norm": 0.6777891079331845,
      "kl": 0.246337890625,
      "learning_rate": 1.7893859284058378e-07,
      "loss": 0.0002,
      "reward": 1.398437537252903,
      "reward_std": 0.2058353153988719,
      "rewards/equation_reward_func": 0.432291675824672,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 272
    },
    {
      "completion_length": 363.75000953674316,
      "epoch": 0.14613333333333334,
      "grad_norm": 0.7035357358785052,
      "kl": 0.238525390625,
      "learning_rate": 1.7549195509265407e-07,
      "loss": 0.0002,
      "reward": 1.2578125484287739,
      "reward_std": 0.27548868488520384,
      "rewards/equation_reward_func": 0.32552084513008595,
      "rewards/format_reward_func": 0.9322916902601719,
      "step": 274
    },
    {
      "completion_length": 320.4010524749756,
      "epoch": 0.1472,
      "grad_norm": 0.5994918354091399,
      "kl": 0.24481201171875,
      "learning_rate": 1.7206079062562536e-07,
      "loss": 0.0002,
      "reward": 1.4062500447034836,
      "reward_std": 0.18554408056661487,
      "rewards/equation_reward_func": 0.4661458421032876,
      "rewards/format_reward_func": 0.9401041865348816,
      "step": 276
    },
    {
      "completion_length": 317.41667556762695,
      "epoch": 0.14826666666666666,
      "grad_norm": 1.0621443788066243,
      "kl": 0.2325439453125,
      "learning_rate": 1.6864581199841226e-07,
      "loss": 0.0002,
      "reward": 1.3697916977107525,
      "reward_std": 0.30007801530882716,
      "rewards/equation_reward_func": 0.42447918374091387,
      "rewards/format_reward_func": 0.9453125149011612,
      "step": 278
    },
    {
      "completion_length": 293.86198806762695,
      "epoch": 0.14933333333333335,
      "grad_norm": 0.6658308659860376,
      "kl": 0.23321533203125,
      "learning_rate": 1.6524772840857388e-07,
      "loss": 0.0002,
      "reward": 1.471354216337204,
      "reward_std": 0.14693401101976633,
      "rewards/equation_reward_func": 0.5052083488553762,
      "rewards/format_reward_func": 0.9661458432674408,
      "step": 280
    },
    {
      "completion_length": 324.47657203674316,
      "epoch": 0.1504,
      "grad_norm": 0.749916869176833,
      "kl": 0.2568359375,
      "learning_rate": 1.6186724554503237e-07,
      "loss": 0.0003,
      "reward": 1.3463542088866234,
      "reward_std": 0.22985868388786912,
      "rewards/equation_reward_func": 0.4192708469927311,
      "rewards/format_reward_func": 0.9270833507180214,
      "step": 282
    },
    {
      "completion_length": 329.3437557220459,
      "epoch": 0.15146666666666667,
      "grad_norm": 0.9887835539251056,
      "kl": 0.2294921875,
      "learning_rate": 1.5850506544152103e-07,
      "loss": 0.0002,
      "reward": 1.3177083656191826,
      "reward_std": 0.26103097246959805,
      "rewards/equation_reward_func": 0.367187506519258,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 284
    },
    {
      "completion_length": 315.1927185058594,
      "epoch": 0.15253333333333333,
      "grad_norm": 0.8107267722492516,
      "kl": 0.2274169921875,
      "learning_rate": 1.5516188633079107e-07,
      "loss": 0.0002,
      "reward": 1.4088542088866234,
      "reward_std": 0.25638267770409584,
      "rewards/equation_reward_func": 0.4609375139698386,
      "rewards/format_reward_func": 0.9479166902601719,
      "step": 286
    },
    {
      "completion_length": 281.1354236602783,
      "epoch": 0.1536,
      "grad_norm": 0.5088155440303934,
      "kl": 0.25,
      "learning_rate": 1.5183840249960784e-07,
      "loss": 0.0003,
      "reward": 1.500000037252903,
      "reward_std": 0.2080402374267578,
      "rewards/equation_reward_func": 0.533854179084301,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 288
    },
    {
      "completion_length": 301.25521659851074,
      "epoch": 0.15466666666666667,
      "grad_norm": 0.625913529917873,
      "kl": 0.2359619140625,
      "learning_rate": 1.4853530414456612e-07,
      "loss": 0.0002,
      "reward": 1.4244792014360428,
      "reward_std": 0.26187680941075087,
      "rewards/equation_reward_func": 0.479166685603559,
      "rewards/format_reward_func": 0.9453125149011612,
      "step": 290
    },
    {
      "completion_length": 311.2500123977661,
      "epoch": 0.15573333333333333,
      "grad_norm": 0.6304718791189173,
      "kl": 0.21844482421875,
      "learning_rate": 1.4525327722875568e-07,
      "loss": 0.0002,
      "reward": 1.4401041977107525,
      "reward_std": 0.25729640712961555,
      "rewards/equation_reward_func": 0.49218750977888703,
      "rewards/format_reward_func": 0.9479166865348816,
      "step": 292
    },
    {
      "completion_length": 305.377610206604,
      "epoch": 0.1568,
      "grad_norm": 0.7466996789785751,
      "kl": 0.2376708984375,
      "learning_rate": 1.4199300333930515e-07,
      "loss": 0.0002,
      "reward": 1.4270833507180214,
