{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.11428571428571428,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 2699.791748046875,
      "entropy": 0.36767578125,
      "epoch": 0.001142857142857143,
      "grad_norm": 0.10378081351518631,
      "kl": 0.0,
      "learning_rate": 1e-07,
      "loss": 0.0593,
      "reward": 0.7708333535119891,
      "reward_std": 0.4629540964961052,
      "rewards/accuracy_reward": 0.25000001303851604,
      "rewards/format_reward": 0.5208333386108279,
      "step": 1
    },
    {
      "completion_length": 3127.3958435058594,
      "entropy": 0.353515625,
      "epoch": 0.002285714285714286,
      "grad_norm": 0.07586322724819183,
      "kl": 0.0,
      "learning_rate": 2e-07,
      "loss": 0.0439,
      "reward": 0.6458333637565374,
      "reward_std": 0.4249730706214905,
      "rewards/accuracy_reward": 0.2812500102445483,
      "rewards/format_reward": 0.3645833386108279,
      "step": 2
    },
    {
      "completion_length": 3761.0208740234375,
      "entropy": 0.4560546875,
      "epoch": 0.0034285714285714284,
      "grad_norm": 0.04515620321035385,
      "kl": 4.029273986816406e-05,
      "learning_rate": 3e-07,
      "loss": 0.0269,
      "reward": 0.1770833432674408,
      "reward_std": 0.2535070404410362,
      "rewards/accuracy_reward": 0.0625,
      "rewards/format_reward": 0.1145833395421505,
      "step": 3
    },
    {
      "completion_length": 2404.6146545410156,
      "entropy": 0.4072265625,
      "epoch": 0.004571428571428572,
      "grad_norm": 0.09874315559864044,
      "kl": 3.263354301452637e-05,
      "learning_rate": 4e-07,
      "loss": 0.0521,
      "reward": 0.8229166716337204,
      "reward_std": 0.48936954140663147,
      "rewards/accuracy_reward": 0.16666667070239782,
      "rewards/format_reward": 0.6562500149011612,
      "step": 4
    },
    {
      "completion_length": 3513.5521850585938,
      "entropy": 0.4580078125,
      "epoch": 0.005714285714285714,
      "grad_norm": 0.07631789147853851,
      "kl": 4.3064355850219727e-05,
      "learning_rate": 5e-07,
      "loss": 0.04,
      "reward": 0.5416666939854622,
      "reward_std": 0.5841840505599976,
      "rewards/accuracy_reward": 0.12500000558793545,
      "rewards/format_reward": 0.4166666865348816,
      "step": 5
    },
    {
      "completion_length": 3405.6251220703125,
      "entropy": 0.45556640625,
      "epoch": 0.006857142857142857,
      "grad_norm": 0.06789246201515198,
      "kl": 4.4912099838256836e-05,
      "learning_rate": 6e-07,
      "loss": 0.0183,
      "reward": 0.38541666977107525,
      "reward_std": 0.46006231755018234,
      "rewards/accuracy_reward": 0.06250000186264515,
      "rewards/format_reward": 0.32291668467223644,
      "step": 6
    },
    {
      "completion_length": 3237.2813720703125,
      "entropy": 0.380859375,
      "epoch": 0.008,
      "grad_norm": 0.0811946764588356,
      "kl": 1.9431114196777344e-05,
      "learning_rate": 7e-07,
      "loss": 0.0457,
      "reward": 0.8750000223517418,
      "reward_std": 0.6384046971797943,
      "rewards/accuracy_reward": 0.2604166716337204,
      "rewards/format_reward": 0.6145833507180214,
      "step": 7
    },
    {
      "completion_length": 2798.9688110351562,
      "entropy": 0.365234375,
      "epoch": 0.009142857142857144,
      "grad_norm": 0.052074775099754333,
      "kl": 1.9818544387817383e-05,
      "learning_rate": 8e-07,
      "loss": 0.0258,
      "reward": 0.9166667014360428,
      "reward_std": 0.3139677122235298,
      "rewards/accuracy_reward": 0.3958333535119891,
      "rewards/format_reward": 0.5208333432674408,
      "step": 8
    },
    {
