{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5599626691553896,
  "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": 502.7198944091797,
      "epoch": 0.002488722974023954,
      "grad_norm": 0.10407081592214139,
      "kl": 0.0,
      "learning_rate": 7.142857142857142e-08,
      "loss": -0.0,
      "reward": 0.30915180034935474,
      "reward_std": 0.45076136849820614,
      "rewards/equation_reward_func": 0.06473214633297175,
      "rewards/format_reward_func": 0.24441965389996767,
      "step": 2
    },
    {
      "completion_length": 493.8493537902832,
      "epoch": 0.004977445948047908,
      "grad_norm": 0.09927009002301451,
      "kl": 0.00039756298065185547,
      "learning_rate": 1.4285714285714285e-07,
      "loss": 0.0,
      "reward": 0.3638393059372902,
      "reward_std": 0.49667514115571976,
      "rewards/equation_reward_func": 0.06696428824216127,
      "rewards/format_reward_func": 0.2968750149011612,
      "step": 4
    },
    {
      "completion_length": 497.8616313934326,
      "epoch": 0.007466168922071862,
      "grad_norm": 0.093370113106332,
      "kl": 0.0003921985626220703,
      "learning_rate": 2.1428571428571426e-07,
      "loss": 0.0,
      "reward": 0.3147321604192257,
      "reward_std": 0.45942891016602516,
      "rewards/equation_reward_func": 0.06138393213041127,
      "rewards/format_reward_func": 0.25334822479635477,
      "step": 6
    },
    {
      "completion_length": 498.6730136871338,
      "epoch": 0.009954891896095816,
      "grad_norm": 0.087715911930183,
      "kl": 0.00040519237518310547,
      "learning_rate": 2.857142857142857e-07,
      "loss": 0.0,
      "reward": 0.3671875223517418,
      "reward_std": 0.510881919413805,
      "rewards/equation_reward_func": 0.07589286006987095,
      "rewards/format_reward_func": 0.2912946566939354,
      "step": 8
    },
    {
      "completion_length": 506.512300491333,
      "epoch": 0.01244361487011977,
      "grad_norm": 0.1032540463580205,
      "kl": 0.00044214725494384766,
      "learning_rate": 3.5714285714285716e-07,
      "loss": 0.0,
      "reward": 0.3303571604192257,
      "reward_std": 0.48546610213816166,
      "rewards/equation_reward_func": 0.06026785948779434,
      "rewards/format_reward_func": 0.2700892947614193,
      "step": 10
    },
    {
      "completion_length": 501.3951072692871,
      "epoch": 0.014932337844143724,
      "grad_norm": 0.10817803852285492,
      "kl": 0.0004922151565551758,
      "learning_rate": 4.285714285714285e-07,
      "loss": 0.0,
      "reward": 0.33258930407464504,
      "reward_std": 0.4802868850529194,
      "rewards/equation_reward_func": 0.03906250197906047,
      "rewards/format_reward_func": 0.2935267984867096,
      "step": 12
    },
    {
      "completion_length": 471.93529510498047,
      "epoch": 0.017421060818167678,
      "grad_norm": 0.09644558381581954,
      "kl": 0.0010251998901367188,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 0.510044664144516,
      "reward_std": 0.5347170457243919,
      "rewards/equation_reward_func": 0.061383931897580624,
      "rewards/format_reward_func": 0.4486607387661934,
      "step": 14
    },
    {
      "completion_length": 457.7031478881836,
      "epoch": 0.019909783792191632,
      "grad_norm": 0.09293320344371607,
      "kl": 0.0017347335815429688,
      "learning_rate": 4.999740409224932e-07,
      "loss": 0.0,
      "reward": 0.5535714514553547,
      "reward_std": 0.5403263326734304,
      "rewards/equation_reward_func": 0.05357143096625805,
      "rewards/format_reward_func": 0.5000000223517418,
      "step": 16
    },
    {
      "completion_length": 449.1908645629883,
      "epoch": 0.022398506766215586,
      "grad_norm": 0.07268783989009871,
      "kl": 0.0044155120849609375,
      "learning_rate": 4.998961690809627e-07,
      "loss": 0.0,
      "reward": 0.7812500447034836,
      "reward_std": 0.5079520158469677,
      "rewards/equation_reward_func": 0.08258928905706853,
      "rewards/format_reward_func": 0.6986607424914837,
      "step": 18
    },
    {
      "completion_length": 439.35828018188477,
      "epoch": 0.02488722974023954,
      "grad_norm": 0.07030613994163097,
      "kl": 0.007472991943359375,
      "learning_rate": 4.997664006472578e-07,
      "loss": 0.0,
      "reward": 0.8560268208384514,
      "reward_std": 0.48480591736733913,
      "rewards/equation_reward_func": 0.09933036286383867,
      "rewards/format_reward_func": 0.7566964626312256,
      "step": 20
    },
    {
      "completion_length": 448.36609268188477,
      "epoch": 0.027375952714263494,
      "grad_norm": 0.0688012680174,
      "kl": 0.0070819854736328125,
      "learning_rate": 4.995847625707292e-07,
      "loss": 0.0,
      "reward": 0.856026828289032,
      "reward_std": 0.45433524437248707,
      "rewards/equation_reward_func": 0.07477678905706853,
      "rewards/format_reward_func": 0.7812500298023224,
      "step": 22
    },
    {
      "completion_length": 449.35046577453613,
      "epoch": 0.029864675688287448,
      "grad_norm": 0.06658434704593806,
      "kl": 0.007633209228515625,
      "learning_rate": 4.993512925726318e-07,
      "loss": 0.0,
      "reward": 0.9084821790456772,
      "reward_std": 0.42067001201212406,
      "rewards/equation_reward_func": 0.0792410756694153,
      "rewards/format_reward_func": 0.8292411155998707,
      "step": 24
    },
    {
      "completion_length": 451.9955539703369,
      "epoch": 0.0323533986623114,
      "grad_norm": 0.059102019385286024,
      "kl": 0.008182525634765625,
      "learning_rate": 4.990660391382923e-07,
      "loss": 0.0,
      "reward": 0.9698661118745804,
      "reward_std": 0.4051512759178877,
      "rewards/equation_reward_func": 0.10714286286383867,
      "rewards/format_reward_func": 0.8627232685685158,
      "step": 26
    },
    {
      "completion_length": 436.1049270629883,
      "epoch": 0.034842121636335356,
      "grad_norm": 0.0505975332867349,
      "kl": 0.010929107666015625,
      "learning_rate": 4.987290615070384e-07,
      "loss": 0.0,
      "reward": 0.9866071864962578,
      "reward_std": 0.30416737496852875,
      "rewards/equation_reward_func": 0.07366071757860482,
      "rewards/format_reward_func": 0.9129464700818062,
      "step": 28
    },
    {
      "completion_length": 415.63952827453613,
      "epoch": 0.03733084461035931,
      "grad_norm": 0.05617076097244933,
      "kl": 0.01288604736328125,
      "learning_rate": 4.983404296598978e-07,
      "loss": 0.0,
      "reward": 1.0814732611179352,
      "reward_std": 0.3300146460533142,
      "rewards/equation_reward_func": 0.1395089360885322,
      "rewards/format_reward_func": 0.9419643245637417,
      "step": 30
    },
    {
      "completion_length": 406.3560447692871,
      "epoch": 0.039819567584383264,
      "grad_norm": 0.04537480362751575,
      "kl": 0.015163421630859375,
      "learning_rate": 4.979002243050646e-07,
      "loss": 0.0,
      "reward": 1.0691964663565159,
      "reward_std": 0.25317980349063873,
      "rewards/equation_reward_func": 0.10602679057046771,
      "rewards/format_reward_func": 0.9631696753203869,
      "step": 32
    },
    {
      "completion_length": 420.8080539703369,
      "epoch": 0.042308290558407215,
      "grad_norm": 0.05671737906093819,
      "kl": 0.015956878662109375,
      "learning_rate": 4.974085368611381e-07,
      "loss": 0.0,
      "reward": 1.0703125521540642,
      "reward_std": 0.28108513820916414,
      "rewards/equation_reward_func": 0.1138392926659435,
      "rewards/format_reward_func": 0.956473246216774,
      "step": 34
    },
    {
      "completion_length": 390.3638553619385,
      "epoch": 0.04479701353243117,
      "grad_norm": 0.057276721518591346,
      "kl": 0.0186004638671875,
      "learning_rate": 4.968654694381379e-07,
      "loss": 0.0,
      "reward": 1.1171875521540642,
      "reward_std": 0.2574427053332329,
      "rewards/equation_reward_func": 0.13616072107106447,
      "rewards/format_reward_func": 0.9810268059372902,
      "step": 36
    },
    {
      "completion_length": 388.2288131713867,
      "epoch": 0.04728573650645512,
      "grad_norm": 0.05608729427224838,
      "kl": 0.025173187255859375,
      "learning_rate": 4.962711348162987e-07,
      "loss": 0.0,
      "reward": 1.0770089775323868,
      "reward_std": 0.2866173293441534,
      "rewards/equation_reward_func": 0.112723219092004,
      "rewards/format_reward_func": 0.9642857387661934,
      "step": 38
    },
    {
      "completion_length": 373.26229095458984,
      "epoch": 0.04977445948047908,
      "grad_norm": 0.05668420494869164,
      "kl": 0.02100372314453125,
      "learning_rate": 4.956256564226487e-07,
      "loss": 0.0,
      "reward": 1.1049107648432255,
      "reward_std": 0.254839526489377,
      "rewards/equation_reward_func": 0.1272321477299556,
      "rewards/format_reward_func": 0.9776786081492901,
      "step": 40
    },
    {
      "completion_length": 381.3013553619385,
      "epoch": 0.05226318245450303,
      "grad_norm": 0.05384466374487655,
      "kl": 0.02074432373046875,
      "learning_rate": 4.949291683053768e-07,
      "loss": 0.0,
      "reward": 1.0825893394649029,
      "reward_std": 0.23148810863494873,
      "rewards/equation_reward_func": 0.10714286239817739,
      "rewards/format_reward_func": 0.9754464626312256,
      "step": 42
    },
    {
      "completion_length": 374.8694362640381,
      "epoch": 0.05475190542852699,
      "grad_norm": 0.05693779183820393,
      "kl": 0.0219879150390625,
      "learning_rate": 4.941818151059955e-07,
      "loss": 0.0,
      "reward": 1.0959821939468384,
      "reward_std": 0.23808866180479527,
      "rewards/equation_reward_func": 0.11607143329456449,
      "rewards/format_reward_func": 0.9799107499420643,
      "step": 44
    },
    {
      "completion_length": 357.81028175354004,
      "epoch": 0.05724062840255094,
      "grad_norm": 0.050061664583350464,
      "kl": 0.025909423828125,
      "learning_rate": 4.933837520293017e-07,
      "loss": 0.0,
      "reward": 1.1026786416769028,
      "reward_std": 0.22216346580535173,
      "rewards/equation_reward_func": 0.1160714344587177,
      "rewards/format_reward_func": 0.9866071753203869,
      "step": 46
    },
    {
      "completion_length": 376.34823417663574,
      "epoch": 0.059729351376574896,
      "grad_norm": 0.05730255508966384,
      "kl": 0.0270843505859375,
      "learning_rate": 4.925351448111454e-07,
      "loss": 0.0,
      "reward": 1.1517857685685158,
      "reward_std": 0.2740681590512395,
      "rewards/equation_reward_func": 0.17299108253791928,
      "rewards/format_reward_func": 0.9787946753203869,
      "step": 48
    },
    {
      "completion_length": 355.76564025878906,
      "epoch": 0.06221807435059885,
      "grad_norm": 0.052885138884175203,
      "kl": 0.03263092041015625,
      "learning_rate": 4.91636169684011e-07,
      "loss": 0.0,
      "reward": 1.13058041036129,
      "reward_std": 0.2741837967187166,
      "rewards/equation_reward_func": 0.1551339344587177,
      "rewards/format_reward_func": 0.9754464514553547,
      "step": 50
    },
    {
      "completion_length": 333.06139945983887,
      "epoch": 0.0647067973246228,
      "grad_norm": 0.06133859386576639,
      "kl": 0.0342864990234375,
      "learning_rate": 4.906870133404186e-07,
      "loss": 0.0,
      "reward": 1.14620541036129,
      "reward_std": 0.26373843336477876,
      "rewards/equation_reward_func": 0.15959822433069348,
      "rewards/format_reward_func": 0.9866071678698063,
      "step": 52
    },
    {
      "completion_length": 359.7198848724365,
      "epoch": 0.06719552029864675,
      "grad_norm": 0.05999061818317411,
      "kl": 0.035614013671875,
      "learning_rate": 4.896878728941531e-07,
      "loss": 0.0,
      "reward": 1.13058041036129,
      "reward_std": 0.30478311236947775,
      "rewards/equation_reward_func": 0.16741072200238705,
      "rewards/format_reward_func": 0.9631696715950966,
      "step": 54
    },
    {
      "completion_length": 339.29354095458984,
      "epoch": 0.06968424327267071,
      "grad_norm": 0.05830073958166394,
      "kl": 0.0386199951171875,
      "learning_rate": 4.886389558393284e-07,
      "loss": 0.0,
      "reward": 1.1930803954601288,
      "reward_std": 0.289231700822711,
      "rewards/equation_reward_func": 0.20982143562287092,
      "rewards/format_reward_func": 0.9832589514553547,
      "step": 56
    },
    {