      "reward_std": 0.24685111083090305,
      "rewards/equation_reward_func": 0.4791666797827929,
      "rewards/format_reward_func": 0.9479166865348816,
      "step": 294
    },
    {
      "completion_length": 277.5703225135803,
      "epoch": 0.15786666666666666,
      "grad_norm": 1.3431386452828957,
      "kl": 0.23419189453125,
      "learning_rate": 1.3875515954583523e-07,
      "loss": 0.0002,
      "reward": 1.4583333730697632,
      "reward_std": 0.2398097929544747,
      "rewards/equation_reward_func": 0.49218751303851604,
      "rewards/format_reward_func": 0.9661458544433117,
      "step": 296
    },
    {
      "completion_length": 302.50001096725464,
      "epoch": 0.15893333333333334,
      "grad_norm": 0.7084022277977373,
      "kl": 0.24224853515625,
      "learning_rate": 1.3554041825985e-07,
      "loss": 0.0002,
      "reward": 1.3958333730697632,
      "reward_std": 0.21419357042759657,
      "rewards/equation_reward_func": 0.43229168094694614,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 298
    },
    {
      "completion_length": 298.9192838668823,
      "epoch": 0.16,
      "grad_norm": 0.6330236043077112,
      "kl": 0.236572265625,
      "learning_rate": 1.323494470950949e-07,
      "loss": 0.0002,
      "reward": 1.4375000335276127,
      "reward_std": 0.19589123968034983,
      "rewards/equation_reward_func": 0.4765625116415322,
      "rewards/format_reward_func": 0.9609375149011612,
      "step": 300
    },
    {
      "completion_length": 304.861985206604,
      "epoch": 0.16106666666666666,
      "grad_norm": 0.8282349724126732,
      "kl": 0.2291259765625,
      "learning_rate": 1.2918290872891236e-07,
      "loss": 0.0002,
      "reward": 1.3671875447034836,
      "reward_std": 0.19775044033303857,
      "rewards/equation_reward_func": 0.416666679084301,
      "rewards/format_reward_func": 0.9505208544433117,
      "step": 302
    },
    {
      "completion_length": 310.7447986602783,
      "epoch": 0.16213333333333332,
      "grad_norm": 0.6732535657390017,
      "kl": 0.258056640625,
      "learning_rate": 1.260414607646213e-07,
      "loss": 0.0003,
      "reward": 1.359375026077032,
      "reward_std": 0.1770839923992753,
      "rewards/equation_reward_func": 0.4062500074505806,
      "rewards/format_reward_func": 0.9531250111758709,
      "step": 304
    },
    {
      "completion_length": 273.6614646911621,
      "epoch": 0.1632,
      "grad_norm": 0.9183577183732234,
      "kl": 0.23602294921875,
      "learning_rate": 1.2292575559495143e-07,
      "loss": 0.0002,
      "reward": 1.4166667014360428,
      "reward_std": 0.18182098167017102,
      "rewards/equation_reward_func": 0.4453125111758709,
      "rewards/format_reward_func": 0.9713541753590107,
      "step": 306
    },
    {
      "completion_length": 247.5677146911621,
      "epoch": 0.16426666666666667,
      "grad_norm": 0.6700271556876387,
      "kl": 0.2451171875,
      "learning_rate": 1.1983644026655835e-07,
      "loss": 0.0002,
      "reward": 1.5364583805203438,
      "reward_std": 0.16835713339969516,
      "rewards/equation_reward_func": 0.5468750132713467,
      "rewards/format_reward_func": 0.9895833395421505,
      "step": 308
    },
    {
      "completion_length": 296.3489637374878,
      "epoch": 0.16533333333333333,
      "grad_norm": 0.7335847558655371,
      "kl": 0.22369384765625,
      "learning_rate": 1.1677415634565066e-07,
      "loss": 0.0002,
      "reward": 1.4244792014360428,
      "reward_std": 0.21954377181828022,
      "rewards/equation_reward_func": 0.4739583469927311,
      "rewards/format_reward_func": 0.9505208469927311,
      "step": 310
    },
    {
      "completion_length": 294.5338611602783,
      "epoch": 0.1664,
      "grad_norm": 0.566093567716464,
      "kl": 0.26129150390625,
      "learning_rate": 1.1373953978475353e-07,
      "loss": 0.0003,
      "reward": 1.3828125447034836,
      "reward_std": 0.15335895912721753,
      "rewards/equation_reward_func": 0.42968750884756446,
      "rewards/format_reward_func": 0.9531250111758709,
      "step": 312
    },
    {
      "completion_length": 319.2760543823242,
      "epoch": 0.16746666666666668,
      "grad_norm": 0.7230399862261075,
      "kl": 0.2347412109375,
      "learning_rate": 1.1073322079063913e-07,
      "loss": 0.0002,
      "reward": 1.2890625335276127,
      "reward_std": 0.1907347133383155,
      "rewards/equation_reward_func": 0.3281250062864274,
      "rewards/format_reward_func": 0.9609375149011612,
      "step": 314
    },
    {
      "completion_length": 277.4401111602783,
      "epoch": 0.16853333333333334,
      "grad_norm": 0.7925402861184135,
      "kl": 0.228271484375,
      "learning_rate": 1.0775582369344946e-07,
      "loss": 0.0002,
      "reward": 1.4869792237877846,
      "reward_std": 0.1919974870979786,
      "rewards/equation_reward_func": 0.5182291828095913,