      "completion_length": 3406.6771850585938,
      "entropy": 0.4599609375,
      "epoch": 0.010285714285714285,
      "grad_norm": 0.10204152017831802,
      "kl": 3.5703182220458984e-05,
      "learning_rate": 9e-07,
      "loss": 0.0666,
      "reward": 0.416666679084301,
      "reward_std": 0.5519952327013016,
      "rewards/accuracy_reward": 0.1041666679084301,
      "rewards/format_reward": 0.3125000111758709,
      "step": 9
    },
    {
      "completion_length": 2986.166748046875,
      "entropy": 0.34423828125,
      "epoch": 0.011428571428571429,
      "grad_norm": 0.0653812512755394,
      "kl": 1.93864107131958e-05,
      "learning_rate": 1e-06,
      "loss": 0.027,
      "reward": 0.5625000111758709,
      "reward_std": 0.47086329013109207,
      "rewards/accuracy_reward": 0.15625000558793545,
      "rewards/format_reward": 0.4062500149011612,
      "step": 10
    },
    {
      "completion_length": 3706.4688110351562,
      "entropy": 0.3798828125,
      "epoch": 0.012571428571428572,
      "grad_norm": 0.05876584351062775,
      "kl": 2.321600914001465e-05,
      "learning_rate": 9.997258721585931e-07,
      "loss": 0.0221,
      "reward": 0.23958334419876337,
      "reward_std": 0.4288342222571373,
      "rewards/accuracy_reward": 0.07291666697710752,
      "rewards/format_reward": 0.16666666977107525,
      "step": 11
    },
    {
      "completion_length": 2727.4896240234375,
      "entropy": 0.40966796875,
      "epoch": 0.013714285714285714,
      "grad_norm": 0.08620024472475052,
      "kl": 3.463029861450195e-05,
      "learning_rate": 9.989038226169207e-07,
      "loss": 0.0421,
      "reward": 0.7291666977107525,
      "reward_std": 0.3988698050379753,
      "rewards/accuracy_reward": 0.13541666697710752,
      "rewards/format_reward": 0.5937500111758709,
      "step": 12
    },
    {
      "completion_length": 3067.4896850585938,
      "entropy": 0.3759765625,
      "epoch": 0.014857142857142857,
      "grad_norm": 0.05084392428398132,
      "kl": 1.9818544387817383e-05,
      "learning_rate": 9.975348529157229e-07,
      "loss": 0.0229,
      "reward": 0.6666666865348816,
      "reward_std": 0.24620391800999641,
      "rewards/accuracy_reward": 0.21875000558793545,
      "rewards/format_reward": 0.4479166716337204,
      "step": 13
    },
    {
      "completion_length": 3072.791748046875,
      "entropy": 0.37548828125,
      "epoch": 0.016,
      "grad_norm": 0.11545959115028381,
      "kl": 1.6961246728897095e-05,
      "learning_rate": 9.956206309337066e-07,
      "loss": 0.0629,
      "reward": 0.7187500149011612,
      "reward_std": 0.5520583093166351,
      "rewards/accuracy_reward": 0.2604166744276881,
      "rewards/format_reward": 0.4583333507180214,
      "step": 14
    },
    {
      "completion_length": 2951.1458740234375,
      "entropy": 0.374755859375,
      "epoch": 0.017142857142857144,
      "grad_norm": 0.050988707691431046,
      "kl": 3.2514333724975586e-05,
      "learning_rate": 9.931634888554935e-07,
      "loss": 0.0276,
      "reward": 0.6770833432674408,
      "reward_std": 0.3433547392487526,
      "rewards/accuracy_reward": 0.2291666716337204,
      "rewards/format_reward": 0.447916679084301,
      "step": 15
    },
    {
      "completion_length": 3903.416748046875,
      "entropy": 0.474609375,
      "epoch": 0.018285714285714287,
      "grad_norm": 0.047724124044179916,
      "kl": 1.6005709767341614e-05,
      "learning_rate": 9.901664203302124e-07,
      "loss": 0.0075,
      "reward": 0.1666666716337204,
      "reward_std": 0.3083270490169525,
      "rewards/accuracy_reward": 0.0729166679084301,
      "rewards/format_reward": 0.0937500037252903,
      "step": 16