      "completion_length": 356.0680980682373,
      "epoch": 0.07217296624669467,
      "grad_norm": 0.0544928686410873,
      "kl": 0.041656494140625,
      "learning_rate": 4.875404800072976e-07,
      "loss": 0.0,
      "reward": 1.1406250596046448,
      "reward_std": 0.22303624730557203,
      "rewards/equation_reward_func": 0.15625000721774995,
      "rewards/format_reward_func": 0.984375037252903,
      "step": 58
    },
    {
      "completion_length": 340.20537185668945,
      "epoch": 0.07466168922071861,
      "grad_norm": 0.05744987372930731,
      "kl": 0.0499420166015625,
      "learning_rate": 4.86392673521415e-07,
      "loss": 0.0,
      "reward": 1.1741071939468384,
      "reward_std": 0.27861948776990175,
      "rewards/equation_reward_func": 0.184151794295758,
      "rewards/format_reward_func": 0.9899553768336773,
      "step": 60
    },
    {
      "completion_length": 340.42635345458984,
      "epoch": 0.07715041219474257,
      "grad_norm": 0.061800785648734306,
      "kl": 0.04815673828125,
      "learning_rate": 4.851957747496606e-07,
      "loss": 0.0,
      "reward": 1.2120536342263222,
      "reward_std": 0.2723410092294216,
      "rewards/equation_reward_func": 0.22656251164153218,
      "rewards/format_reward_func": 0.9854911006987095,
      "step": 62
    },
    {
      "completion_length": 360.4308204650879,
      "epoch": 0.07963913516876653,
      "grad_norm": 0.054162346491789554,
      "kl": 0.05035400390625,
      "learning_rate": 4.839500322551386e-07,
      "loss": 0.0001,
      "reward": 1.1752232611179352,
      "reward_std": 0.2869079066440463,
      "rewards/equation_reward_func": 0.19419643748551607,
      "rewards/format_reward_func": 0.9810268133878708,
      "step": 64
    },
    {
      "completion_length": 350.62166595458984,
      "epoch": 0.08212785814279049,
      "grad_norm": 0.06519334959182187,
      "kl": 0.0565338134765625,
      "learning_rate": 4.826557047444563e-07,
      "loss": 0.0001,
      "reward": 1.1953125670552254,
      "reward_std": 0.30847367085516453,
      "rewards/equation_reward_func": 0.21763393934816122,
      "rewards/format_reward_func": 0.9776786006987095,
      "step": 66
    },
    {
      "completion_length": 353.1428756713867,
      "epoch": 0.08461658111681443,
      "grad_norm": 0.06183291246364765,
      "kl": 0.0570526123046875,
      "learning_rate": 4.813130610139993e-07,
      "loss": 0.0001,
      "reward": 1.2187500447034836,
      "reward_std": 0.31583766732364893,
      "rewards/equation_reward_func": 0.23549107927829027,
      "rewards/format_reward_func": 0.9832589626312256,
      "step": 68
    },
    {
      "completion_length": 374.2712211608887,
      "epoch": 0.08710530409083839,
      "grad_norm": 0.057224664305537236,
      "kl": 0.0590667724609375,
      "learning_rate": 4.799223798941089e-07,
      "loss": 0.0001,
      "reward": 1.188616119325161,
      "reward_std": 0.3013247288763523,
      "rewards/equation_reward_func": 0.21875001047737896,
      "rewards/format_reward_func": 0.9698660969734192,
      "step": 70
    },
    {
      "completion_length": 371.8750114440918,
      "epoch": 0.08959402706486234,
      "grad_norm": 0.058154083828801245,
      "kl": 0.064361572265625,
      "learning_rate": 4.78483950191177e-07,
      "loss": 0.0001,
      "reward": 1.2187500447034836,
      "reward_std": 0.32332514226436615,
      "rewards/equation_reward_func": 0.2433035826543346,
      "rewards/format_reward_func": 0.9754464589059353,
      "step": 72
    },
    {
      "completion_length": 365.0033664703369,
      "epoch": 0.0920827500388863,
      "grad_norm": 0.05964598037277607,
      "kl": 0.0686798095703125,
      "learning_rate": 4.769980706276687e-07,
      "loss": 0.0001,
      "reward": 1.243303619325161,
      "reward_std": 0.3098542420193553,
      "rewards/equation_reward_func": 0.2622767958091572,
      "rewards/format_reward_func": 0.9810268133878708,
      "step": 74
    },
    {
      "completion_length": 369.5089473724365,
      "epoch": 0.09457147301291025,
      "grad_norm": 0.051790769083910046,
      "kl": 0.074310302734375,
      "learning_rate": 4.7546504978008595e-07,
      "loss": 0.0001,
      "reward": 1.2645089775323868,
      "reward_std": 0.30733966641128063,
      "rewards/equation_reward_func": 0.2890625153668225,
      "rewards/format_reward_func": 0.9754464700818062,
      "step": 76
    },
    {
      "completion_length": 384.2756881713867,
      "epoch": 0.0970601959869342,
      "grad_norm": 0.05484455622025143,
      "kl": 0.07171630859375,
      "learning_rate": 4.738852060148848e-07,
      "loss": 0.0001,
      "reward": 1.250000063329935,
      "reward_std": 0.3137952769175172,
      "rewards/equation_reward_func": 0.2756696529686451,
      "rewards/format_reward_func": 0.9743303991854191,
      "step": 78
    },
    {
      "completion_length": 408.2500171661377,
      "epoch": 0.09954891896095816,
      "grad_norm": 0.0574310858584145,
      "kl": 0.075714111328125,
      "learning_rate": 4.722588674223593e-07,
      "loss": 0.0001,
      "reward": 1.2343750521540642,
      "reward_std": 0.3266843259334564,
      "rewards/equation_reward_func": 0.2801339440047741,
      "rewards/format_reward_func": 0.9542410969734192,
      "step": 80
    },
    {
      "completion_length": 418.1774711608887,
      "epoch": 0.10203764193498212,
      "grad_norm": 0.04959040686750888,
      "kl": 0.07958984375,
      "learning_rate": 4.70586371748506e-07,
      "loss": 0.0001,
      "reward": 1.2377232685685158,
      "reward_std": 0.3039839044213295,
      "rewards/equation_reward_func": 0.27455358440056443,
      "rewards/format_reward_func": 0.9631696678698063,
      "step": 82
    },
    {
      "completion_length": 392.39064598083496,
      "epoch": 0.10452636490900606,
      "grad_norm": 0.05828296572641528,
      "kl": 0.086578369140625,
      "learning_rate": 4.6886806632488363e-07,
      "loss": 0.0001,
      "reward": 1.2935268506407738,
      "reward_std": 0.3257301030680537,
      "rewards/equation_reward_func": 0.32254465483129025,
      "rewards/format_reward_func": 0.9709821790456772,
      "step": 84
    },
    {
      "completion_length": 393.9118461608887,
      "epoch": 0.10701508788303002,
      "grad_norm": 0.05583356462661088,
      "kl": 0.100616455078125,
      "learning_rate": 4.6710430799648143e-07,
      "loss": 0.0001,
      "reward": 1.2979911342263222,
      "reward_std": 0.2977573899552226,
      "rewards/equation_reward_func": 0.3370535937137902,
      "rewards/format_reward_func": 0.9609375298023224,
      "step": 86
    },
    {
      "completion_length": 418.97546768188477,
      "epoch": 0.10950381085705398,
      "grad_norm": 0.05984739996252054,
      "kl": 0.091064453125,
      "learning_rate": 4.652954630476127e-07,
      "loss": 0.0001,
      "reward": 1.2444196939468384,
      "reward_std": 0.347733149304986,
      "rewards/equation_reward_func": 0.29241072945296764,
      "rewards/format_reward_func": 0.9520089663565159,
      "step": 88
    },
    {
      "completion_length": 412.0435428619385,
      "epoch": 0.11199253383107793,
      "grad_norm": 0.05993124016275902,
      "kl": 0.096588134765625,
      "learning_rate": 4.6344190712584713e-07,
      "loss": 0.0001,
      "reward": 1.3292411416769028,
      "reward_std": 0.36256127804517746,
      "rewards/equation_reward_func": 0.37388394493609667,
      "rewards/format_reward_func": 0.9553571864962578,
      "step": 90
    },
    {
      "completion_length": 417.2634086608887,
      "epoch": 0.11448125680510188,
      "grad_norm": 0.05382564496764054,
      "kl": 0.09246826171875,
      "learning_rate": 4.615440251639995e-07,
      "loss": 0.0001,
      "reward": 1.290178619325161,
      "reward_std": 0.30793262645602226,
      "rewards/equation_reward_func": 0.33147323224693537,
      "rewards/format_reward_func": 0.9587053991854191,
      "step": 92
    },
    {
      "completion_length": 443.2835006713867,
      "epoch": 0.11696997977912583,
      "grad_norm": 0.06169358639422184,
      "kl": 0.100860595703125,
      "learning_rate": 4.596022113001894e-07,
      "loss": 0.0001,
      "reward": 1.2935268580913544,
      "reward_std": 0.35633409582078457,
      "rewards/equation_reward_func": 0.34821430686861277,
      "rewards/format_reward_func": 0.9453125447034836,
      "step": 94
    },
    {
      "completion_length": 408.9777011871338,
      "epoch": 0.11945870275314979,
      "grad_norm": 0.06289641800144408,
      "kl": 0.10186767578125,
      "learning_rate": 4.576168687959895e-07,
      "loss": 0.0001,
      "reward": 1.3582589849829674,
      "reward_std": 0.3258579559624195,
      "rewards/equation_reward_func": 0.40178573317825794,
      "rewards/format_reward_func": 0.9564732536673546,
      "step": 96
    },
    {
      "completion_length": 431.306941986084,
      "epoch": 0.12194742572717375,
      "grad_norm": 0.05214216522023378,
      "kl": 0.101165771484375,
      "learning_rate": 4.555884099526793e-07,
      "loss": 0.0001,
      "reward": 1.3091518506407738,
      "reward_std": 0.28289179550483823,
      "rewards/equation_reward_func": 0.3560268050059676,
      "rewards/format_reward_func": 0.9531250260770321,
      "step": 98
    },
    {
      "completion_length": 427.6640796661377,
      "epoch": 0.1244361487011977,
      "grad_norm": 0.06593974374565598,
      "kl": 0.106964111328125,
      "learning_rate": 4.5351725602562174e-07,
      "loss": 0.0001,
      "reward": 1.311383992433548,
      "reward_std": 0.3198657017201185,
      "rewards/equation_reward_func": 0.36272323317825794,
      "rewards/format_reward_func": 0.9486607573926449,
      "step": 100
    },
    {
      "completion_length": 412.4486827850342,
      "epoch": 0.12692487167522165,
      "grad_norm": 0.057534299379812,
      "kl": 0.104766845703125,
      "learning_rate": 4.514038371367791e-07,
      "loss": 0.0001,
      "reward": 1.4162947088479996,
      "reward_std": 0.2974701370112598,
      "rewards/equation_reward_func": 0.44531251955777407,
      "rewards/format_reward_func": 0.9709821678698063,
      "step": 102
    },
    {
      "completion_length": 412.8136329650879,
      "epoch": 0.1294135946492456,
      "grad_norm": 0.047305658451965514,
      "kl": 0.11578369140625,
      "learning_rate": 4.4924859218538936e-07,
      "loss": 0.0001,
      "reward": 1.358258992433548,
      "reward_std": 0.31219158321619034,
      "rewards/equation_reward_func": 0.40513394866138697,
      "rewards/format_reward_func": 0.9531250409781933,
      "step": 104
    },
    {
      "completion_length": 414.66296195983887,
      "epoch": 0.13190231762326957,
      "grad_norm": 0.05814046051575021,
      "kl": 0.118560791015625,
      "learning_rate": 4.470519687568185e-07,
      "loss": 0.0001,
      "reward": 1.3627232760190964,
      "reward_std": 0.28798825573176146,
      "rewards/equation_reward_func": 0.3973214505240321,
      "rewards/format_reward_func": 0.9654018133878708,
      "step": 106
    },
    {
      "completion_length": 408.1540355682373,
      "epoch": 0.1343910405972935,
      "grad_norm": 0.056119658206168294,
      "kl": 0.114013671875,
      "learning_rate": 4.4481442302960923e-07,
      "loss": 0.0001,
      "reward": 1.3459822088479996,
      "reward_std": 0.27404902782291174,
      "rewards/equation_reward_func": 0.3761160969734192,
      "rewards/format_reward_func": 0.9698660969734192,
      "step": 108
    },
    {
      "completion_length": 389.2522506713867,
      "epoch": 0.13687976357131748,
      "grad_norm": 0.06337568584346356,
      "kl": 0.118377685546875,
      "learning_rate": 4.4253641968074505e-07,
      "loss": 0.0001,
      "reward": 1.4229911267757416,
      "reward_std": 0.2757099084556103,
      "rewards/equation_reward_func": 0.4609375149011612,
      "rewards/format_reward_func": 0.9620536044239998,
      "step": 110
    },
    {
      "completion_length": 378.68305015563965,
      "epoch": 0.13936848654534142,
      "grad_norm": 0.055648006379385793,
      "kl": 0.134857177734375,
      "learning_rate": 4.402184317891501e-07,
      "loss": 0.0001,
      "reward": 1.439732201397419,
      "reward_std": 0.2977797780185938,
      "rewards/equation_reward_func": 0.46651787497103214,
      "rewards/format_reward_func": 0.9732143171131611,
      "step": 112
    },
    {
      "completion_length": 407.8962211608887,
      "epoch": 0.14185720951936537,