      "rewards/format_reward_func": 0.9687500111758709,
      "step": 316
    },
    {
      "completion_length": 314.05209255218506,
      "epoch": 0.1696,
      "grad_norm": 0.7822619678777167,
      "kl": 0.21319580078125,
      "learning_rate": 1.0480796681704077e-07,
      "loss": 0.0002,
      "reward": 1.4088542126119137,
      "reward_std": 0.21224576281383634,
      "rewards/equation_reward_func": 0.4375000149011612,
      "rewards/format_reward_func": 0.9713541828095913,
      "step": 318
    },
    {
      "completion_length": 263.19011211395264,
      "epoch": 0.17066666666666666,
      "grad_norm": 0.6210402052679087,
      "kl": 0.23016357421875,
      "learning_rate": 1.018902623505741e-07,
      "loss": 0.0002,
      "reward": 1.4791667088866234,
      "reward_std": 0.2531423852778971,
      "rewards/equation_reward_func": 0.5208333446644247,
      "rewards/format_reward_func": 0.9583333507180214,
      "step": 320
    },
    {
      "completion_length": 287.1484489440918,
      "epoch": 0.17173333333333332,
      "grad_norm": 0.634776746244374,
      "kl": 0.2230224609375,
      "learning_rate": 9.900331622138063e-08,
      "loss": 0.0002,
      "reward": 1.4348958730697632,
      "reward_std": 0.15947270533069968,
      "rewards/equation_reward_func": 0.4557291779201478,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 322
    },
    {
      "completion_length": 284.5286521911621,
      "epoch": 0.1728,
      "grad_norm": 0.8088754769939193,
      "kl": 0.21588134765625,
      "learning_rate": 9.614772796912681e-08,
      "loss": 0.0002,
      "reward": 1.4348958656191826,
      "reward_std": 0.23451161291450262,
      "rewards/equation_reward_func": 0.4791666802484542,
      "rewards/format_reward_func": 0.9557291828095913,
      "step": 324
    },
    {
      "completion_length": 297.3776135444641,
      "epoch": 0.17386666666666667,
      "grad_norm": 0.9105446106335183,
      "kl": 0.22491455078125,
      "learning_rate": 9.332409062130686e-08,
      "loss": 0.0002,
      "reward": 1.3177083805203438,
      "reward_std": 0.1509189954958856,
      "rewards/equation_reward_func": 0.3515625116415322,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 326
    },
    {
      "completion_length": 298.77084398269653,
      "epoch": 0.17493333333333333,
      "grad_norm": 0.7126661333306704,
      "kl": 0.22003173828125,
      "learning_rate": 9.053299057008699e-08,
      "loss": 0.0002,
      "reward": 1.3958333730697632,
      "reward_std": 0.15720906713977456,
      "rewards/equation_reward_func": 0.4401041779201478,
      "rewards/format_reward_func": 0.9557291865348816,
      "step": 328
    },
    {
      "completion_length": 308.93490505218506,
      "epoch": 0.176,
      "grad_norm": 0.7401905494769911,
      "kl": 0.2979736328125,
      "learning_rate": 8.777500745052743e-08,
      "loss": 0.0003,
      "reward": 1.3385416977107525,
      "reward_std": 0.17720176372677088,
      "rewards/equation_reward_func": 0.3750000111758709,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 330
    },
    {
      "completion_length": 315.75782012939453,
      "epoch": 0.17706666666666668,
      "grad_norm": 1.0292481385344092,
      "kl": 0.2142333984375,
      "learning_rate": 8.505071402020892e-08,
      "loss": 0.0002,
      "reward": 1.3567708730697632,
      "reward_std": 0.16194288805127144,
      "rewards/equation_reward_func": 0.3854166786186397,
      "rewards/format_reward_func": 0.9713541828095913,
      "step": 332
    },
    {
      "completion_length": 261.085946559906,
      "epoch": 0.17813333333333334,
      "grad_norm": 0.6313338932442103,
      "kl": 0.27130126953125,
      "learning_rate": 8.236067604028562e-08,
      "loss": 0.0003,
      "reward": 1.5260417088866234,
      "reward_std": 0.19619298679754138,
      "rewards/equation_reward_func": 0.5703125167638063,
      "rewards/format_reward_func": 0.955729179084301,
      "step": 334
    },
    {
      "completion_length": 299.28125858306885,
      "epoch": 0.1792,
      "grad_norm": 0.767319955612925,
      "kl": 0.2216796875,
      "learning_rate": 7.970545215799327e-08,
      "loss": 0.0002,
      "reward": 1.3984375521540642,
      "reward_std": 0.1856266581453383,
      "rewards/equation_reward_func": 0.43489584513008595,
      "rewards/format_reward_func": 0.9635416865348816,
      "step": 336
    },
    {
      "completion_length": 289.1171979904175,
      "epoch": 0.18026666666666666,
      "grad_norm": 0.5279846258585172,
      "kl": 0.22845458984375,
      "learning_rate": 7.708559379063204e-08,
      "loss": 0.0002,
      "reward": 1.4010417088866234,
      "reward_std": 0.19252674980089068,
      "rewards/equation_reward_func": 0.432291679084301,
      "rewards/format_reward_func": 0.9687500186264515,
      "step": 338
    },
    {
      "completion_length": 261.9843816757202,