    },
    {
      "completion_length": 2442.3334350585938,
      "entropy": 0.443359375,
      "epoch": 0.019428571428571427,
      "grad_norm": 0.07992254197597504,
      "kl": 0.0001729130744934082,
      "learning_rate": 9.866330768241983e-07,
      "loss": 0.0298,
      "reward": 0.9375000149011612,
      "reward_std": 0.46943993866443634,
      "rewards/accuracy_reward": 0.3125000074505806,
      "rewards/format_reward": 0.6250000074505806,
      "step": 17
    },
    {
      "completion_length": 3152.9896850585938,
      "entropy": 0.38623046875,
      "epoch": 0.02057142857142857,
      "grad_norm": 0.058529071509838104,
      "kl": 0.000102996826171875,
      "learning_rate": 9.825677631722435e-07,
      "loss": 0.0241,
      "reward": 0.5937500298023224,
      "reward_std": 0.40303635224699974,
      "rewards/accuracy_reward": 0.14583333861082792,
      "rewards/format_reward": 0.447916679084301,
      "step": 18
    },
    {
      "completion_length": 3366.354248046875,
      "entropy": 0.41357421875,
      "epoch": 0.021714285714285714,
      "grad_norm": 0.057065945118665695,
      "kl": 7.18533992767334e-05,
      "learning_rate": 9.779754323328192e-07,
      "loss": 0.033,
      "reward": 0.7291666697710752,
      "reward_std": 0.42073434591293335,
      "rewards/accuracy_reward": 0.34375000558793545,
      "rewards/format_reward": 0.38541667722165585,
      "step": 19
    },
    {
      "completion_length": 2634.4896545410156,
      "entropy": 0.354736328125,
      "epoch": 0.022857142857142857,
      "grad_norm": 0.08138567209243774,
      "kl": 0.00015538930892944336,
      "learning_rate": 9.728616793536587e-07,
      "loss": 0.0625,
      "reward": 0.9375000298023224,
      "reward_std": 0.45325468480587006,
      "rewards/accuracy_reward": 0.2812500102445483,
      "rewards/format_reward": 0.6562500149011612,
      "step": 20
    },
    {
      "completion_length": 2729.7083740234375,
      "entropy": 0.44775390625,
      "epoch": 0.024,
      "grad_norm": 0.05598609521985054,
      "kl": 0.0005742311477661133,
      "learning_rate": 9.672327345550543e-07,
      "loss": 0.0194,
      "reward": 0.791666692122817,
      "reward_std": 0.3458455428481102,
      "rewards/accuracy_reward": 0.250000006519258,
      "rewards/format_reward": 0.5416666669771075,
      "step": 21
    },
    {
      "completion_length": 1972.1771545410156,
      "entropy": 0.4150390625,
      "epoch": 0.025142857142857144,
      "grad_norm": 0.045343004167079926,
      "kl": 0.0006244182586669922,
      "learning_rate": 9.610954559391704e-07,
      "loss": 0.0245,
      "reward": 1.0520833432674408,
      "reward_std": 0.3538536988198757,
      "rewards/accuracy_reward": 0.26041667722165585,
      "rewards/format_reward": 0.7916666716337204,
      "step": 22
    },
    {
      "completion_length": 2651.7813415527344,
      "entropy": 0.38134765625,
      "epoch": 0.026285714285714287,
      "grad_norm": 0.09305568039417267,
      "kl": 0.0003466606140136719,
      "learning_rate": 9.54457320834625e-07,
      "loss": 0.0566,
      "reward": 0.833333358168602,
      "reward_std": 0.563101589679718,
      "rewards/accuracy_reward": 0.23958334233611822,
      "rewards/format_reward": 0.5937500223517418,
      "step": 23
    },
    {
      "completion_length": 3014.854248046875,
      "entropy": 0.3828125,
      "epoch": 0.027428571428571427,
      "grad_norm": 0.08546774834394455,
      "kl": 0.00020091235637664795,
      "learning_rate": 9.473264167865171e-07,
      "loss": 0.0708,
      "reward": 0.7187500298023224,
      "reward_std": 0.5017067566514015,
      "rewards/accuracy_reward": 0.2291666679084301,