      "grad_norm": 0.0591540848415508,
      "kl": 0.126220703125,
      "learning_rate": 4.37860940737443e-07,
      "loss": 0.0001,
      "reward": 1.3571429178118706,
      "reward_std": 0.24285848252475262,
      "rewards/equation_reward_func": 0.39174109138548374,
      "rewards/format_reward_func": 0.9654018133878708,
      "step": 114
    },
    {
      "completion_length": 360.5067090988159,
      "epoch": 0.14434593249338934,
      "grad_norm": 0.05663015266712956,
      "kl": 0.133392333984375,
      "learning_rate": 4.354644361119671e-07,
      "loss": 0.0001,
      "reward": 1.4698661342263222,
      "reward_std": 0.21007388643920422,
      "rewards/equation_reward_func": 0.49330359790474176,
      "rewards/format_reward_func": 0.9765625298023224,
      "step": 116
    },
    {
      "completion_length": 376.5457820892334,
      "epoch": 0.14683465546741328,
      "grad_norm": 0.05339975098270581,
      "kl": 0.1224365234375,
      "learning_rate": 4.3302941560111716e-07,
      "loss": 0.0001,
      "reward": 1.434151865541935,
      "reward_std": 0.21489905659109354,
      "rewards/equation_reward_func": 0.45647323597222567,
      "rewards/format_reward_func": 0.9776785932481289,
      "step": 118
    },
    {
      "completion_length": 365.3828287124634,
      "epoch": 0.14932337844143723,
      "grad_norm": 0.05910197740570998,
      "kl": 0.1470947265625,
      "learning_rate": 4.3055638489198236e-07,
      "loss": 0.0001,
      "reward": 1.3984375670552254,
      "reward_std": 0.2542656445875764,
      "rewards/equation_reward_func": 0.42857144493609667,
      "rewards/format_reward_func": 0.9698660969734192,
      "step": 120
    },
    {
      "completion_length": 352.23662662506104,
      "epoch": 0.1518121014154612,
      "grad_norm": 0.05864149302937991,
      "kl": 0.146240234375,
      "learning_rate": 4.280458575653296e-07,
      "loss": 0.0001,
      "reward": 1.456473283469677,
      "reward_std": 0.18524617236107588,
      "rewards/equation_reward_func": 0.4754464505240321,
      "rewards/format_reward_func": 0.9810268171131611,
      "step": 122
    },
    {
      "completion_length": 348.39287090301514,
      "epoch": 0.15430082438948514,
      "grad_norm": 0.0663797306790776,
      "kl": 0.15472412109375,
      "learning_rate": 4.2549835498894665e-07,
      "loss": 0.0002,
      "reward": 1.4453125670552254,
      "reward_std": 0.23717239312827587,
      "rewards/equation_reward_func": 0.4732143087312579,
      "rewards/format_reward_func": 0.972098246216774,
      "step": 124
    },
    {
      "completion_length": 356.01117610931396,
      "epoch": 0.1567895473635091,
      "grad_norm": 0.05497099225448875,
      "kl": 0.13714599609375,
      "learning_rate": 4.229144062093679e-07,
      "loss": 0.0001,
      "reward": 1.4397322051227093,
      "reward_std": 0.19848236767575145,
      "rewards/equation_reward_func": 0.46540180779993534,
      "rewards/format_reward_func": 0.9743303880095482,
      "step": 126
    },
    {
      "completion_length": 381.28573417663574,
      "epoch": 0.15927827033753306,
      "grad_norm": 0.05653433573745518,
      "kl": 0.13873291015625,
      "learning_rate": 4.2029454784200675e-07,
      "loss": 0.0001,
      "reward": 1.3180804327130318,
      "reward_std": 0.18374477326869965,
      "rewards/equation_reward_func": 0.34598215762525797,
      "rewards/format_reward_func": 0.9720982387661934,
      "step": 128
    },
    {
      "completion_length": 343.4129581451416,
      "epoch": 0.161766993311557,
      "grad_norm": 0.06565562774179554,
      "kl": 0.165069580078125,
      "learning_rate": 4.1763932395971433e-07,
      "loss": 0.0002,
      "reward": 1.4308036342263222,
      "reward_std": 0.16549686901271343,
      "rewards/equation_reward_func": 0.4441964477300644,
      "rewards/format_reward_func": 0.9866071678698063,
      "step": 130
    },
    {
      "completion_length": 348.75224781036377,
      "epoch": 0.16425571628558097,
      "grad_norm": 0.06462157522635509,
      "kl": 0.1668701171875,
      "learning_rate": 4.1494928597979117e-07,
      "loss": 0.0002,
      "reward": 1.4352679252624512,
      "reward_std": 0.1786380740813911,
      "rewards/equation_reward_func": 0.45424109883606434,
      "rewards/format_reward_func": 0.9810268096625805,
      "step": 132
    },
    {
      "completion_length": 308.1205520629883,
      "epoch": 0.16674443925960492,
      "grad_norm": 0.05800701976146753,
      "kl": 0.151275634765625,
      "learning_rate": 4.122249925494726e-07,
      "loss": 0.0002,
      "reward": 1.4765625670552254,
      "reward_std": 0.17193927755579352,
      "rewards/equation_reward_func": 0.48214288242161274,
      "rewards/format_reward_func": 0.994419664144516,
      "step": 134
    },
    {
      "completion_length": 346.0546998977661,
      "epoch": 0.16923316223362886,
      "grad_norm": 0.04492148649071812,
      "kl": 0.15283203125,
      "learning_rate": 4.094670094299131e-07,
      "loss": 0.0002,
      "reward": 1.3660714849829674,
      "reward_std": 0.161542855668813,
      "rewards/equation_reward_func": 0.38058037986047566,
      "rewards/format_reward_func": 0.9854911044239998,
      "step": 136
    },
    {
      "completion_length": 331.13618087768555,
      "epoch": 0.17172188520765283,
      "grad_norm": 0.2212276658068583,
      "kl": 0.537384033203125,
      "learning_rate": 4.066759093786931e-07,
      "loss": 0.0005,
      "reward": 1.41964291036129,
      "reward_std": 0.19885259168222547,
      "rewards/equation_reward_func": 0.43861609138548374,
      "rewards/format_reward_func": 0.9810268059372902,
      "step": 138
    },
    {
      "completion_length": 344.59264945983887,
      "epoch": 0.17421060818167677,
      "grad_norm": 0.05665698464447795,
      "kl": 0.151947021484375,
      "learning_rate": 4.038522720308732e-07,
      "loss": 0.0002,
      "reward": 1.3805804178118706,
      "reward_std": 0.18276562774553895,
      "rewards/equation_reward_func": 0.39508930686861277,
      "rewards/format_reward_func": 0.9854910969734192,
      "step": 140
    },
    {
      "completion_length": 328.3872928619385,
      "epoch": 0.17669933115570072,
      "grad_norm": 0.05858138418747856,
      "kl": 0.16424560546875,
      "learning_rate": 4.009966837786194e-07,
      "loss": 0.0002,
      "reward": 1.4631697237491608,
      "reward_std": 0.21255239751189947,
      "rewards/equation_reward_func": 0.47879466973245144,
      "rewards/format_reward_func": 0.9843750335276127,
      "step": 142
    },
    {
      "completion_length": 322.76898860931396,
      "epoch": 0.1791880541297247,
      "grad_norm": 0.051814888933446955,
      "kl": 0.157470703125,
      "learning_rate": 3.981097376494259e-07,
      "loss": 0.0002,
      "reward": 1.4575893506407738,
      "reward_std": 0.19401998538523912,
      "rewards/equation_reward_func": 0.4743303768336773,
      "rewards/format_reward_func": 0.9832589514553547,
      "step": 144
    },
    {
      "completion_length": 333.7712211608887,
      "epoch": 0.18167677710374863,
      "grad_norm": 0.05624207982164332,
      "kl": 0.160125732421875,
      "learning_rate": 3.951920331829592e-07,
      "loss": 0.0002,
      "reward": 1.4207589775323868,
      "reward_std": 0.15926225204020739,
      "rewards/equation_reward_func": 0.43861609138548374,
      "rewards/format_reward_func": 0.9821428842842579,
      "step": 146
    },
    {
      "completion_length": 319.8225564956665,
      "epoch": 0.1841655000777726,
      "grad_norm": 0.057600263729418885,
      "kl": 0.154571533203125,
      "learning_rate": 3.922441763065506e-07,
      "loss": 0.0002,
      "reward": 1.4151786416769028,
      "reward_std": 0.1929843365214765,
      "rewards/equation_reward_func": 0.43191966600716114,
      "rewards/format_reward_func": 0.9832589589059353,
      "step": 148
    },
    {
      "completion_length": 341.7154140472412,
      "epoch": 0.18665422305179655,
      "grad_norm": 0.05008851715635548,
      "kl": 0.17266845703125,
      "learning_rate": 3.8926677920936093e-07,
      "loss": 0.0002,
      "reward": 1.390625074505806,
      "reward_std": 0.1591361197642982,
      "rewards/equation_reward_func": 0.40290180686861277,
      "rewards/format_reward_func": 0.9877232424914837,
      "step": 150
    },
    {
      "completion_length": 308.49889850616455,
      "epoch": 0.1891429460258205,
      "grad_norm": 0.04651438810073453,
      "kl": 0.1688232421875,
      "learning_rate": 3.862604602152464e-07,
      "loss": 0.0002,
      "reward": 1.472098283469677,
      "reward_std": 0.17588552320376039,
      "rewards/equation_reward_func": 0.4854910932481289,
      "rewards/format_reward_func": 0.9866071790456772,
      "step": 152
    },
    {
      "completion_length": 329.714298248291,
      "epoch": 0.19163166899984446,
      "grad_norm": 0.05395346273397737,
      "kl": 0.1552734375,
      "learning_rate": 3.8322584365434934e-07,
      "loss": 0.0002,
      "reward": 1.409598283469677,
      "reward_std": 0.1818417957983911,
      "rewards/equation_reward_func": 0.43080359511077404,
      "rewards/format_reward_func": 0.9787946678698063,
      "step": 154
    },
    {
      "completion_length": 297.5982275009155,
      "epoch": 0.1941203919738684,
      "grad_norm": 0.04614157256005193,
      "kl": 0.17120361328125,
      "learning_rate": 3.8016355973344173e-07,
      "loss": 0.0002,
      "reward": 1.5055804252624512,
      "reward_std": 0.17592015489935875,
      "rewards/equation_reward_func": 0.5178571604192257,
      "rewards/format_reward_func": 0.9877232424914837,
      "step": 156
    },
    {
      "completion_length": 305.8538112640381,
      "epoch": 0.19660911494789235,
      "grad_norm": 0.0694599272801652,
      "kl": 0.2060546875,
      "learning_rate": 3.7707424440504863e-07,
      "loss": 0.0002,
      "reward": 1.4587054252624512,
      "reward_std": 0.19024762557819486,
      "rewards/equation_reward_func": 0.47098217345774174,
      "rewards/format_reward_func": 0.9877232350409031,
      "step": 158
    },
    {
      "completion_length": 326.53796005249023,
      "epoch": 0.19909783792191632,
      "grad_norm": 0.06216275817115459,
      "kl": 0.17303466796875,
      "learning_rate": 3.739585392353787e-07,
      "loss": 0.0002,
      "reward": 1.3604911342263222,
      "reward_std": 0.16597195249050856,
      "rewards/equation_reward_func": 0.37946430314332247,
      "rewards/format_reward_func": 0.9810268208384514,
      "step": 160
    },
    {
      "completion_length": 289.9274663925171,
      "epoch": 0.20158656089594026,
      "grad_norm": 0.07236670443181108,
      "kl": 0.18756103515625,
      "learning_rate": 3.7081709127108767e-07,
      "loss": 0.0002,
      "reward": 1.4866072162985802,
      "reward_std": 0.2302361116744578,
      "rewards/equation_reward_func": 0.5033482387661934,
      "rewards/format_reward_func": 0.9832589589059353,
      "step": 162
    },
    {
      "completion_length": 323.6227798461914,
      "epoch": 0.20407528386996424,
      "grad_norm": 0.048051970688120026,
      "kl": 0.172607421875,
      "learning_rate": 3.6765055290490513e-07,
      "loss": 0.0002,
      "reward": 1.3683036342263222,
      "reward_std": 0.17599576339125633,
      "rewards/equation_reward_func": 0.38504466286394745,
      "rewards/format_reward_func": 0.9832589589059353,
      "step": 164
    },
    {
      "completion_length": 287.5881814956665,
      "epoch": 0.20656400684398818,
      "grad_norm": 0.05486464300817734,
      "kl": 0.18212890625,
      "learning_rate": 3.644595817401501e-07,
      "loss": 0.0002,
      "reward": 1.4843750670552254,
      "reward_std": 0.17731093661859632,
      "rewards/equation_reward_func": 0.5011160932481289,
      "rewards/format_reward_func": 0.9832589626312256,
      "step": 166
    },
    {
      "completion_length": 284.36050605773926,
      "epoch": 0.20905272981801212,
      "grad_norm": 0.06808427865251532,
      "kl": 0.195556640625,
      "learning_rate": 3.6124484045416483e-07,
      "loss": 0.0002,
      "reward": 1.4665179327130318,
      "reward_std": 0.20187203446403146,
      "rewards/equation_reward_func": 0.48660716600716114,
      "rewards/format_reward_func": 0.9799107611179352,
      "step": 168
    },
    {
      "completion_length": 291.9732255935669,
      "epoch": 0.2115414527920361,
      "grad_norm": 0.04519403336857609,
      "kl": 0.18017578125,
      "learning_rate": 3.580069966606949e-07,