      "epoch": 0.18133333333333335,
      "grad_norm": 1.0110560575317675,
      "kl": 0.392333984375,
      "learning_rate": 7.45016450110534e-08,
      "loss": 0.0004,
      "reward": 1.4687500447034836,
      "reward_std": 0.16133547388017178,
      "rewards/equation_reward_func": 0.5000000167638063,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 340
    },
    {
      "completion_length": 264.8333406448364,
      "epoch": 0.1824,
      "grad_norm": 0.8570941539149842,
      "kl": 0.22308349609375,
      "learning_rate": 7.195414243467029e-08,
      "loss": 0.0002,
      "reward": 1.4583333805203438,
      "reward_std": 0.13745049014687538,
      "rewards/equation_reward_func": 0.4687500139698386,
      "rewards/format_reward_func": 0.9895833395421505,
      "step": 342
    },
    {
      "completion_length": 230.47396659851074,
      "epoch": 0.18346666666666667,
      "grad_norm": 0.7319616485107195,
      "kl": 0.24261474609375,
      "learning_rate": 6.944361510801763e-08,
      "loss": 0.0002,
      "reward": 1.6041667237877846,
      "reward_std": 0.09940882679075003,
      "rewards/equation_reward_func": 0.6197916846722364,
      "rewards/format_reward_func": 0.9843750074505806,
      "step": 344
    },
    {
      "completion_length": 325.17448902130127,
      "epoch": 0.18453333333333333,
      "grad_norm": 0.8899948638554169,
      "kl": 0.22564697265625,
      "learning_rate": 6.697058439888283e-08,
      "loss": 0.0002,
      "reward": 1.2890625335276127,
      "reward_std": 0.17243655677884817,
      "rewards/equation_reward_func": 0.32552084047347307,
      "rewards/format_reward_func": 0.9635416902601719,
      "step": 346
    },
    {
      "completion_length": 277.57032012939453,
      "epoch": 0.1856,
      "grad_norm": 0.9596879937723287,
      "kl": 0.23382568359375,
      "learning_rate": 6.453556388803288e-08,
      "loss": 0.0002,
      "reward": 1.4322917237877846,
      "reward_std": 0.2332175257615745,
      "rewards/equation_reward_func": 0.46875001303851604,
      "rewards/format_reward_func": 0.9635416865348816,
      "step": 348
    },
    {
      "completion_length": 247.51823616027832,
      "epoch": 0.18666666666666668,
      "grad_norm": 0.7147233096894462,
      "kl": 0.23199462890625,
      "learning_rate": 6.213905926255697e-08,
      "loss": 0.0002,
      "reward": 1.572916716337204,
      "reward_std": 0.13177413679659367,
      "rewards/equation_reward_func": 0.5885416772216558,
      "rewards/format_reward_func": 0.9843750037252903,
      "step": 350
    },
    {
      "completion_length": 254.29688453674316,
      "epoch": 0.18773333333333334,
      "grad_norm": 0.5156264024880255,
      "kl": 0.2138671875,
      "learning_rate": 5.978156821084987e-08,
      "loss": 0.0002,
      "reward": 1.5338542014360428,
      "reward_std": 0.09953805012628436,
      "rewards/equation_reward_func": 0.570312513038516,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 352
    },
    {
      "completion_length": 306.208345413208,
      "epoch": 0.1888,
      "grad_norm": 0.7349538500183934,
      "kl": 0.22650146484375,
      "learning_rate": 5.7463580319254853e-08,
      "loss": 0.0002,
      "reward": 1.2916666977107525,
      "reward_std": 0.23371722968295217,
      "rewards/equation_reward_func": 0.34375000814907253,
      "rewards/format_reward_func": 0.9479166939854622,
      "step": 354
    },
    {
      "completion_length": 250.97396755218506,
      "epoch": 0.18986666666666666,
      "grad_norm": 0.9539899909075816,
      "kl": 0.2208251953125,
      "learning_rate": 5.518557697039081e-08,
      "loss": 0.0002,
      "reward": 1.5442708805203438,
      "reward_std": 0.12091149017214775,
      "rewards/equation_reward_func": 0.5598958432674408,
      "rewards/format_reward_func": 0.9843750111758709,
      "step": 356
    },
    {
      "completion_length": 247.60937976837158,
      "epoch": 0.19093333333333334,
      "grad_norm": 0.5833784789045885,
      "kl": 0.24102783203125,
      "learning_rate": 5.294803124318145e-08,
      "loss": 0.0002,
      "reward": 1.520833395421505,
      "reward_std": 0.08653856860473752,
      "rewards/equation_reward_func": 0.5364583507180214,
      "rewards/format_reward_func": 0.9843750111758709,
      "step": 358
    },
    {
      "completion_length": 277.3724021911621,
      "epoch": 0.192,
      "grad_norm": 0.7440919143169152,
      "kl": 0.2235107421875,
      "learning_rate": 5.07514078146106e-08,
      "loss": 0.0002,
      "reward": 1.476562537252903,
      "reward_std": 0.14440069487318397,
      "rewards/equation_reward_func": 0.5000000121071935,
      "rewards/format_reward_func": 0.9765625149011612,
      "step": 360
    },
    {
      "completion_length": 226.549485206604,
      "epoch": 0.19306666666666666,
      "grad_norm": 0.6991350685338126,
      "kl": 0.2354736328125,