      "rewards/format_reward": 0.4895833432674408,
      "step": 24
    },
    {
      "completion_length": 2801.6250610351562,
      "entropy": 0.44580078125,
      "epoch": 0.02857142857142857,
      "grad_norm": 0.047619644552469254,
      "kl": 0.0005018115043640137,
      "learning_rate": 9.397114317029974e-07,
      "loss": 0.025,
      "reward": 0.614583358168602,
      "reward_std": 0.30002114176750183,
      "rewards/accuracy_reward": 0.1562500074505806,
      "rewards/format_reward": 0.4583333432674408,
      "step": 25
    },
    {
      "completion_length": 3229.6250610351562,
      "entropy": 0.43603515625,
      "epoch": 0.029714285714285714,
      "grad_norm": 0.030185947194695473,
      "kl": 0.00011354684829711914,
      "learning_rate": 9.316216432703916e-07,
      "loss": 0.0144,
      "reward": 0.6562500223517418,
      "reward_std": 0.21994514763355255,
      "rewards/accuracy_reward": 0.2083333358168602,
      "rewards/format_reward": 0.447916679084301,
      "step": 26
    },
    {
      "completion_length": 3260.0938110351562,
      "entropy": 0.4638671875,
      "epoch": 0.030857142857142857,
      "grad_norm": 0.1223464086651802,
      "kl": 0.0009315013885498047,
      "learning_rate": 9.230669076497687e-07,
      "loss": 0.0537,
      "reward": 0.5520833507180214,
      "reward_std": 0.5293410122394562,
      "rewards/accuracy_reward": 0.1458333358168602,
      "rewards/format_reward": 0.4062500223517418,
      "step": 27
    },
    {
      "completion_length": 3152.3021240234375,
      "entropy": 0.43310546875,
      "epoch": 0.032,
      "grad_norm": 0.05098699778318405,
      "kl": 0.0033740997314453125,
      "learning_rate": 9.140576474687263e-07,
      "loss": 0.0395,
      "reward": 0.6875000149011612,
      "reward_std": 0.34270069003105164,
      "rewards/accuracy_reward": 0.29166667722165585,
      "rewards/format_reward": 0.3958333358168602,
      "step": 28
    },
    {
      "completion_length": 3795.7708740234375,
      "entropy": 0.47900390625,
      "epoch": 0.03314285714285714,
      "grad_norm": 0.05339181423187256,
      "kl": 0.0014908313751220703,
      "learning_rate": 9.046048391230247e-07,
      "loss": 0.0268,
      "reward": 0.21875000558793545,
      "reward_std": 0.36166706681251526,
      "rewards/accuracy_reward": 0.09375000093132257,
      "rewards/format_reward": 0.12500000651925802,
      "step": 29
    },
    {
      "completion_length": 3244.0313110351562,
      "entropy": 0.42431640625,
      "epoch": 0.03428571428571429,
      "grad_norm": 0.08664752542972565,
      "kl": 0.0019369125366210938,
      "learning_rate": 8.9471999940354e-07,
      "loss": 0.0417,
      "reward": 0.6145833386108279,
      "reward_std": 0.5796928033232689,
      "rewards/accuracy_reward": 0.21875000279396772,
      "rewards/format_reward": 0.3958333386108279,
      "step": 30
    },
    {
      "completion_length": 3305.9063110351562,
      "entropy": 0.423828125,
      "epoch": 0.03542857142857143,
      "grad_norm": 0.0605226494371891,
      "kl": 0.000820159912109375,
      "learning_rate": 8.844151714648274e-07,
      "loss": 0.0404,
      "reward": 0.4791666865348816,
      "reward_std": 0.4150131866335869,
      "rewards/accuracy_reward": 0.1875000037252903,
      "rewards/format_reward": 0.2916666716337204,
      "step": 31
    },
    {
      "completion_length": 3559.572998046875,
      "entropy": 0.470703125,
      "epoch": 0.036571428571428574,
      "grad_norm": 0.08260089159011841,
      "kl": 0.004940032958984375,
      "learning_rate": 8.737029101523929e-07,
      "loss": 0.0361,
      "reward": 0.6770833507180214,
      "reward_std": 0.5618596971035004,