      "loss": 0.0002,
      "reward": 1.4084821939468384,
      "reward_std": 0.19656919362023473,
      "rewards/equation_reward_func": 0.44308037497103214,
      "rewards/format_reward_func": 0.9654018208384514,
      "step": 170
    },
    {
      "completion_length": 318.1685400009155,
      "epoch": 0.21403017576606004,
      "grad_norm": 0.061705947760315766,
      "kl": 0.1832275390625,
      "learning_rate": 3.547467227712444e-07,
      "loss": 0.0002,
      "reward": 1.3526786342263222,
      "reward_std": 0.19870673725381494,
      "rewards/equation_reward_func": 0.37834823690354824,
      "rewards/format_reward_func": 0.9743303954601288,
      "step": 172
    },
    {
      "completion_length": 301.315860748291,
      "epoch": 0.21651889874008398,
      "grad_norm": 0.054334606936889365,
      "kl": 0.1953125,
      "learning_rate": 3.5146469585543386e-07,
      "loss": 0.0002,
      "reward": 1.4017857760190964,
      "reward_std": 0.20476062037050724,
      "rewards/equation_reward_func": 0.42075894959270954,
      "rewards/format_reward_func": 0.9810268096625805,
      "step": 174
    },
    {
      "completion_length": 304.6763563156128,
      "epoch": 0.21900762171410795,
      "grad_norm": 0.05052301150918478,
      "kl": 0.1910400390625,
      "learning_rate": 3.481615975003922e-07,
      "loss": 0.0002,
      "reward": 1.3560268431901932,
      "reward_std": 0.19252017885446548,
      "rewards/equation_reward_func": 0.38169644493609667,
      "rewards/format_reward_func": 0.9743303842842579,
      "step": 176
    },
    {
      "completion_length": 269.1328239440918,
      "epoch": 0.2214963446881319,
      "grad_norm": 0.05172823495274905,
      "kl": 0.2069091796875,
      "learning_rate": 3.448381136692089e-07,
      "loss": 0.0002,
      "reward": 1.4609375596046448,
      "reward_std": 0.16476588556542993,
      "rewards/equation_reward_func": 0.48883930779993534,
      "rewards/format_reward_func": 0.972098246216774,
      "step": 178
    },
    {
      "completion_length": 308.0904140472412,
      "epoch": 0.22398506766215587,
      "grad_norm": 0.0613229880527607,
      "kl": 0.19024658203125,
      "learning_rate": 3.4149493455847897e-07,
      "loss": 0.0002,
      "reward": 1.3571429178118706,
      "reward_std": 0.22263891389593482,
      "rewards/equation_reward_func": 0.393973232829012,
      "rewards/format_reward_func": 0.963169664144516,
      "step": 180
    },
    {
      "completion_length": 265.180814743042,
      "epoch": 0.2264737906361798,
      "grad_norm": 0.05586824711527208,
      "kl": 0.20489501953125,
      "learning_rate": 3.3813275445496766e-07,
      "loss": 0.0002,
      "reward": 1.520089354366064,
      "reward_std": 0.18474067840725183,
      "rewards/equation_reward_func": 0.5479911003494635,
      "rewards/format_reward_func": 0.972098246216774,
      "step": 182
    },
    {
      "completion_length": 303.05358600616455,
      "epoch": 0.22896251361020376,
      "grad_norm": 0.04781670355873072,
      "kl": 0.19744873046875,
      "learning_rate": 3.347522715914262e-07,
      "loss": 0.0002,
      "reward": 1.379464365541935,
      "reward_std": 0.21752706728875637,
      "rewards/equation_reward_func": 0.4174107350409031,
      "rewards/format_reward_func": 0.9620536006987095,
      "step": 184
    },
    {
      "completion_length": 266.5502338409424,
      "epoch": 0.23145123658422773,
      "grad_norm": 0.07219267256948605,
      "kl": 0.21258544921875,
      "learning_rate": 3.313541880015877e-07,
      "loss": 0.0002,
      "reward": 1.476562574505806,
      "reward_std": 0.1959572951309383,
      "rewards/equation_reward_func": 0.5033482350409031,
      "rewards/format_reward_func": 0.9732143245637417,
      "step": 186
    },
    {
      "completion_length": 285.2198791503906,
      "epoch": 0.23393995955825167,
      "grad_norm": 0.1000308934340073,
      "kl": 0.215087890625,
      "learning_rate": 3.279392093743747e-07,
      "loss": 0.0002,
      "reward": 1.3883929252624512,
      "reward_std": 0.21638350887224078,
      "rewards/equation_reward_func": 0.4252232303842902,
      "rewards/format_reward_func": 0.9631696715950966,
      "step": 188
    },
    {
      "completion_length": 275.9531412124634,
      "epoch": 0.2364286825322756,
      "grad_norm": 0.063450820634565,
      "kl": 0.2064208984375,
      "learning_rate": 3.245080449073459e-07,
      "loss": 0.0002,
      "reward": 1.4296875670552254,
      "reward_std": 0.18975705141201615,
      "rewards/equation_reward_func": 0.46093752048909664,
      "rewards/format_reward_func": 0.9687500409781933,
      "step": 190
    },
    {
      "completion_length": 281.63282680511475,
      "epoch": 0.23891740550629958,
      "grad_norm": 0.07927927952984541,
      "kl": 0.23663330078125,
      "learning_rate": 3.210614071594162e-07,
      "loss": 0.0002,
      "reward": 1.4776786491274834,
      "reward_std": 0.2009006328880787,
      "rewards/equation_reward_func": 0.5212053842842579,
      "rewards/format_reward_func": 0.9564732536673546,
      "step": 192
    },
    {
      "completion_length": 306.393985748291,
      "epoch": 0.24140612848032353,
      "grad_norm": 0.06500632509866108,
      "kl": 0.2120361328125,
      "learning_rate": 3.1760001190287695e-07,
      "loss": 0.0002,
      "reward": 1.3917411416769028,
      "reward_std": 0.23199980147182941,
      "rewards/equation_reward_func": 0.44084823317825794,
      "rewards/format_reward_func": 0.9508928880095482,
      "step": 194
    },
    {
      "completion_length": 313.911847114563,
      "epoch": 0.2438948514543475,
      "grad_norm": 0.07817216779042299,
      "kl": 0.22210693359375,
      "learning_rate": 3.141245779747502e-07,
      "loss": 0.0002,
      "reward": 1.3649554178118706,
      "reward_std": 0.21487555792555213,
      "rewards/equation_reward_func": 0.4207589514553547,
      "rewards/format_reward_func": 0.9441964663565159,
      "step": 196
    },
    {
      "completion_length": 304.01787090301514,
      "epoch": 0.24638357442837144,
      "grad_norm": 0.059524282155233844,
      "kl": 0.23980712890625,
      "learning_rate": 3.106358271275056e-07,
      "loss": 0.0002,
      "reward": 1.3984375596046448,
      "reward_std": 0.21076519135385752,
      "rewards/equation_reward_func": 0.44196430314332247,
      "rewards/format_reward_func": 0.9564732424914837,
      "step": 198
    },
    {
      "completion_length": 292.20983505249023,
      "epoch": 0.2488722974023954,
      "grad_norm": 0.05679318708182419,
      "kl": 0.21978759765625,
      "learning_rate": 3.0713448387917227e-07,
      "loss": 0.0002,
      "reward": 1.4107143580913544,
      "reward_std": 0.1913211834616959,
      "rewards/equation_reward_func": 0.4508928805589676,
      "rewards/format_reward_func": 0.9598214589059353,
      "step": 200
    },
    {
      "completion_length": 279.8493461608887,
      "epoch": 0.25136102037641933,
      "grad_norm": 0.05632547780170071,
      "kl": 0.2144775390625,
      "learning_rate": 3.0362127536287636e-07,
      "loss": 0.0002,
      "reward": 1.4609375670552254,
      "reward_std": 0.19915124960243702,
      "rewards/equation_reward_func": 0.49665181152522564,
      "rewards/format_reward_func": 0.964285746216774,
      "step": 202
    },
    {
      "completion_length": 305.8973331451416,
      "epoch": 0.2538497433504433,
      "grad_norm": 0.06737671533384239,
      "kl": 0.23370361328125,
      "learning_rate": 3.0009693117583523e-07,
      "loss": 0.0002,
      "reward": 1.3426339998841286,
      "reward_std": 0.23496287874877453,
      "rewards/equation_reward_func": 0.3895089477300644,
      "rewards/format_reward_func": 0.9531250335276127,
      "step": 204
    },
    {
      "completion_length": 306.7031373977661,
      "epoch": 0.2563384663244673,
      "grad_norm": 0.055756372882820274,
      "kl": 0.2335205078125,
      "learning_rate": 2.965621832278401e-07,
      "loss": 0.0002,
      "reward": 1.3973214998841286,
      "reward_std": 0.21409290935844183,
      "rewards/equation_reward_func": 0.435267879627645,
      "rewards/format_reward_func": 0.9620536081492901,
      "step": 206
    },
    {
      "completion_length": 301.6216621398926,
      "epoch": 0.2588271892984912,
      "grad_norm": 0.057990490708177245,
      "kl": 0.210205078125,
      "learning_rate": 2.9301776558925875e-07,
      "loss": 0.0002,
      "reward": 1.3593750596046448,
      "reward_std": 0.18663123482838273,
      "rewards/equation_reward_func": 0.39397323317825794,
      "rewards/format_reward_func": 0.9654018208384514,
      "step": 208
    },
    {
      "completion_length": 302.0312662124634,
      "epoch": 0.26131591227251516,
      "grad_norm": 0.05591978437529843,
      "kl": 0.19818115234375,
      "learning_rate": 2.894644143385885e-07,
      "loss": 0.0002,
      "reward": 1.3549107983708382,
      "reward_std": 0.181817049626261,
      "rewards/equation_reward_func": 0.38839287823066115,
      "rewards/format_reward_func": 0.9665178917348385,
      "step": 210
    },
    {
      "completion_length": 295.438627243042,
      "epoch": 0.26380463524653913,
      "grad_norm": 0.05880963721504364,
      "kl": 0.20281982421875,
      "learning_rate": 2.859028674095937e-07,
      "loss": 0.0002,
      "reward": 1.3861607611179352,
      "reward_std": 0.1711200401186943,
      "rewards/equation_reward_func": 0.415178588591516,
      "rewards/format_reward_func": 0.9709821790456772,
      "step": 212
    },
    {
      "completion_length": 280.2243413925171,
      "epoch": 0.26629335822056305,
      "grad_norm": 0.05541782868549402,
      "kl": 0.2681884765625,
      "learning_rate": 2.823338644380566e-07,
      "loss": 0.0003,
      "reward": 1.4341518506407738,
      "reward_std": 0.1762135154567659,
      "rewards/equation_reward_func": 0.45758930779993534,
      "rewards/format_reward_func": 0.9765625186264515,
      "step": 214
    },
    {
      "completion_length": 262.88952255249023,
      "epoch": 0.268782081194587,
      "grad_norm": 0.06876739528559901,
      "kl": 0.1982421875,
      "learning_rate": 2.7875814660817504e-07,
      "loss": 0.0002,
      "reward": 1.4810268580913544,
      "reward_std": 0.19189182948321104,
      "rewards/equation_reward_func": 0.5044643133878708,
      "rewards/format_reward_func": 0.976562537252903,
      "step": 216
    },
    {
      "completion_length": 268.20090198516846,
      "epoch": 0.271270804168611,
      "grad_norm": 0.06507593507837049,
      "kl": 0.1932373046875,
      "learning_rate": 2.751764564986396e-07,
      "loss": 0.0002,
      "reward": 1.4252232760190964,
      "reward_std": 0.1531409532763064,
      "rewards/equation_reward_func": 0.4464285960420966,
      "rewards/format_reward_func": 0.9787946864962578,
      "step": 218
    },
    {
      "completion_length": 248.7288064956665,
      "epoch": 0.27375952714263496,
      "grad_norm": 0.04947068793737493,
      "kl": 0.22991943359375,
      "learning_rate": 2.715895379284194e-07,
      "loss": 0.0002,
      "reward": 1.4598215073347092,
      "reward_std": 0.17193997697904706,
      "rewards/equation_reward_func": 0.4776785857975483,
      "rewards/format_reward_func": 0.9821428768336773,
      "step": 220
    },
    {
      "completion_length": 250.73104000091553,
      "epoch": 0.2762482501166589,
      "grad_norm": 0.05521345272079342,
      "kl": 0.218994140625,
      "learning_rate": 2.6799813580229174e-07,
      "loss": 0.0002,
      "reward": 1.464285783469677,
      "reward_std": 0.14387341123074293,
      "rewards/equation_reward_func": 0.47991073690354824,
      "rewards/format_reward_func": 0.9843750186264515,
      "step": 222
    },
    {
      "completion_length": 250.93973922729492,
      "epoch": 0.27873697309068285,
      "grad_norm": 0.0483398884521067,
      "kl": 0.1998291015625,
      "learning_rate": 2.6440299595614606e-07,
      "loss": 0.0002,
      "reward": 1.4575893431901932,
      "reward_std": 0.1358907464891672,
      "rewards/equation_reward_func": 0.4720982280559838,
      "rewards/format_reward_func": 0.9854911118745804,
      "step": 224
    },
    {
      "completion_length": 255.92411518096924,
      "epoch": 0.2812256960647068,
      "grad_norm": 0.05969943458746344,
      "kl": 0.20379638671875,
      "learning_rate": 2.6080486500209347e-07,
      "loss": 0.0002,
      "reward": 1.4352679252624512,
      "reward_std": 0.15126124769449234,