      "learning_rate": 4.859616286322094e-08,
      "loss": 0.0002,
      "reward": 1.6692708656191826,
      "reward_std": 0.1841339054517448,
      "rewards/equation_reward_func": 0.6875000204890966,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 362
    },
    {
      "completion_length": 257.7474002838135,
      "epoch": 0.19413333333333332,
      "grad_norm": 0.6684660994634332,
      "kl": 0.22576904296875,
      "learning_rate": 4.648274397437829e-08,
      "loss": 0.0002,
      "reward": 1.4817708656191826,
      "reward_std": 0.14901333069428802,
      "rewards/equation_reward_func": 0.49739584466442466,
      "rewards/format_reward_func": 0.9843750074505806,
      "step": 364
    },
    {
      "completion_length": 247.213547706604,
      "epoch": 0.1952,
      "grad_norm": 1.3498805283474344,
      "kl": 0.2373046875,
      "learning_rate": 4.4411590047320617e-08,
      "loss": 0.0002,
      "reward": 1.5416667088866234,
      "reward_std": 0.20480258110910654,
      "rewards/equation_reward_func": 0.5677083497866988,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 366
    },
    {
      "completion_length": 271.8333387374878,
      "epoch": 0.19626666666666667,
      "grad_norm": 0.4853801837605491,
      "kl": 0.2344970703125,
      "learning_rate": 4.2383131204010494e-08,
      "loss": 0.0002,
      "reward": 1.4765625223517418,
      "reward_std": 0.13479887740686536,
      "rewards/equation_reward_func": 0.5052083446644247,
      "rewards/format_reward_func": 0.9713541753590107,
      "step": 368
    },
    {
      "completion_length": 304.4739694595337,
      "epoch": 0.19733333333333333,
      "grad_norm": 0.6611011259992957,
      "kl": 0.22882080078125,
      "learning_rate": 4.039778869981064e-08,
      "loss": 0.0002,
      "reward": 1.3151042014360428,
      "reward_std": 0.2048181821592152,
      "rewards/equation_reward_func": 0.3567708428017795,
      "rewards/format_reward_func": 0.9583333432674408,
      "step": 370
    },
    {
      "completion_length": 256.5833406448364,
      "epoch": 0.1984,
      "grad_norm": 0.7998496404585249,
      "kl": 0.22052001953125,
      "learning_rate": 3.845597483600049e-08,
      "loss": 0.0002,
      "reward": 1.5052083767950535,
      "reward_std": 0.17007605405524373,
      "rewards/equation_reward_func": 0.5260416804812849,
      "rewards/format_reward_func": 0.9791666716337204,
      "step": 372
    },
    {
      "completion_length": 288.625009059906,
      "epoch": 0.19946666666666665,
      "grad_norm": 0.6923241167833261,
      "kl": 0.228271484375,
      "learning_rate": 3.655809287415284e-08,
      "loss": 0.0002,
      "reward": 1.4192708730697632,
      "reward_std": 0.18929289653897285,
      "rewards/equation_reward_func": 0.4531250118743628,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 374
    },
    {
      "completion_length": 299.89323329925537,
      "epoch": 0.20053333333333334,
      "grad_norm": 0.8024434586179425,
      "kl": 0.21966552734375,
      "learning_rate": 3.4704536952387285e-08,
      "loss": 0.0002,
      "reward": 1.341145858168602,
      "reward_std": 0.2507268921472132,
      "rewards/equation_reward_func": 0.3828125037252903,
      "rewards/format_reward_func": 0.9583333544433117,
      "step": 376
    },
    {
      "completion_length": 293.13542556762695,
      "epoch": 0.2016,
      "grad_norm": 0.7670744626957942,
      "kl": 0.23614501953125,
      "learning_rate": 3.2895692003518575e-08,
      "loss": 0.0002,
      "reward": 1.3541666977107525,
      "reward_std": 0.26297286711633205,
      "rewards/equation_reward_func": 0.40364584466442466,
      "rewards/format_reward_func": 0.9505208544433117,
      "step": 378
    },
    {
      "completion_length": 267.18490409851074,
      "epoch": 0.20266666666666666,
      "grad_norm": 18.75450521285186,
      "kl": 0.29046630859375,
      "learning_rate": 3.113193367511635e-08,
      "loss": 0.0003,
      "reward": 1.468750037252903,
      "reward_std": 0.20902110496535897,
      "rewards/equation_reward_func": 0.5026041809469461,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 380
    },
    {
      "completion_length": 287.71354961395264,
      "epoch": 0.20373333333333332,
      "grad_norm": 0.8438907529414764,
      "kl": 0.236083984375,
      "learning_rate": 2.9413628251493934e-08,
      "loss": 0.0002,
      "reward": 1.335937537252903,
      "reward_std": 0.18686132272705436,
      "rewards/equation_reward_func": 0.37760417233221233,
      "rewards/format_reward_func": 0.9583333469927311,
      "step": 382
    },
    {
      "completion_length": 272.40365409851074,
      "epoch": 0.2048,
      "grad_norm": 1.0640930657692547,
      "kl": 0.232177734375,
      "learning_rate": 2.774113257764066e-08,
      "loss": 0.0002,
      "reward": 1.468750037252903,
      "reward_std": 0.259894413407892,