      "rewards/accuracy_reward": 0.30208333395421505,
      "rewards/format_reward": 0.3750000074505806,
      "step": 32
    },
    {
      "completion_length": 3600.5208740234375,
      "entropy": 0.4111328125,
      "epoch": 0.037714285714285714,
      "grad_norm": 0.07252604514360428,
      "kl": 0.0011572837829589844,
      "learning_rate": 8.625962667065487e-07,
      "loss": 0.0397,
      "reward": 0.5416666828095913,
      "reward_std": 0.5172309577465057,
      "rewards/accuracy_reward": 0.21875000558793545,
      "rewards/format_reward": 0.32291667722165585,
      "step": 33
    },
    {
      "completion_length": 2965.9896240234375,
      "entropy": 0.5205078125,
      "epoch": 0.038857142857142854,
      "grad_norm": 0.11166317760944366,
      "kl": 0.015777587890625,
      "learning_rate": 8.511087728614862e-07,
      "loss": 0.0416,
      "reward": 0.854166679084301,
      "reward_std": 0.46539078652858734,
      "rewards/accuracy_reward": 0.3750000111758709,
      "rewards/format_reward": 0.4791666828095913,
      "step": 34
    },
    {
      "completion_length": 3255.885467529297,
      "entropy": 0.43994140625,
      "epoch": 0.04,
      "grad_norm": 0.08642137050628662,
      "kl": 0.0021219253540039062,
      "learning_rate": 8.392544243589427e-07,
      "loss": 0.0466,
      "reward": 0.6562500223517418,
      "reward_std": 0.5902808457612991,
      "rewards/accuracy_reward": 0.26041668094694614,
      "rewards/format_reward": 0.3958333469927311,
      "step": 35
    },
    {
      "completion_length": 3611.8646240234375,
      "entropy": 0.57421875,
      "epoch": 0.04114285714285714,
      "grad_norm": 0.04823267087340355,
      "kl": 0.0031194686889648438,
      "learning_rate": 8.270476638965461e-07,
      "loss": 0.0177,
      "reward": 0.1666666679084301,
      "reward_std": 0.24164992570877075,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.14583333674818277,
      "step": 36
    },
    {
      "completion_length": 3503.0625,
      "entropy": 0.4638671875,
      "epoch": 0.04228571428571429,
      "grad_norm": 0.052913472056388855,
      "kl": 0.001556396484375,
      "learning_rate": 8.145033635316128e-07,
      "loss": 0.0278,
      "reward": 0.1875000074505806,
      "reward_std": 0.11020193248987198,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.1875000074505806,
      "step": 37
    },
    {
      "completion_length": 3664.5625610351562,
      "entropy": 0.513671875,
      "epoch": 0.04342857142857143,
      "grad_norm": 0.05083264037966728,
      "kl": 0.0012769699096679688,
      "learning_rate": 8.01636806561836e-07,
      "loss": 0.0244,
      "reward": 0.3229166716337204,
      "reward_std": 0.27773431688547134,
      "rewards/accuracy_reward": 0.1458333358168602,
      "rewards/format_reward": 0.1770833358168602,
      "step": 38
    },
    {
      "completion_length": 2993.385498046875,
      "entropy": 0.39794921875,
      "epoch": 0.044571428571428574,
      "grad_norm": 0.0582449734210968,
      "kl": 0.0028791427612304688,
      "learning_rate": 7.884636689049422e-07,
      "loss": 0.0389,
      "reward": 0.729166679084301,
      "reward_std": 0.34878823161125183,
      "rewards/accuracy_reward": 0.25000000558793545,
      "rewards/format_reward": 0.479166679084301,
      "step": 39
    },
    {
      "completion_length": 3047.6563720703125,
      "entropy": 0.455078125,
      "epoch": 0.045714285714285714,
      "grad_norm": 0.1776048094034195,
      "kl": 0.033954620361328125,
      "learning_rate": 7.75e-07,
      "loss": 0.0499,
      "reward": 0.6145833656191826,
      "reward_std": 0.4401242658495903,
      "rewards/accuracy_reward": 0.1354166679084301,