      "rewards/equation_reward_func": 0.4542410960420966,
      "rewards/format_reward_func": 0.9810268133878708,
      "step": 226
    },
    {
      "completion_length": 242.58260345458984,
      "epoch": 0.28371441903873074,
      "grad_norm": 0.07222088082221652,
      "kl": 0.2523193359375,
      "learning_rate": 2.572044901734166e-07,
      "loss": 0.0003,
      "reward": 1.4486607760190964,
      "reward_std": 0.1779097393155098,
      "rewards/equation_reward_func": 0.4642857387661934,
      "rewards/format_reward_func": 0.9843750260770321,
      "step": 228
    },
    {
      "completion_length": 241.76117038726807,
      "epoch": 0.2862031420127547,
      "grad_norm": 0.06362399984743052,
      "kl": 0.2340087890625,
      "learning_rate": 2.536026191693893e-07,
      "loss": 0.0002,
      "reward": 1.4531250670552254,
      "reward_std": 0.16605611657723784,
      "rewards/equation_reward_func": 0.4743303768336773,
      "rewards/format_reward_func": 0.9787946715950966,
      "step": 230
    },
    {
      "completion_length": 232.6540288925171,
      "epoch": 0.2886918649867787,
      "grad_norm": 0.08974695425968056,
      "kl": 0.20977783203125,
      "learning_rate": 2.5e-07,
      "loss": 0.0002,
      "reward": 1.4899554252624512,
      "reward_std": 0.14485393464565277,
      "rewards/equation_reward_func": 0.5100446660071611,
      "rewards/format_reward_func": 0.9799107611179352,
      "step": 232
    },
    {
      "completion_length": 238.89398288726807,
      "epoch": 0.2911805879608026,
      "grad_norm": 0.056325369308057746,
      "kl": 0.2130126953125,
      "learning_rate": 2.4639738083061073e-07,
      "loss": 0.0002,
      "reward": 1.492187574505806,
      "reward_std": 0.1786569063551724,
      "rewards/equation_reward_func": 0.5200893115252256,
      "rewards/format_reward_func": 0.9720982536673546,
      "step": 234
    },
    {
      "completion_length": 275.3326053619385,
      "epoch": 0.29366931093482657,
      "grad_norm": 0.055885259957339305,
      "kl": 0.19952392578125,
      "learning_rate": 2.4279550982658345e-07,
      "loss": 0.0002,
      "reward": 1.3493304178118706,
      "reward_std": 0.19797838106751442,
      "rewards/equation_reward_func": 0.3794643059372902,
      "rewards/format_reward_func": 0.9698661006987095,
      "step": 236
    },
    {
      "completion_length": 248.7734498977661,
      "epoch": 0.29615803390885054,
      "grad_norm": 0.053249216235646074,
      "kl": 0.24993896484375,
      "learning_rate": 2.3919513499790646e-07,
      "loss": 0.0002,
      "reward": 1.4843750521540642,
      "reward_std": 0.20823450107127428,
      "rewards/equation_reward_func": 0.5189732387661934,
      "rewards/format_reward_func": 0.9654018133878708,
      "step": 238
    },
    {
      "completion_length": 274.2745695114136,
      "epoch": 0.29864675688287445,
      "grad_norm": 0.05354241002117662,
      "kl": 0.23797607421875,
      "learning_rate": 2.3559700404385394e-07,
      "loss": 0.0002,
      "reward": 1.3593750596046448,
      "reward_std": 0.18844150891527534,
      "rewards/equation_reward_func": 0.39732144959270954,
      "rewards/format_reward_func": 0.9620536006987095,
      "step": 240
    },
    {
      "completion_length": 278.19643783569336,
      "epoch": 0.3011354798568984,
      "grad_norm": 0.07271477039344029,
      "kl": 0.29461669921875,
      "learning_rate": 2.3200186419770823e-07,
      "loss": 0.0003,
      "reward": 1.3883929252624512,
      "reward_std": 0.2401469461619854,
      "rewards/equation_reward_func": 0.43526787869632244,
      "rewards/format_reward_func": 0.9531250335276127,
      "step": 242
    },
    {
      "completion_length": 240.8593873977661,
      "epoch": 0.3036242028309224,
      "grad_norm": 0.04915715622456297,
      "kl": 0.24176025390625,
      "learning_rate": 2.284104620715807e-07,
      "loss": 0.0002,
      "reward": 1.4765625670552254,
      "reward_std": 0.14837059983983636,
      "rewards/equation_reward_func": 0.5033482387661934,
      "rewards/format_reward_func": 0.9732143133878708,
      "step": 244
    },
    {
      "completion_length": 270.9040307998657,
      "epoch": 0.3061129258049463,
      "grad_norm": 0.07553072814564832,
      "kl": 0.22186279296875,
      "learning_rate": 2.2482354350136043e-07,
      "loss": 0.0002,
      "reward": 1.3917411267757416,
      "reward_std": 0.24061778374016285,
      "rewards/equation_reward_func": 0.4386160895228386,
      "rewards/format_reward_func": 0.9531250335276127,
      "step": 246
    },
    {
      "completion_length": 259.24108505249023,
      "epoch": 0.3086016487789703,
      "grad_norm": 0.06131285978491122,
      "kl": 0.22003173828125,
      "learning_rate": 2.2124185339182496e-07,
      "loss": 0.0002,
      "reward": 1.4140625596046448,
      "reward_std": 0.19331888854503632,
      "rewards/equation_reward_func": 0.44754466973245144,
      "rewards/format_reward_func": 0.9665178842842579,
      "step": 248
    },
    {
      "completion_length": 266.12836265563965,
      "epoch": 0.31109037175299425,
      "grad_norm": 0.057251447829503296,
      "kl": 0.211181640625,
      "learning_rate": 2.1766613556194344e-07,
      "loss": 0.0002,
      "reward": 1.431919701397419,
      "reward_std": 0.19034441327676177,
      "rewards/equation_reward_func": 0.4698660969734192,
      "rewards/format_reward_func": 0.9620536044239998,
      "step": 250
    },
    {
      "completion_length": 263.9140787124634,
      "epoch": 0.3135790947270182,
      "grad_norm": 0.06634536235536961,
      "kl": 0.22381591796875,
      "learning_rate": 2.1409713259040628e-07,
      "loss": 0.0002,
      "reward": 1.4631696939468384,
      "reward_std": 0.20181358978152275,
      "rewards/equation_reward_func": 0.49776787031441927,
      "rewards/format_reward_func": 0.9654018357396126,
      "step": 252
    },
    {
      "completion_length": 270.25447845458984,
      "epoch": 0.31606781770104214,
      "grad_norm": 0.08857117161748256,
      "kl": 0.21771240234375,
      "learning_rate": 2.105355856614115e-07,
      "loss": 0.0002,
      "reward": 1.408482201397419,
      "reward_std": 0.20398310013115406,
      "rewards/equation_reward_func": 0.4520089514553547,
      "rewards/format_reward_func": 0.9564732536673546,
      "step": 254
    },
    {
      "completion_length": 248.61050510406494,
      "epoch": 0.3185565406750661,
      "grad_norm": 0.05771718831012432,
      "kl": 0.21124267578125,
      "learning_rate": 2.069822344107413e-07,
      "loss": 0.0002,
      "reward": 1.4832590073347092,
      "reward_std": 0.2012131493538618,
      "rewards/equation_reward_func": 0.5267857331782579,
      "rewards/format_reward_func": 0.9564732499420643,
      "step": 256
    },
    {
      "completion_length": 265.1105041503906,
      "epoch": 0.3210452636490901,
      "grad_norm": 0.07137640163322823,
      "kl": 0.22540283203125,
      "learning_rate": 2.034378167721599e-07,
      "loss": 0.0002,
      "reward": 1.4263393580913544,
      "reward_std": 0.22233180236071348,
      "rewards/equation_reward_func": 0.47656252421438694,
      "rewards/format_reward_func": 0.9497768171131611,
      "step": 258
    },
    {
      "completion_length": 274.2232265472412,
      "epoch": 0.323533986623114,
      "grad_norm": 0.048682994811717506,
      "kl": 0.2178955078125,
      "learning_rate": 1.9990306882416485e-07,
      "loss": 0.0002,
      "reward": 1.3839286342263222,
      "reward_std": 0.15760588087141514,
      "rewards/equation_reward_func": 0.4285714477300644,
      "rewards/format_reward_func": 0.9553571715950966,
      "step": 260
    },
    {
      "completion_length": 264.10491943359375,
      "epoch": 0.32602270959713797,
      "grad_norm": 0.059864933831037394,
      "kl": 0.20758056640625,
      "learning_rate": 1.9637872463712362e-07,
      "loss": 0.0002,
      "reward": 1.4531250670552254,
      "reward_std": 0.2065706574358046,
      "rewards/equation_reward_func": 0.48883931152522564,
      "rewards/format_reward_func": 0.9642857536673546,
      "step": 262
    },
    {
      "completion_length": 242.52344799041748,
      "epoch": 0.32851143257116194,
      "grad_norm": 0.05470521183712939,
      "kl": 0.20501708984375,
      "learning_rate": 1.9286551612082773e-07,
      "loss": 0.0002,
      "reward": 1.5312500819563866,
      "reward_std": 0.14761846419423819,
      "rewards/equation_reward_func": 0.5636160988360643,
      "rewards/format_reward_func": 0.9676339514553547,
      "step": 264
    },
    {
      "completion_length": 257.9397439956665,
      "epoch": 0.33100015554518586,
      "grad_norm": 0.05856528610006056,
      "kl": 0.19775390625,
      "learning_rate": 1.8936417287249446e-07,
      "loss": 0.0002,
      "reward": 1.4542411416769028,
      "reward_std": 0.19231480779126287,
      "rewards/equation_reward_func": 0.4921875223517418,
      "rewards/format_reward_func": 0.9620536081492901,
      "step": 266
    },
    {
      "completion_length": 260.13952255249023,
      "epoch": 0.33348887851920983,
      "grad_norm": 0.06519402114882171,
      "kl": 0.20098876953125,
      "learning_rate": 1.8587542202524985e-07,
      "loss": 0.0002,
      "reward": 1.433035783469677,
      "reward_std": 0.1624811594374478,
      "rewards/equation_reward_func": 0.46093752421438694,
      "rewards/format_reward_func": 0.9720982499420643,
      "step": 268
    },
    {
      "completion_length": 241.5424222946167,
      "epoch": 0.3359776014932338,
      "grad_norm": 0.06174357121788231,
      "kl": 0.2017822265625,
      "learning_rate": 1.82399988097123e-07,
      "loss": 0.0002,
      "reward": 1.5156250670552254,
      "reward_std": 0.1582701876759529,
      "rewards/equation_reward_func": 0.5513393171131611,
      "rewards/format_reward_func": 0.9642857387661934,
      "step": 270
    },
    {
      "completion_length": 256.6517972946167,
      "epoch": 0.3384663244672577,
      "grad_norm": 0.038421184229451566,
      "kl": 0.20526123046875,
      "learning_rate": 1.7893859284058378e-07,
      "loss": 0.0002,
      "reward": 1.4754464998841286,
      "reward_std": 0.1418059840798378,
      "rewards/equation_reward_func": 0.5033482369035482,
      "rewards/format_reward_func": 0.9720982387661934,
      "step": 272
    },
    {
      "completion_length": 261.31139755249023,
      "epoch": 0.3409550474412817,
      "grad_norm": 0.06938416420126632,
      "kl": 0.19525146484375,
      "learning_rate": 1.7549195509265407e-07,
      "loss": 0.0002,
      "reward": 1.437500074505806,
      "reward_std": 0.15865246811881661,
      "rewards/equation_reward_func": 0.45758930779993534,
      "rewards/format_reward_func": 0.9799107275903225,
      "step": 274
    },
    {
      "completion_length": 237.3839406967163,
      "epoch": 0.34344377041530566,
      "grad_norm": 0.05279986019801015,
      "kl": 0.19287109375,
      "learning_rate": 1.7206079062562536e-07,
      "loss": 0.0002,
      "reward": 1.5078125670552254,
      "reward_std": 0.14753996301442385,
      "rewards/equation_reward_func": 0.5323660974390805,
      "rewards/format_reward_func": 0.9754464589059353,
      "step": 276
    },
    {
      "completion_length": 243.16295623779297,
      "epoch": 0.3459324933893296,
      "grad_norm": 0.07742881946994733,
      "kl": 0.19915771484375,
      "learning_rate": 1.6864581199841226e-07,
      "loss": 0.0002,
      "reward": 1.512276865541935,
      "reward_std": 0.18381929537281394,
      "rewards/equation_reward_func": 0.5345982387661934,
      "rewards/format_reward_func": 0.9776785969734192,
      "step": 278
    },
    {
      "completion_length": 267.8125104904175,
      "epoch": 0.34842121636335355,
      "grad_norm": 0.06365543898835445,
      "kl": 0.1842041015625,
      "learning_rate": 1.6524772840857388e-07,
      "loss": 0.0002,
      "reward": 1.381696492433548,
      "reward_std": 0.1500142039731145,
      "rewards/equation_reward_func": 0.40513394586741924,
      "rewards/format_reward_func": 0.9765625409781933,
      "step": 280
    },
    {
      "completion_length": 260.2701005935669,
      "epoch": 0.3509099393373775,
      "grad_norm": 0.06454931261369504,
      "kl": 0.19287109375,
      "learning_rate": 1.6186724554503237e-07,
      "loss": 0.0002,
      "reward": 1.4531250596046448,
      "reward_std": 0.16177314054220915,
      "rewards/equation_reward_func": 0.47433037776499987,
      "rewards/format_reward_func": 0.9787946753203869,