      "rewards/equation_reward_func": 0.49739584885537624,
      "rewards/format_reward_func": 0.9713541902601719,
      "step": 384
    },
    {
      "completion_length": 294.9479236602783,
      "epoch": 0.20586666666666667,
      "grad_norm": 0.5559880495074878,
      "kl": 0.2174072265625,
      "learning_rate": 2.611479398511518e-08,
      "loss": 0.0002,
      "reward": 1.3619792088866234,
      "reward_std": 0.11354415956884623,
      "rewards/equation_reward_func": 0.3854166741948575,
      "rewards/format_reward_func": 0.9765625111758709,
      "step": 386
    },
    {
      "completion_length": 283.9375123977661,
      "epoch": 0.20693333333333333,
      "grad_norm": 0.6076443550251356,
      "kl": 0.22991943359375,
      "learning_rate": 2.4534950219914057e-08,
      "loss": 0.0002,
      "reward": 1.382812537252903,
      "reward_std": 0.14025551825761795,
      "rewards/equation_reward_func": 0.408854179084301,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 388
    },
    {
      "completion_length": 279.5729293823242,
      "epoch": 0.208,
      "grad_norm": 0.8924913049812959,
      "kl": 0.22430419921875,
      "learning_rate": 2.300192937233128e-08,
      "loss": 0.0002,
      "reward": 1.4427083656191826,
      "reward_std": 0.24587285611778498,
      "rewards/equation_reward_func": 0.47135418094694614,
      "rewards/format_reward_func": 0.9713541865348816,
      "step": 390
    },
    {
      "completion_length": 260.111985206604,
      "epoch": 0.20906666666666668,
      "grad_norm": 0.6858880533326718,
      "kl": 0.298583984375,
      "learning_rate": 2.1516049808822935e-08,
      "loss": 0.0003,
      "reward": 1.5052083730697632,
      "reward_std": 0.1241533849388361,
      "rewards/equation_reward_func": 0.5182291865348816,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 392
    },
    {
      "completion_length": 266.47657012939453,
      "epoch": 0.21013333333333334,
      "grad_norm": 0.6402109636411236,
      "kl": 0.2220458984375,
      "learning_rate": 2.007762010589098e-08,
      "loss": 0.0002,
      "reward": 1.4375000447034836,
      "reward_std": 0.20047930860891938,
      "rewards/equation_reward_func": 0.4765625181607902,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 394
    },
    {
      "completion_length": 270.7604217529297,
      "epoch": 0.2112,
      "grad_norm": 0.906287924542971,
      "kl": 0.2283935546875,
      "learning_rate": 1.8686938986000627e-08,
      "loss": 0.0002,
      "reward": 1.502604205161333,
      "reward_std": 0.13055545324459672,
      "rewards/equation_reward_func": 0.5208333441987634,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 396
    },
    {
      "completion_length": 296.5468854904175,
      "epoch": 0.21226666666666666,
      "grad_norm": 0.7760203626767087,
      "kl": 0.24676513671875,
      "learning_rate": 1.734429525554365e-08,
      "loss": 0.0002,
      "reward": 1.3385417088866234,
      "reward_std": 0.18930128030478954,
      "rewards/equation_reward_func": 0.36197917931713164,
      "rewards/format_reward_func": 0.9765625223517418,
      "step": 398
    },
    {
      "completion_length": 272.8854250907898,
      "epoch": 0.21333333333333335,
      "grad_norm": 1.5742500055580204,
      "kl": 0.24517822265625,
      "learning_rate": 1.604996774486145e-08,
      "loss": 0.0002,
      "reward": 1.4661458656191826,
      "reward_std": 0.20536453556269407,
      "rewards/equation_reward_func": 0.5052083469927311,
      "rewards/format_reward_func": 0.9609375111758709,
      "step": 400
    },
    {
      "completion_length": 267.20052909851074,
      "epoch": 0.2144,
      "grad_norm": 0.6444833598398121,
      "kl": 0.29022216796875,
      "learning_rate": 1.4804225250339281e-08,
      "loss": 0.0003,
      "reward": 1.4557292126119137,
      "reward_std": 0.20770712802186608,
      "rewards/equation_reward_func": 0.48958334466442466,
      "rewards/format_reward_func": 0.9661458395421505,
      "step": 402
    },
    {
      "completion_length": 300.9765691757202,
      "epoch": 0.21546666666666667,
      "grad_norm": 1.4114979925483315,
      "kl": 0.25653076171875,
      "learning_rate": 1.360732647858498e-08,
      "loss": 0.0003,
      "reward": 1.3906250596046448,
      "reward_std": 0.22731021884828806,
      "rewards/equation_reward_func": 0.4401041760575026,
      "rewards/format_reward_func": 0.950520858168602,
      "step": 404
    },
    {
      "completion_length": 282.7500057220459,
      "epoch": 0.21653333333333333,
      "grad_norm": 0.6696240626293803,
      "kl": 0.22821044921875,
      "learning_rate": 1.2459519992702311e-08,
      "loss": 0.0002,
      "reward": 1.4453125298023224,
      "reward_std": 0.20481601962819695,
      "rewards/equation_reward_func": 0.4661458432674408,
      "rewards/format_reward_func": 0.9791666753590107,
      "step": 406
    },
    {