      "rewards/format_reward": 0.4791666828095913,
      "step": 40
    },
    {
      "completion_length": 3675.8021240234375,
      "entropy": 0.455078125,
      "epoch": 0.046857142857142854,
      "grad_norm": 0.0947902724146843,
      "kl": 0.0025844573974609375,
      "learning_rate": 7.612622032536507e-07,
      "loss": 0.034,
      "reward": 0.3958333358168602,
      "reward_std": 0.4633646681904793,
      "rewards/accuracy_reward": 0.11458333861082792,
      "rewards/format_reward": 0.2812500149011612,
      "step": 41
    },
    {
      "completion_length": 3077.5521850585938,
      "entropy": 0.49755859375,
      "epoch": 0.048,
      "grad_norm": 0.043859515339136124,
      "kl": 0.0016508102416992188,
      "learning_rate": 7.472670160550848e-07,
      "loss": 0.0086,
      "reward": 0.3437500037252903,
      "reward_std": 0.19173795729875565,
      "rewards/accuracy_reward": 0.031250000931322575,
      "rewards/format_reward": 0.3125,
      "step": 42
    },
    {
      "completion_length": 3321.8750610351562,
      "entropy": 0.49462890625,
      "epoch": 0.04914285714285714,
      "grad_norm": 0.07441486418247223,
      "kl": 0.0028228759765625,
      "learning_rate": 7.330314893841101e-07,
      "loss": 0.0332,
      "reward": 0.5000000074505806,
      "reward_std": 0.4754345528781414,
      "rewards/accuracy_reward": 0.18750000279396772,
      "rewards/format_reward": 0.3125000074505806,
      "step": 43
    },
    {
      "completion_length": 2997.7709350585938,
      "entropy": 0.466796875,
      "epoch": 0.05028571428571429,
      "grad_norm": 0.08534099161624908,
      "kl": 0.0044193267822265625,
      "learning_rate": 7.185729670371604e-07,
      "loss": 0.0371,
      "reward": 0.6979166865348816,
      "reward_std": 0.4120652452111244,
      "rewards/accuracy_reward": 0.2500000037252903,
      "rewards/format_reward": 0.447916679084301,
      "step": 44
    },
    {
      "completion_length": 3698.2396850585938,
      "entropy": 0.4775390625,
      "epoch": 0.05142857142857143,
      "grad_norm": 0.052873097360134125,
      "kl": 0.0031747817993164062,
      "learning_rate": 7.039090644965509e-07,
      "loss": 0.0083,
      "reward": 0.40625001676380634,
      "reward_std": 0.4317762479186058,
      "rewards/accuracy_reward": 0.15625000651925802,
      "rewards/format_reward": 0.250000006519258,
      "step": 45
    },
    {
      "completion_length": 3654.0521240234375,
      "entropy": 0.5927734375,
      "epoch": 0.052571428571428575,
      "grad_norm": 0.06117837131023407,
      "kl": 0.0036220550537109375,
      "learning_rate": 6.890576474687263e-07,
      "loss": 0.0261,
      "reward": 0.2187500074505806,
      "reward_std": 0.3954956606030464,
      "rewards/accuracy_reward": 0.02083333395421505,
      "rewards/format_reward": 0.1979166716337204,
      "step": 46
    },
    {
      "completion_length": 3092.8750610351562,
      "entropy": 0.455078125,
      "epoch": 0.053714285714285714,
      "grad_norm": 0.10391207039356232,
      "kl": 0.0032706260681152344,
      "learning_rate": 6.740368101176495e-07,
      "loss": 0.0729,
      "reward": 0.7812500298023224,
      "reward_std": 0.5671325027942657,
      "rewards/accuracy_reward": 0.2916666716337204,
      "rewards/format_reward": 0.4895833507180214,
      "step": 47
    },
    {
      "completion_length": 3051.979248046875,
      "entropy": 0.4609375,
      "epoch": 0.054857142857142854,
      "grad_norm": 0.03355824202299118,
      "kl": 0.0069446563720703125,
      "learning_rate": 6.588648530198504e-07,
      "loss": 0.0042,
      "reward": 0.5625000055879354,
      "reward_std": 0.3032701760530472,