      "step": 282
    },
    {
      "completion_length": 264.14621353149414,
      "epoch": 0.35339866231140143,
      "grad_norm": 0.06827460094739726,
      "kl": 0.19805908203125,
      "learning_rate": 1.5850506544152103e-07,
      "loss": 0.0002,
      "reward": 1.444196492433548,
      "reward_std": 0.169928221963346,
      "rewards/equation_reward_func": 0.47098216973245144,
      "rewards/format_reward_func": 0.9732143208384514,
      "step": 284
    },
    {
      "completion_length": 267.815860748291,
      "epoch": 0.3558873852854254,
      "grad_norm": 0.07748655693200346,
      "kl": 0.196533203125,
      "learning_rate": 1.5516188633079107e-07,
      "loss": 0.0002,
      "reward": 1.4006696864962578,
      "reward_std": 0.18472299817949533,
      "rewards/equation_reward_func": 0.42968752048909664,
      "rewards/format_reward_func": 0.9709821790456772,
      "step": 286
    },
    {
      "completion_length": 252.82367324829102,
      "epoch": 0.3583761082594494,
      "grad_norm": 0.06854165504600304,
      "kl": 0.20758056640625,
      "learning_rate": 1.5183840249960784e-07,
      "loss": 0.0002,
      "reward": 1.5178572162985802,
      "reward_std": 0.19481780100613832,
      "rewards/equation_reward_func": 0.5491071660071611,
      "rewards/format_reward_func": 0.9687500298023224,
      "step": 288
    },
    {
      "completion_length": 247.53460693359375,
      "epoch": 0.36086483123347335,
      "grad_norm": 0.057981171534578976,
      "kl": 0.20849609375,
      "learning_rate": 1.4853530414456612e-07,
      "loss": 0.0002,
      "reward": 1.4799107685685158,
      "reward_std": 0.1738822301849723,
      "rewards/equation_reward_func": 0.506696448661387,
      "rewards/format_reward_func": 0.9732143208384514,
      "step": 290
    },
    {
      "completion_length": 228.368314743042,
      "epoch": 0.36335355420749726,
      "grad_norm": 0.049516342449076566,
      "kl": 0.2069091796875,
      "learning_rate": 1.4525327722875568e-07,
      "loss": 0.0002,
      "reward": 1.4899554327130318,
      "reward_std": 0.13066139770671725,
      "rewards/equation_reward_func": 0.5133928805589676,
      "rewards/format_reward_func": 0.976562537252903,
      "step": 292
    },
    {
      "completion_length": 251.85492134094238,
      "epoch": 0.36584227718152124,
      "grad_norm": 0.061436014992367165,
      "kl": 0.1915283203125,
      "learning_rate": 1.4199300333930515e-07,
      "loss": 0.0002,
      "reward": 1.4508929178118706,
      "reward_std": 0.13965235324576497,
      "rewards/equation_reward_func": 0.4665178805589676,
      "rewards/format_reward_func": 0.9843750447034836,
      "step": 294
    },
    {
      "completion_length": 237.25224208831787,
      "epoch": 0.3683310001555452,
      "grad_norm": 0.061911720113752254,
      "kl": 0.21185302734375,
      "learning_rate": 1.3875515954583523e-07,
      "loss": 0.0002,
      "reward": 1.5156250670552254,
      "reward_std": 0.17848694045096636,
      "rewards/equation_reward_func": 0.5401785969734192,
      "rewards/format_reward_func": 0.9754464663565159,
      "step": 296
    },
    {
      "completion_length": 260.84599208831787,
      "epoch": 0.3708197231295691,
      "grad_norm": 0.06594279646081479,
      "kl": 0.19842529296875,
      "learning_rate": 1.3554041825985e-07,
      "loss": 0.0002,
      "reward": 1.4151786342263222,
      "reward_std": 0.20021545700728893,
      "rewards/equation_reward_func": 0.4464285932481289,
      "rewards/format_reward_func": 0.9687500186264515,
      "step": 298
    },
    {
      "completion_length": 260.6317090988159,
      "epoch": 0.3733084461035931,
      "grad_norm": 0.06036144065480427,
      "kl": 0.19525146484375,
      "learning_rate": 1.323494470950949e-07,
      "loss": 0.0002,
      "reward": 1.4419643506407738,
      "reward_std": 0.1797226625494659,
      "rewards/equation_reward_func": 0.47098216600716114,
      "rewards/format_reward_func": 0.9709821790456772,
      "step": 300
    },
    {
      "completion_length": 241.9788055419922,
      "epoch": 0.37579716907761707,
      "grad_norm": 0.05826454315321918,
      "kl": 0.19757080078125,
      "learning_rate": 1.2918290872891236e-07,
      "loss": 0.0002,
      "reward": 1.4810268506407738,
      "reward_std": 0.13545973878353834,
      "rewards/equation_reward_func": 0.5011160913854837,
      "rewards/format_reward_func": 0.9799107499420643,
      "step": 302
    },
    {
      "completion_length": 257.9218854904175,
      "epoch": 0.378285892051641,
      "grad_norm": 0.08618938040431284,
      "kl": 0.20965576171875,
      "learning_rate": 1.260414607646213e-07,
      "loss": 0.0002,
      "reward": 1.4888393506407738,
      "reward_std": 0.17534363875165582,
      "rewards/equation_reward_func": 0.5122768077999353,
      "rewards/format_reward_func": 0.9765625223517418,
      "step": 304
    },
    {
      "completion_length": 252.98550128936768,
      "epoch": 0.38077461502566495,
      "grad_norm": 0.06784584776225837,
      "kl": 0.2144775390625,
      "learning_rate": 1.2292575559495143e-07,
      "loss": 0.0002,
      "reward": 1.4765625670552254,
      "reward_std": 0.18450129078701138,
      "rewards/equation_reward_func": 0.5066964533179998,
      "rewards/format_reward_func": 0.9698661006987095,
      "step": 306
    },
    {
      "completion_length": 271.24443435668945,
      "epoch": 0.3832633379996889,
      "grad_norm": 0.036854119619471624,
      "kl": 0.19024658203125,
      "learning_rate": 1.1983644026655835e-07,
      "loss": 0.0002,
      "reward": 1.4185268580913544,
      "reward_std": 0.123422771692276,
      "rewards/equation_reward_func": 0.439732164144516,
      "rewards/format_reward_func": 0.9787946715950966,
      "step": 308
    },
    {
      "completion_length": 263.0301446914673,
      "epoch": 0.38575206097371284,
      "grad_norm": 0.06348479835149756,
      "kl": 0.1917724609375,
      "learning_rate": 1.1677415634565066e-07,
      "loss": 0.0002,
      "reward": 1.4821429252624512,
      "reward_std": 0.17120425822213292,
      "rewards/equation_reward_func": 0.5011160876601934,
      "rewards/format_reward_func": 0.9810268245637417,
      "step": 310
    },
    {
      "completion_length": 271.0122890472412,
      "epoch": 0.3882407839477368,
      "grad_norm": 0.06288609866959954,
      "kl": 0.25067138671875,
      "learning_rate": 1.1373953978475353e-07,
      "loss": 0.0003,
      "reward": 1.4196429252624512,
      "reward_std": 0.1885003875941038,
      "rewards/equation_reward_func": 0.4575893102446571,
      "rewards/format_reward_func": 0.9620535932481289,
      "step": 312
    },
    {
      "completion_length": 270.0948781967163,
      "epoch": 0.3907295069217608,
      "grad_norm": 0.05814773664760414,
      "kl": 0.18487548828125,
      "learning_rate": 1.1073322079063913e-07,
      "loss": 0.0002,
      "reward": 1.4341518431901932,
      "reward_std": 0.17814994137734175,
      "rewards/equation_reward_func": 0.45424109511077404,
      "rewards/format_reward_func": 0.9799107536673546,
      "step": 314
    },
    {
      "completion_length": 248.2522430419922,
      "epoch": 0.3932182298957847,
      "grad_norm": 0.0693174506427365,
      "kl": 0.18951416015625,
      "learning_rate": 1.0775582369344946e-07,
      "loss": 0.0002,
      "reward": 1.4531250596046448,
      "reward_std": 0.18309945985674858,
      "rewards/equation_reward_func": 0.48214287776499987,
      "rewards/format_reward_func": 0.9709821790456772,
      "step": 316
    },
    {
      "completion_length": 241.57367038726807,
      "epoch": 0.39570695286980867,
      "grad_norm": 0.07528066976062558,
      "kl": 0.2047119140625,
      "learning_rate": 1.0480796681704077e-07,
      "loss": 0.0002,
      "reward": 1.4877232685685158,
      "reward_std": 0.15524994302541018,
      "rewards/equation_reward_func": 0.5122768115252256,
      "rewards/format_reward_func": 0.9754464663565159,
      "step": 318
    },
    {
      "completion_length": 239.8069314956665,
      "epoch": 0.39819567584383264,
      "grad_norm": 0.05012947107458731,
      "kl": 0.20562744140625,
      "learning_rate": 1.018902623505741e-07,
      "loss": 0.0002,
      "reward": 1.4888393580913544,
      "reward_std": 0.12663634540513158,
      "rewards/equation_reward_func": 0.5100446697324514,
      "rewards/format_reward_func": 0.9787946827709675,
      "step": 320
    },
    {
      "completion_length": 239.58148384094238,
      "epoch": 0.4006843988178566,
      "grad_norm": 0.0628656210330094,
      "kl": 0.213134765625,
      "learning_rate": 9.900331622138063e-08,
      "loss": 0.0002,
      "reward": 1.5000000596046448,
      "reward_std": 0.1728154099546373,
      "rewards/equation_reward_func": 0.525669670663774,
      "rewards/format_reward_func": 0.9743303954601288,
      "step": 322
    },
    {
      "completion_length": 245.7187614440918,
      "epoch": 0.40317312179188053,
      "grad_norm": 0.06489486125976161,
      "kl": 0.20806884765625,
      "learning_rate": 9.614772796912681e-08,
      "loss": 0.0002,
      "reward": 1.428571492433548,
      "reward_std": 0.1652527879923582,
      "rewards/equation_reward_func": 0.4531250214204192,
      "rewards/format_reward_func": 0.9754464589059353,
      "step": 324
    },
    {
      "completion_length": 224.60268878936768,
      "epoch": 0.4056618447659045,
      "grad_norm": 0.0573415983961628,
      "kl": 0.2230224609375,
      "learning_rate": 9.332409062130686e-08,
      "loss": 0.0002,
      "reward": 1.564732201397419,
      "reward_std": 0.14763088943436742,
      "rewards/equation_reward_func": 0.5792410932481289,
      "rewards/format_reward_func": 0.9854911044239998,
      "step": 326
    },
    {
      "completion_length": 240.5245656967163,
      "epoch": 0.40815056773992847,
      "grad_norm": 0.06956094753946698,
      "kl": 0.1995849609375,
      "learning_rate": 9.053299057008699e-08,
      "loss": 0.0002,
      "reward": 1.4330357909202576,
      "reward_std": 0.15731613663956523,
      "rewards/equation_reward_func": 0.4520089477300644,
      "rewards/format_reward_func": 0.9810268096625805,
      "step": 328
    },
    {
      "completion_length": 270.4955463409424,
      "epoch": 0.4106392907139524,
      "grad_norm": 0.07509531278447272,
      "kl": 0.19549560546875,
      "learning_rate": 8.777500745052743e-08,
      "loss": 0.0002,
      "reward": 1.3649554327130318,
      "reward_std": 0.17564317397773266,
      "rewards/equation_reward_func": 0.3984375214204192,
      "rewards/format_reward_func": 0.9665178805589676,
      "step": 330
    },
    {
      "completion_length": 266.61942863464355,
      "epoch": 0.41312801368797636,
      "grad_norm": 0.058100347807015534,
      "kl": 0.2001953125,
      "learning_rate": 8.505071402020892e-08,
      "loss": 0.0002,
      "reward": 1.350446492433548,
      "reward_std": 0.17841344932094216,
      "rewards/equation_reward_func": 0.38504466228187084,
      "rewards/format_reward_func": 0.965401828289032,
      "step": 332
    },
    {
      "completion_length": 248.6127347946167,
      "epoch": 0.41561673666200033,
      "grad_norm": 0.05834481650049284,
      "kl": 0.198974609375,
      "learning_rate": 8.236067604028562e-08,
      "loss": 0.0002,
      "reward": 1.4609375670552254,
      "reward_std": 0.17589116655290127,
      "rewards/equation_reward_func": 0.5022321678698063,
      "rewards/format_reward_func": 0.9587053917348385,
      "step": 334
    },
    {
      "completion_length": 247.92746686935425,
      "epoch": 0.41810545963602425,
      "grad_norm": 0.0591341402478618,
      "kl": 0.21661376953125,
      "learning_rate": 7.970545215799327e-08,
      "loss": 0.0002,
      "reward": 1.4944197162985802,
      "reward_std": 0.17136807506904006,
      "rewards/equation_reward_func": 0.5212053768336773,
      "rewards/format_reward_func": 0.9732143208384514,
      "step": 336
    },
    {
      "completion_length": 275.84488105773926,
      "epoch": 0.4205941826100482,
      "grad_norm": 0.04884352625738446,
      "kl": 0.20855712890625,
      "learning_rate": 7.708559379063204e-08,
      "loss": 0.0002,
      "reward": 1.3694196976721287,
      "reward_std": 0.17837184108793736,
      "rewards/equation_reward_func": 0.4107143059372902,
      "rewards/format_reward_func": 0.9587053880095482,
      "step": 338
    },
    {
      "completion_length": 243.42411851882935,