      "completion_length": 280.0234456062317,
      "epoch": 0.2176,
      "grad_norm": 0.6398611308559539,
      "kl": 0.2344970703125,
      "learning_rate": 1.1361044160671629e-08,
      "loss": 0.0002,
      "reward": 1.4114583656191826,
      "reward_std": 0.16066451743245125,
      "rewards/equation_reward_func": 0.4322916760575026,
      "rewards/format_reward_func": 0.979166679084301,
      "step": 408
    },
    {
      "completion_length": 283.92448902130127,
      "epoch": 0.21866666666666668,
      "grad_norm": 0.7894317667037624,
      "kl": 0.2291259765625,
      "learning_rate": 1.0312127105846947e-08,
      "loss": 0.0002,
      "reward": 1.4583333730697632,
      "reward_std": 0.1464251554571092,
      "rewards/equation_reward_func": 0.48958334466442466,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 410
    },
    {
      "completion_length": 269.15105152130127,
      "epoch": 0.21973333333333334,
      "grad_norm": 0.6521973364777306,
      "kl": 0.2493896484375,
      "learning_rate": 9.312986659581301e-09,
      "loss": 0.0002,
      "reward": 1.4218750409781933,
      "reward_std": 0.20944805443286896,
      "rewards/equation_reward_func": 0.4583333421032876,
      "rewards/format_reward_func": 0.9635416902601719,
      "step": 412
    },
    {
      "completion_length": 303.3698024749756,
      "epoch": 0.2208,
      "grad_norm": 0.928412007143324,
      "kl": 0.22808837890625,
      "learning_rate": 8.363830315988945e-09,
      "loss": 0.0002,
      "reward": 1.3854167088866234,
      "reward_std": 0.22633069939911366,
      "rewards/equation_reward_func": 0.4140625100117177,
      "rewards/format_reward_func": 0.9713541828095913,
      "step": 414
    },
    {
      "completion_length": 279.8463611602783,
      "epoch": 0.22186666666666666,
      "grad_norm": 0.6471319326937831,
      "kl": 0.2322998046875,
      "learning_rate": 7.46485518885462e-09,
      "loss": 0.0002,
      "reward": 1.393229216337204,
      "reward_std": 0.19153147842735052,
      "rewards/equation_reward_func": 0.4140625111758709,
      "rewards/format_reward_func": 0.9791666753590107,
      "step": 416
    },
    {
      "completion_length": 286.213547706604,
      "epoch": 0.22293333333333334,
      "grad_norm": 1.030967274909426,
      "kl": 0.22314453125,
      "learning_rate": 6.616247970698319e-09,
      "loss": 0.0002,
      "reward": 1.4739583805203438,
      "reward_std": 0.18684497009962797,
      "rewards/equation_reward_func": 0.5104166779201478,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 418
    },
    {
      "completion_length": 268.1250114440918,
      "epoch": 0.224,
      "grad_norm": 0.687917802429927,
      "kl": 0.2392578125,
      "learning_rate": 5.8181848940044855e-09,
      "loss": 0.0002,
      "reward": 1.4583333730697632,
      "reward_std": 0.19611964747309685,
      "rewards/equation_reward_func": 0.4791666818782687,
      "rewards/format_reward_func": 0.979166679084301,
      "step": 420
    },
    {
      "completion_length": 286.42188358306885,
      "epoch": 0.22506666666666666,
      "grad_norm": 1.003000954678048,
      "kl": 0.24884033203125,
      "learning_rate": 5.070831694623135e-09,
      "loss": 0.0002,
      "reward": 1.4244792014360428,
      "reward_std": 0.24217658629640937,
      "rewards/equation_reward_func": 0.45833334093913436,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 422
    },
    {
      "completion_length": 279.4114656448364,
      "epoch": 0.22613333333333333,
      "grad_norm": 0.7365913580715772,
      "kl": 0.22650146484375,
      "learning_rate": 4.374343577351336e-09,
      "loss": 0.0002,
      "reward": 1.3828125521540642,
      "reward_std": 0.1807353887706995,
      "rewards/equation_reward_func": 0.40625000768341124,
      "rewards/format_reward_func": 0.9765625186264515,
      "step": 424
    },
    {
      "completion_length": 285.20312881469727,
      "epoch": 0.2272,
      "grad_norm": 1.1325349820394892,
      "kl": 0.2252197265625,
      "learning_rate": 3.7288651837012745e-09,
      "loss": 0.0002,
      "reward": 1.4322917088866234,
      "reward_std": 0.24207678250968456,
      "rewards/equation_reward_func": 0.45572918676771224,
      "rewards/format_reward_func": 0.9765625186264515,
      "step": 426
    },
    {
      "completion_length": 295.6354269981384,
      "epoch": 0.22826666666666667,
      "grad_norm": 0.6262258188992635,
      "kl": 0.224609375,
      "learning_rate": 3.134530561862081e-09,
      "loss": 0.0002,
      "reward": 1.4192708656191826,
      "reward_std": 0.21014545997604728,
      "rewards/equation_reward_func": 0.4583333386108279,
      "rewards/format_reward_func": 0.9609375149011612,
      "step": 428
    },
    {
      "completion_length": 295.90625858306885,
      "epoch": 0.22933333333333333,
      "grad_norm": 0.6142531023657672,
      "kl": 0.22576904296875,