      "rewards/accuracy_reward": 0.16666666697710752,
      "rewards/format_reward": 0.39583333395421505,
      "step": 48
    },
    {
      "completion_length": 2525.9063720703125,
      "entropy": 0.42431640625,
      "epoch": 0.056,
      "grad_norm": 0.0799187645316124,
      "kl": 0.002651214599609375,
      "learning_rate": 6.435602608679916e-07,
      "loss": 0.0455,
      "reward": 0.7916666865348816,
      "reward_std": 0.45931655168533325,
      "rewards/accuracy_reward": 0.21875000651925802,
      "rewards/format_reward": 0.5729166865348816,
      "step": 49
    },
    {
      "completion_length": 3233.9896240234375,
      "entropy": 0.4208984375,
      "epoch": 0.05714285714285714,
      "grad_norm": 0.06934866309165955,
      "kl": 0.0029773712158203125,
      "learning_rate": 6.281416799501187e-07,
      "loss": 0.0284,
      "reward": 0.6562500111758709,
      "reward_std": 0.39541390538215637,
      "rewards/accuracy_reward": 0.2916666716337204,
      "rewards/format_reward": 0.3645833544433117,
      "step": 50
    },
    {
      "completion_length": 2639.4376220703125,
      "entropy": 0.548828125,
      "epoch": 0.05828571428571429,
      "grad_norm": 0.05569971352815628,
      "kl": 0.007915496826171875,
      "learning_rate": 6.126278954320294e-07,
      "loss": 0.0188,
      "reward": 0.6250000149011612,
      "reward_std": 0.31560593098402023,
      "rewards/accuracy_reward": 0.1145833395421505,
      "rewards/format_reward": 0.5104166716337204,
      "step": 51
    },
    {
      "completion_length": 3200.7083740234375,
      "entropy": 0.4736328125,
      "epoch": 0.05942857142857143,
      "grad_norm": 0.10418283194303513,
      "kl": 0.0034847259521484375,
      "learning_rate": 5.97037808470444e-07,
      "loss": 0.0464,
      "reward": 0.7187500149011612,
      "reward_std": 0.5199342221021652,
      "rewards/accuracy_reward": 0.3020833432674408,
      "rewards/format_reward": 0.416666679084301,
      "step": 52
    },
    {
      "completion_length": 2897.354248046875,
      "entropy": 0.46533203125,
      "epoch": 0.060571428571428575,
      "grad_norm": 0.0808432400226593,
      "kl": 0.0022940635681152344,
      "learning_rate": 5.813904131848564e-07,
      "loss": 0.0321,
      "reward": 0.8125000149011612,
      "reward_std": 0.5500369966030121,
      "rewards/accuracy_reward": 0.2812500037252903,
      "rewards/format_reward": 0.5312500074505806,
      "step": 53
    },
    {
      "completion_length": 3137.5834350585938,
      "entropy": 0.48291015625,
      "epoch": 0.061714285714285715,
      "grad_norm": 0.14153237640857697,
      "kl": 0.0033512115478515625,
      "learning_rate": 5.657047735161255e-07,
      "loss": 0.0765,
      "reward": 0.8437500149011612,
      "reward_std": 0.568057231605053,
      "rewards/accuracy_reward": 0.354166679084301,
      "rewards/format_reward": 0.4895833358168602,
      "step": 54
    },
    {
      "completion_length": 3302.8021850585938,
      "entropy": 0.5078125,
      "epoch": 0.06285714285714286,
      "grad_norm": 0.06538388878107071,
      "kl": 0.004062652587890625,
      "learning_rate": 5.5e-07,
      "loss": 0.0373,
      "reward": 0.6145833358168602,
      "reward_std": 0.49769678711891174,
      "rewards/accuracy_reward": 0.19791667070239782,
      "rewards/format_reward": 0.4166666753590107,
      "step": 55
    },
    {
      "completion_length": 3132.322998046875,
      "entropy": 0.474609375,
      "epoch": 0.064,
      "grad_norm": 0.10536816716194153,
      "kl": 0.00362396240234375,
      "learning_rate": 5.342952264838747e-07,
      "loss": 0.034,
      "reward": 0.6041666669771075,
      "reward_std": 0.31034940481185913,