      "epoch": 0.4230829055840722,
      "grad_norm": 0.06899085908963953,
      "kl": 48.954345703125,
      "learning_rate": 7.45016450110534e-08,
      "loss": 0.0489,
      "reward": 1.4654018580913544,
      "reward_std": 0.1875977530144155,
      "rewards/equation_reward_func": 0.4933036006987095,
      "rewards/format_reward_func": 0.9720982499420643,
      "step": 340
    },
    {
      "completion_length": 265.886173248291,
      "epoch": 0.4255716285580961,
      "grad_norm": 0.05980414020403841,
      "kl": 0.19793701171875,
      "learning_rate": 7.195414243467029e-08,
      "loss": 0.0002,
      "reward": 1.3984375596046448,
      "reward_std": 0.175481291487813,
      "rewards/equation_reward_func": 0.43750001955777407,
      "rewards/format_reward_func": 0.9609375335276127,
      "step": 342
    },
    {
      "completion_length": 255.4665298461914,
      "epoch": 0.4280603515321201,
      "grad_norm": 0.05974337860010266,
      "kl": 0.197998046875,
      "learning_rate": 6.944361510801763e-08,
      "loss": 0.0002,
      "reward": 1.4419643431901932,
      "reward_std": 0.1873517008498311,
      "rewards/equation_reward_func": 0.4787946669384837,
      "rewards/format_reward_func": 0.9631696790456772,
      "step": 344
    },
    {
      "completion_length": 237.57590103149414,
      "epoch": 0.43054907450614405,
      "grad_norm": 0.04126838575736974,
      "kl": 0.19622802734375,
      "learning_rate": 6.697058439888283e-08,
      "loss": 0.0002,
      "reward": 1.4720982760190964,
      "reward_std": 0.14282790152356029,
      "rewards/equation_reward_func": 0.4866071632131934,
      "rewards/format_reward_func": 0.9854910857975483,
      "step": 346
    },
    {
      "completion_length": 245.25224494934082,
      "epoch": 0.43303779748016796,
      "grad_norm": 0.08845105411407385,
      "kl": 0.1953125,
      "learning_rate": 6.453556388803288e-08,
      "loss": 0.0002,
      "reward": 1.4330357685685158,
      "reward_std": 0.17873004684224725,
      "rewards/equation_reward_func": 0.45535716973245144,
      "rewards/format_reward_func": 0.9776786044239998,
      "step": 348
    },
    {
      "completion_length": 259.90849781036377,
      "epoch": 0.43552652045419193,
      "grad_norm": 0.06619651517208977,
      "kl": 0.20709228515625,
      "learning_rate": 6.213905926255697e-08,
      "loss": 0.0002,
      "reward": 1.38839291036129,
      "reward_std": 0.17326576448976994,
      "rewards/equation_reward_func": 0.4196428721770644,
      "rewards/format_reward_func": 0.9687500298023224,
      "step": 350
    },
    {
      "completion_length": 246.68862628936768,
      "epoch": 0.4380152434282159,
      "grad_norm": 0.07466580623927803,
      "kl": 0.21270751953125,
      "learning_rate": 5.978156821084987e-08,
      "loss": 0.0002,
      "reward": 1.4709822162985802,
      "reward_std": 0.17257908964529634,
      "rewards/equation_reward_func": 0.49776787869632244,
      "rewards/format_reward_func": 0.973214328289032,
      "step": 352
    },
    {
      "completion_length": 238.16072463989258,
      "epoch": 0.4405039664022399,
      "grad_norm": 0.05555433456578972,
      "kl": 0.20025634765625,
      "learning_rate": 5.7463580319254853e-08,
      "loss": 0.0002,
      "reward": 1.464285783469677,
      "reward_std": 0.18330077826976776,
      "rewards/equation_reward_func": 0.4821428805589676,
      "rewards/format_reward_func": 0.9821428917348385,
      "step": 354
    },
    {
      "completion_length": 238.5982265472412,
      "epoch": 0.4429926893762638,
      "grad_norm": 0.10010877276744827,
      "kl": 0.20880126953125,
      "learning_rate": 5.518557697039081e-08,
      "loss": 0.0002,
      "reward": 1.5022322162985802,
      "reward_std": 0.15463243331760168,
      "rewards/equation_reward_func": 0.5234375204890966,
      "rewards/format_reward_func": 0.9787946604192257,
      "step": 356
    },
    {
      "completion_length": 235.23773384094238,
      "epoch": 0.44548141235028776,
      "grad_norm": 0.05922980250758421,
      "kl": 0.208984375,
      "learning_rate": 5.294803124318145e-08,
      "loss": 0.0002,
      "reward": 1.512276865541935,
      "reward_std": 0.15547437546774745,
      "rewards/equation_reward_func": 0.5323661006987095,
      "rewards/format_reward_func": 0.9799107387661934,
      "step": 358
    },
    {
      "completion_length": 247.35046005249023,
      "epoch": 0.44797013532431174,
      "grad_norm": 0.040689373027519675,
      "kl": 0.19696044921875,
      "learning_rate": 5.07514078146106e-08,
      "loss": 0.0002,
      "reward": 1.4263393357396126,
      "reward_std": 0.1481092283502221,
      "rewards/equation_reward_func": 0.454241088591516,
      "rewards/format_reward_func": 0.9720982536673546,
      "step": 360
    },
    {
      "completion_length": 242.27567958831787,
      "epoch": 0.45045885829833565,
      "grad_norm": 0.07393799743280928,
      "kl": 0.20159912109375,
      "learning_rate": 4.859616286322094e-08,
      "loss": 0.0002,
      "reward": 1.4620536342263222,
      "reward_std": 0.1555815995670855,
      "rewards/equation_reward_func": 0.47656251676380634,
      "rewards/format_reward_func": 0.9854910932481289,
      "step": 362
    },
    {
      "completion_length": 248.84710788726807,
      "epoch": 0.4529475812723596,
      "grad_norm": 0.04906536376122032,
      "kl": 0.20123291015625,
      "learning_rate": 4.648274397437829e-08,
      "loss": 0.0002,
      "reward": 1.4654018506407738,
      "reward_std": 0.1682790988124907,
      "rewards/equation_reward_func": 0.49218752793967724,
      "rewards/format_reward_func": 0.9732143133878708,
      "step": 364
    },
    {
      "completion_length": 245.94532299041748,
      "epoch": 0.4554363042463836,
      "grad_norm": 0.05802144393054711,
      "kl": 0.19244384765625,
      "learning_rate": 4.4411590047320617e-08,
      "loss": 0.0002,
      "reward": 1.4464286342263222,
      "reward_std": 0.15317056560888886,
      "rewards/equation_reward_func": 0.4676339514553547,
      "rewards/format_reward_func": 0.9787946604192257,
      "step": 366
    },
    {
      "completion_length": 246.37612628936768,
      "epoch": 0.4579250272204075,
      "grad_norm": 0.060239468744741904,
      "kl": 0.20068359375,
      "learning_rate": 4.2383131204010494e-08,
      "loss": 0.0002,
      "reward": 1.422991156578064,
      "reward_std": 0.15117083815857768,
      "rewards/equation_reward_func": 0.45424109790474176,
      "rewards/format_reward_func": 0.9687500335276127,
      "step": 368
    },
    {
      "completion_length": 242.1529130935669,
      "epoch": 0.4604137501944315,
      "grad_norm": 0.04267260387428845,
      "kl": 0.19769287109375,
      "learning_rate": 4.039778869981064e-08,
      "loss": 0.0002,
      "reward": 1.4732143506407738,
      "reward_std": 0.12463709712028503,
      "rewards/equation_reward_func": 0.49107145331799984,
      "rewards/format_reward_func": 0.9821428842842579,
      "step": 370
    },
    {
      "completion_length": 260.5067090988159,
      "epoch": 0.46290247316845545,
      "grad_norm": 0.0544495720289963,
      "kl": 0.199462890625,
      "learning_rate": 3.845597483600049e-08,
      "loss": 0.0002,
      "reward": 1.4017857909202576,
      "reward_std": 0.1715238355100155,
      "rewards/equation_reward_func": 0.4386160932481289,
      "rewards/format_reward_func": 0.9631696753203869,
      "step": 372
    },
    {
      "completion_length": 253.75782299041748,
      "epoch": 0.46539119614247937,
      "grad_norm": 0.05723699687752985,
      "kl": 0.2158203125,
      "learning_rate": 3.655809287415284e-08,
      "loss": 0.0002,
      "reward": 1.3939732871949673,
      "reward_std": 0.17196187004446983,
      "rewards/equation_reward_func": 0.4207589514553547,
      "rewards/format_reward_func": 0.9732143245637417,
      "step": 374
    },
    {
      "completion_length": 259.69532108306885,
      "epoch": 0.46787991911650334,
      "grad_norm": 0.05101025382090558,
      "kl": 0.1917724609375,
      "learning_rate": 3.4704536952387285e-08,
      "loss": 0.0002,
      "reward": 1.3772322088479996,
      "reward_std": 0.17750998865813017,
      "rewards/equation_reward_func": 0.40513394703157246,
      "rewards/format_reward_func": 0.9720982387661934,
      "step": 376
    },
    {
      "completion_length": 236.41072368621826,
      "epoch": 0.4703686420905273,
      "grad_norm": 0.07195347219746893,
      "kl": 0.1944580078125,
      "learning_rate": 3.2895692003518575e-08,
      "loss": 0.0002,
      "reward": 1.4944197162985802,
      "reward_std": 0.17748092440888286,
      "rewards/equation_reward_func": 0.5200893096625805,
      "rewards/format_reward_func": 0.9743303954601288,
      "step": 378
    },
    {
      "completion_length": 259.48438835144043,
      "epoch": 0.4728573650645512,
      "grad_norm": 0.07172484472660048,
      "kl": 0.19696044921875,
      "learning_rate": 3.113193367511635e-08,
      "loss": 0.0002,
      "reward": 1.3895090073347092,
      "reward_std": 0.1647741007618606,
      "rewards/equation_reward_func": 0.41852680779993534,
      "rewards/format_reward_func": 0.9709821753203869,
      "step": 380
    },
    {
      "completion_length": 245.80804824829102,
      "epoch": 0.4753460880385752,
      "grad_norm": 0.04427418304284531,
      "kl": 0.20074462890625,
      "learning_rate": 2.9413628251493934e-08,
      "loss": 0.0002,
      "reward": 1.4587054327130318,
      "reward_std": 0.15418589767068624,
      "rewards/equation_reward_func": 0.4888393096625805,
      "rewards/format_reward_func": 0.9698661081492901,
      "step": 382
    },
    {
      "completion_length": 241.82032108306885,
      "epoch": 0.47783481101259917,
      "grad_norm": 0.05413458692040878,
      "kl": 0.19281005859375,
      "learning_rate": 2.774113257764066e-08,
      "loss": 0.0002,
      "reward": 1.4654018580913544,
      "reward_std": 0.1285329912789166,
      "rewards/equation_reward_func": 0.48995537497103214,
      "rewards/format_reward_func": 0.975446455180645,
      "step": 384
    },
    {
      "completion_length": 235.2399663925171,
      "epoch": 0.48032353398662314,
      "grad_norm": 0.07013851232759652,
      "kl": 0.20245361328125,
      "learning_rate": 2.611479398511518e-08,
      "loss": 0.0002,
      "reward": 1.439732201397419,
      "reward_std": 0.18071282655000687,
      "rewards/equation_reward_func": 0.4687500251457095,
      "rewards/format_reward_func": 0.9709821790456772,
      "step": 386
    },
    {
      "completion_length": 253.3303689956665,
      "epoch": 0.48281225696064706,
      "grad_norm": 0.05817044023141003,
      "kl": 0.19342041015625,
      "learning_rate": 2.4534950219914057e-08,
      "loss": 0.0002,
      "reward": 1.3939732909202576,
      "reward_std": 0.1655187914147973,
      "rewards/equation_reward_func": 0.4185268059372902,
      "rewards/format_reward_func": 0.9754464738070965,
      "step": 388
    },
    {
      "completion_length": 220.24219703674316,
      "epoch": 0.48530097993467103,
      "grad_norm": 0.06057036471445671,
      "kl": 0.2003173828125,
      "learning_rate": 2.300192937233128e-08,
      "loss": 0.0002,
      "reward": 1.4854911267757416,
      "reward_std": 0.1361873117275536,
      "rewards/equation_reward_func": 0.49888395704329014,
      "rewards/format_reward_func": 0.9866071678698063,
      "step": 390
    },
    {
      "completion_length": 238.94085693359375,
      "epoch": 0.487789702908695,
      "grad_norm": 0.059472704194779645,
      "kl": 0.229736328125,
      "learning_rate": 2.1516049808822935e-08,
      "loss": 0.0002,
      "reward": 1.4229911342263222,
      "reward_std": 0.19178591016680002,
      "rewards/equation_reward_func": 0.4486607378348708,
      "rewards/format_reward_func": 0.9743303880095482,
      "step": 392
    },
    {
      "completion_length": 214.02791023254395,
      "epoch": 0.4902784258827189,
      "grad_norm": 0.04800855566257019,
      "kl": 0.21405029296875,
      "learning_rate": 2.007762010589098e-08,
      "loss": 0.0002,
      "reward": 1.506696492433548,
      "reward_std": 0.09478258900344372,
      "rewards/equation_reward_func": 0.5245536025613546,
      "rewards/format_reward_func": 0.9821428768336773,
      "step": 394
    },
    {
      "completion_length": 247.6830472946167,
      "epoch": 0.4927671488567429,
      "grad_norm": 0.07541466310359685,
      "kl": 0.2535400390625,