      "learning_rate": 2.5914631388619103e-09,
      "loss": 0.0002,
      "reward": 1.3385417088866234,
      "reward_std": 0.16080457624047995,
      "rewards/equation_reward_func": 0.3802083416376263,
      "rewards/format_reward_func": 0.9583333507180214,
      "step": 430
    },
    {
      "completion_length": 263.97396755218506,
      "epoch": 0.2304,
      "grad_norm": 0.6976796846300636,
      "kl": 0.2528076171875,
      "learning_rate": 2.0997756949353297e-09,
      "loss": 0.0003,
      "reward": 1.4322917014360428,
      "reward_std": 0.18798484792932868,
      "rewards/equation_reward_func": 0.46093751094304025,
      "rewards/format_reward_func": 0.971354179084301,
      "step": 432
    },
    {
      "completion_length": 238.1224021911621,
      "epoch": 0.23146666666666665,
      "grad_norm": 0.899293286345609,
      "kl": 0.23345947265625,
      "learning_rate": 1.6595703401020844e-09,
      "loss": 0.0002,
      "reward": 1.4895833730697632,
      "reward_std": 0.16196552151814103,
      "rewards/equation_reward_func": 0.5130208421032876,
      "rewards/format_reward_func": 0.9765625111758709,
      "step": 434
    },
    {
      "completion_length": 300.72136402130127,
      "epoch": 0.23253333333333334,
      "grad_norm": 0.8491958193033893,
      "kl": 0.214111328125,
      "learning_rate": 1.2709384929615596e-09,
      "loss": 0.0002,
      "reward": 1.3567708730697632,
      "reward_std": 0.1968450741842389,
      "rewards/equation_reward_func": 0.3906250107102096,
      "rewards/format_reward_func": 0.9661458544433117,
      "step": 436
    },
    {
      "completion_length": 281.16667556762695,
      "epoch": 0.2336,
      "grad_norm": 0.5991014285626669,
      "kl": 0.23944091796875,
      "learning_rate": 9.339608617077165e-10,
      "loss": 0.0002,
      "reward": 1.3958333656191826,
      "reward_std": 0.19733294053003192,
      "rewards/equation_reward_func": 0.4401041781529784,
      "rewards/format_reward_func": 0.9557291865348816,
      "step": 438
    },
    {
      "completion_length": 271.0625057220459,
      "epoch": 0.23466666666666666,
      "grad_norm": 0.5794558183115414,
      "kl": 0.22308349609375,
      "learning_rate": 6.487074273681114e-10,
      "loss": 0.0002,
      "reward": 1.4661458656191826,
      "reward_std": 0.17582183191552758,
      "rewards/equation_reward_func": 0.4869791818782687,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 440
    },
    {
      "completion_length": 261.28386306762695,
      "epoch": 0.23573333333333332,
      "grad_norm": 0.7209816553467453,
      "kl": 0.248779296875,
      "learning_rate": 4.152374292708538e-10,
      "loss": 0.0002,
      "reward": 1.471354205161333,
      "reward_std": 0.16987955244258046,
      "rewards/equation_reward_func": 0.4973958469927311,
      "rewards/format_reward_func": 0.9739583432674408,
      "step": 442
    },
    {
      "completion_length": 231.79427671432495,
      "epoch": 0.2368,
      "grad_norm": 0.6551899921842496,
      "kl": 0.2474365234375,
      "learning_rate": 2.3359935274214204e-10,
      "loss": 0.0002,
      "reward": 1.5885417088866234,
      "reward_std": 0.18360807234421372,
      "rewards/equation_reward_func": 0.6223958495538682,
      "rewards/format_reward_func": 0.9661458544433117,
      "step": 444
    },
    {
      "completion_length": 284.0859432220459,
      "epoch": 0.23786666666666667,
      "grad_norm": 0.6724399152364189,
      "kl": 0.231201171875,
      "learning_rate": 1.0383091903720665e-10,
      "loss": 0.0002,
      "reward": 1.411458358168602,
      "reward_std": 0.22678064601495862,
      "rewards/equation_reward_func": 0.4401041748933494,
      "rewards/format_reward_func": 0.9713541828095913,
      "step": 446
    },
    {
      "completion_length": 276.44792556762695,
      "epoch": 0.23893333333333333,
      "grad_norm": 0.98764526989421,
      "kl": 0.33917236328125,
      "learning_rate": 2.595907750671533e-11,
      "loss": 0.0003,
      "reward": 1.4427083767950535,
      "reward_std": 0.17404404049739242,
      "rewards/equation_reward_func": 0.4661458453629166,
      "rewards/format_reward_func": 0.9765625149011612,
      "step": 448
    },
    {
      "completion_length": 276.0937604904175,
      "epoch": 0.24,
      "grad_norm": 0.844438465358864,
      "kl": 0.215087890625,
      "learning_rate": 0.0,
      "loss": 0.0002,
      "reward": 1.5312500298023224,
      "reward_std": 0.17535703862085938,
      "rewards/equation_reward_func": 0.5364583441987634,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 450
    },
    {
      "epoch": 0.24,
      "step": 450,
      "total_flos": 0.0,
      "train_loss": 0.0001681143713844217,
      "train_runtime": 19587.5481,
      "train_samples_per_second": 0.551,
      "train_steps_per_second": 0.023
    }
  ],
  "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
}