      "learning_rate": 1.8686938986000627e-08,
      "loss": 0.0003,
      "reward": 1.3705357685685158,
      "reward_std": 0.1484149764291942,
      "rewards/equation_reward_func": 0.39174109196756035,
      "rewards/format_reward_func": 0.9787946715950966,
      "step": 396
    },
    {
      "completion_length": 239.485502243042,
      "epoch": 0.49525587183076686,
      "grad_norm": 0.05438411307581503,
      "kl": 0.208740234375,
      "learning_rate": 1.734429525554365e-08,
      "loss": 0.0002,
      "reward": 1.4174107760190964,
      "reward_std": 0.15924348449334502,
      "rewards/equation_reward_func": 0.4397321632131934,
      "rewards/format_reward_func": 0.9776785932481289,
      "step": 398
    },
    {
      "completion_length": 240.2142972946167,
      "epoch": 0.4977445948047908,
      "grad_norm": 0.05085143905737969,
      "kl": 0.2021484375,
      "learning_rate": 1.604996774486145e-08,
      "loss": 0.0002,
      "reward": 1.389508992433548,
      "reward_std": 0.1306094191968441,
      "rewards/equation_reward_func": 0.407366088591516,
      "rewards/format_reward_func": 0.9821428805589676,
      "step": 400
    },
    {
      "completion_length": 243.95648193359375,
      "epoch": 0.5002333177788147,
      "grad_norm": 0.05035017375480962,
      "kl": 0.2027587890625,
      "learning_rate": 1.4804225250339281e-08,
      "loss": 0.0002,
      "reward": 1.4475447162985802,
      "reward_std": 0.16129027446731925,
      "rewards/equation_reward_func": 0.478794664144516,
      "rewards/format_reward_func": 0.9687500335276127,
      "step": 402
    },
    {
      "completion_length": 237.8091640472412,
      "epoch": 0.5027220407528387,
      "grad_norm": 0.054861170642348474,
      "kl": 0.1966552734375,
      "learning_rate": 1.360732647858498e-08,
      "loss": 0.0002,
      "reward": 1.4263393431901932,
      "reward_std": 0.1515321801416576,
      "rewards/equation_reward_func": 0.44754466880112886,
      "rewards/format_reward_func": 0.9787946715950966,
      "step": 404
    },
    {
      "completion_length": 219.90625953674316,
      "epoch": 0.5052107637268627,
      "grad_norm": 0.07477628715165045,
      "kl": 0.21026611328125,
      "learning_rate": 1.2459519992702311e-08,
      "loss": 0.0002,
      "reward": 1.5011161491274834,
      "reward_std": 0.1352310930378735,
      "rewards/equation_reward_func": 0.5145089495927095,
      "rewards/format_reward_func": 0.9866071715950966,
      "step": 406
    },
    {
      "completion_length": 255.5993413925171,
      "epoch": 0.5076994867008866,
      "grad_norm": 0.05849730835576962,
      "kl": 0.1883544921875,
      "learning_rate": 1.1361044160671629e-08,
      "loss": 0.0002,
      "reward": 1.3504464849829674,
      "reward_std": 0.14734927052631974,
      "rewards/equation_reward_func": 0.37834823317825794,
      "rewards/format_reward_func": 0.972098246216774,
      "step": 408
    },
    {
      "completion_length": 223.27344799041748,
      "epoch": 0.5101882096749105,
      "grad_norm": 0.06665066337341637,
      "kl": 0.20648193359375,
      "learning_rate": 1.0312127105846947e-08,
      "loss": 0.0002,
      "reward": 1.4966518431901932,
      "reward_std": 0.15878778649494052,
      "rewards/equation_reward_func": 0.5122768119908869,
      "rewards/format_reward_func": 0.9843750409781933,
      "step": 410
    },
    {
      "completion_length": 234.1026906967163,
      "epoch": 0.5126769326489345,
      "grad_norm": 0.05933115040400525,
      "kl": 0.204345703125,
      "learning_rate": 9.312986659581301e-09,
      "loss": 0.0002,
      "reward": 1.4542411416769028,
      "reward_std": 0.17036408884450793,
      "rewards/equation_reward_func": 0.47321429941803217,
      "rewards/format_reward_func": 0.9810268096625805,
      "step": 412
    },
    {
      "completion_length": 231.8727788925171,
      "epoch": 0.5151656556229585,
      "grad_norm": 0.06260514018048406,
      "kl": 0.2193603515625,
      "learning_rate": 8.363830315988945e-09,
      "loss": 0.0002,
      "reward": 1.4832590073347092,
      "reward_std": 0.1522468039765954,
      "rewards/equation_reward_func": 0.4988839514553547,
      "rewards/format_reward_func": 0.9843750335276127,
      "step": 414
    },
    {
      "completion_length": 240.4453239440918,
      "epoch": 0.5176543785969824,
      "grad_norm": 0.034718644688255154,
      "kl": 0.20037841796875,
      "learning_rate": 7.46485518885462e-09,
      "loss": 0.0002,
      "reward": 1.4151786491274834,
      "reward_std": 0.13582705985754728,
      "rewards/equation_reward_func": 0.43750002048909664,
      "rewards/format_reward_func": 0.9776785969734192,
      "step": 416
    },
    {
      "completion_length": 227.1093864440918,
      "epoch": 0.5201431015710064,
      "grad_norm": 0.05669666354887382,
      "kl": 0.234375,
      "learning_rate": 6.616247970698319e-09,
      "loss": 0.0002,
      "reward": 1.4988840073347092,
      "reward_std": 0.13698529778048396,
      "rewards/equation_reward_func": 0.5189732406288385,
      "rewards/format_reward_func": 0.9799107424914837,
      "step": 418
    },
    {
      "completion_length": 262.71653270721436,
      "epoch": 0.5226318245450303,
      "grad_norm": 0.054401514129177465,
      "kl": 0.18310546875,
      "learning_rate": 5.8181848940044855e-09,
      "loss": 0.0002,
      "reward": 1.3281250670552254,
      "reward_std": 0.1366122462786734,
      "rewards/equation_reward_func": 0.34375001303851604,
      "rewards/format_reward_func": 0.9843750298023224,
      "step": 420
    },
    {
      "completion_length": 235.43639755249023,
      "epoch": 0.5251205475190542,
      "grad_norm": 0.07834852740210009,
      "kl": 0.21807861328125,
      "learning_rate": 5.070831694623135e-09,
      "loss": 0.0002,
      "reward": 1.470982201397419,
      "reward_std": 0.14967900328338146,
      "rewards/equation_reward_func": 0.48995538521558046,
      "rewards/format_reward_func": 0.9810268096625805,
      "step": 422
    },
    {
      "completion_length": 250.70983409881592,
      "epoch": 0.5276092704930783,
      "grad_norm": 0.06584760953247815,
      "kl": 0.19866943359375,
      "learning_rate": 4.374343577351336e-09,
      "loss": 0.0002,
      "reward": 1.416294701397419,
      "reward_std": 0.18149724043905735,
      "rewards/equation_reward_func": 0.44866074062883854,
      "rewards/format_reward_func": 0.9676339700818062,
      "step": 424
    },
    {
      "completion_length": 242.2321538925171,
      "epoch": 0.5300979934671022,
      "grad_norm": 0.05239977644165029,
      "kl": 0.2098388671875,
      "learning_rate": 3.7288651837012745e-09,
      "loss": 0.0002,
      "reward": 1.4330357685685158,
      "reward_std": 0.16506521217525005,
      "rewards/equation_reward_func": 0.45312501955777407,
      "rewards/format_reward_func": 0.979910746216774,
      "step": 426
    },
    {
      "completion_length": 215.3493413925171,
      "epoch": 0.5325867164411261,
      "grad_norm": 0.08270584148740566,
      "kl": 0.202392578125,
      "learning_rate": 3.134530561862081e-09,
      "loss": 0.0002,
      "reward": 1.5446429327130318,
      "reward_std": 0.14123681280761957,
      "rewards/equation_reward_func": 0.5558035969734192,
      "rewards/format_reward_func": 0.9888393133878708,
      "step": 428
    },
    {
      "completion_length": 241.829252243042,
      "epoch": 0.5350754394151501,
      "grad_norm": 0.054858889512319424,
      "kl": 0.1895751953125,
      "learning_rate": 2.5914631388619103e-09,
      "loss": 0.0002,
      "reward": 1.4241072088479996,
      "reward_std": 0.10743419965729117,
      "rewards/equation_reward_func": 0.43750002048909664,
      "rewards/format_reward_func": 0.9866071715950966,
      "step": 430
    },
    {
      "completion_length": 227.63059043884277,
      "epoch": 0.537564162389174,
      "grad_norm": 0.06298373988036782,
      "kl": 0.2347412109375,
      "learning_rate": 2.0997756949353297e-09,
      "loss": 0.0002,
      "reward": 1.4654018580913544,
      "reward_std": 0.13623460568487644,
      "rewards/equation_reward_func": 0.482142879627645,
      "rewards/format_reward_func": 0.983258955180645,
      "step": 432
    },
    {
      "completion_length": 231.58482933044434,
      "epoch": 0.540052885363198,
      "grad_norm": 0.05735572859329674,
      "kl": 0.241455078125,
      "learning_rate": 1.6595703401020844e-09,
      "loss": 0.0002,
      "reward": 1.487723283469677,
      "reward_std": 0.1775469919666648,
      "rewards/equation_reward_func": 0.5156250279396772,
      "rewards/format_reward_func": 0.9720982536673546,
      "step": 434
    },
    {
      "completion_length": 226.0413064956665,
      "epoch": 0.542541608337222,
      "grad_norm": 0.07098124510409774,
      "kl": 0.3662109375,
      "learning_rate": 1.2709384929615596e-09,
      "loss": 0.0004,
      "reward": 1.4821429252624512,
      "reward_std": 0.14848907012492418,
      "rewards/equation_reward_func": 0.49888394959270954,
      "rewards/format_reward_func": 0.983258955180645,
      "step": 436
    },
    {
      "completion_length": 233.2154140472412,
      "epoch": 0.5450303313112459,
      "grad_norm": 0.0583032288090882,
      "kl": 0.234619140625,
      "learning_rate": 9.339608617077165e-10,
      "loss": 0.0002,
      "reward": 1.4352679178118706,
      "reward_std": 0.12030374212190509,
      "rewards/equation_reward_func": 0.4564732341095805,
      "rewards/format_reward_func": 0.978794664144516,
      "step": 438
    },
    {
      "completion_length": 231.55581378936768,
      "epoch": 0.5475190542852699,
      "grad_norm": 0.05202154689567301,
      "kl": 0.19830322265625,
      "learning_rate": 6.487074273681114e-10,
      "loss": 0.0002,
      "reward": 1.475446492433548,
      "reward_std": 0.16154285473749042,
      "rewards/equation_reward_func": 0.49107144586741924,
      "rewards/format_reward_func": 0.9843750298023224,
      "step": 440
    },
    {
      "completion_length": 226.89844608306885,
      "epoch": 0.5500077772592938,
      "grad_norm": 0.06559019258824787,
      "kl": 0.212890625,
      "learning_rate": 4.152374292708538e-10,
      "loss": 0.0002,
      "reward": 1.455357201397419,
      "reward_std": 0.10218051401898265,
      "rewards/equation_reward_func": 0.4654018096625805,
      "rewards/format_reward_func": 0.9899553880095482,
      "step": 442
    },
    {
      "completion_length": 220.54799938201904,
      "epoch": 0.5524965002333178,
      "grad_norm": 0.05408063170536412,
      "kl": 0.2149658203125,
      "learning_rate": 2.3359935274214204e-10,
      "loss": 0.0002,
      "reward": 1.5022322237491608,
      "reward_std": 0.14629516191780567,
      "rewards/equation_reward_func": 0.5167410969734192,
      "rewards/format_reward_func": 0.9854911044239998,
      "step": 444
    },
    {
      "completion_length": 228.62389469146729,
      "epoch": 0.5549852232073418,
      "grad_norm": 0.04759889084008438,
      "kl": 0.21331787109375,
      "learning_rate": 1.0383091903720665e-10,
      "loss": 0.0002,
      "reward": 1.507812574505806,
      "reward_std": 0.13805820420384407,
      "rewards/equation_reward_func": 0.5245535988360643,
      "rewards/format_reward_func": 0.9832589700818062,
      "step": 446
    },
    {
      "completion_length": 239.22768783569336,
      "epoch": 0.5574739461813657,
      "grad_norm": 0.06951974665374468,
      "kl": 0.20892333984375,
      "learning_rate": 2.595907750671533e-11,
      "loss": 0.0002,
      "reward": 1.4174107685685158,
      "reward_std": 0.1395715670660138,
      "rewards/equation_reward_func": 0.43638394959270954,
      "rewards/format_reward_func": 0.981026828289032,
      "step": 448
    },
    {
      "completion_length": 230.02009963989258,
      "epoch": 0.5599626691553896,
      "grad_norm": 0.06979530747254646,
      "kl": 0.21221923828125,
      "learning_rate": 0.0,
      "loss": 0.0002,
      "reward": 1.4553571939468384,
      "reward_std": 0.12265546387061477,
      "rewards/equation_reward_func": 0.4776785969734192,
      "rewards/format_reward_func": 0.9776786081492901,
      "step": 450
    },
    {
      "epoch": 0.5599626691553896,
      "step": 450,
      "total_flos": 0.0,
      "train_loss": 0.0003818092903999215,
      "train_runtime": 35419.6415,
      "train_samples_per_second": 0.711,
      "train_steps_per_second": 0.013
    }
  ],
  "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
}