      "rewards/accuracy_reward": 0.1875,
      "rewards/format_reward": 0.4166666669771075,
      "step": 56
    },
    {
      "completion_length": 3647.697998046875,
      "entropy": 0.39453125,
      "epoch": 0.06514285714285714,
      "grad_norm": 0.056989964097738266,
      "kl": 0.002593994140625,
      "learning_rate": 5.186095868151436e-07,
      "loss": 0.0117,
      "reward": 0.46875002048909664,
      "reward_std": 0.5007961094379425,
      "rewards/accuracy_reward": 0.15625000279396772,
      "rewards/format_reward": 0.3125000176951289,
      "step": 57
    },
    {
      "completion_length": 2546.010498046875,
      "entropy": 0.3984375,
      "epoch": 0.06628571428571428,
      "grad_norm": 0.06985888630151749,
      "kl": 0.005336761474609375,
      "learning_rate": 5.02962191529556e-07,
      "loss": 0.0615,
      "reward": 0.9062500409781933,
      "reward_std": 0.49653898924589157,
      "rewards/accuracy_reward": 0.2708333395421505,
      "rewards/format_reward": 0.6354166828095913,
      "step": 58
    },
    {
      "completion_length": 3380.75,
      "entropy": 0.4609375,
      "epoch": 0.06742857142857143,
      "grad_norm": 0.05882963165640831,
      "kl": 0.004192352294921875,
      "learning_rate": 4.873721045679706e-07,
      "loss": 0.0238,
      "reward": 0.4687500298023224,
      "reward_std": 0.2912241667509079,
      "rewards/accuracy_reward": 0.1770833432674408,
      "rewards/format_reward": 0.2916666716337204,
      "step": 59
    },
    {
      "completion_length": 3201.0208740234375,
      "entropy": 0.45556640625,
      "epoch": 0.06857142857142857,
      "grad_norm": 0.04845750704407692,
      "kl": 0.003467559814453125,
      "learning_rate": 4.7185832004988133e-07,
      "loss": 0.0125,
      "reward": 0.5208333488553762,
      "reward_std": 0.3204401135444641,
      "rewards/accuracy_reward": 0.1354166679084301,
      "rewards/format_reward": 0.3854166669771075,
      "step": 60
    },
    {
      "completion_length": 3301.5626220703125,
      "entropy": 0.44921875,
      "epoch": 0.06971428571428571,
      "grad_norm": 0.1134149506688118,
      "kl": 0.00356292724609375,
      "learning_rate": 4.5643973913200837e-07,
      "loss": 0.077,
      "reward": 0.6666666893288493,
      "reward_std": 0.5132372379302979,
      "rewards/accuracy_reward": 0.2291666716337204,
      "rewards/format_reward": 0.4375000176951289,
      "step": 61
    },
    {
      "completion_length": 2941.3958740234375,
      "entropy": 0.437744140625,
      "epoch": 0.07085714285714285,
      "grad_norm": 0.11249862611293793,
      "kl": 0.014247894287109375,
      "learning_rate": 4.4113514698014953e-07,
      "loss": 0.0926,
      "reward": 0.833333358168602,
      "reward_std": 0.6926020532846451,
      "rewards/accuracy_reward": 0.28125000558793545,
      "rewards/format_reward": 0.5520833507180214,
      "step": 62
    },
    {
      "completion_length": 2724.9063110351562,
      "entropy": 0.4765625,
      "epoch": 0.072,
      "grad_norm": 0.12434863299131393,
      "kl": 0.00435638427734375,
      "learning_rate": 4.2596318988235037e-07,
      "loss": 0.0403,
      "reward": 0.9791666865348816,
      "reward_std": 0.46333979070186615,
      "rewards/accuracy_reward": 0.3229166669771075,
      "rewards/format_reward": 0.6562500074505806,
      "step": 63
    },
    {
      "completion_length": 3465.1563720703125,
      "entropy": 0.52001953125,
      "epoch": 0.07314285714285715,
      "grad_norm": 0.09894906729459763,
      "kl": 0.006351470947265625,
      "learning_rate": 4.1094235253127374e-07,
      "loss": 0.079,
      "reward": 0.5000000074505806,
      "reward_std": 0.564013384282589,
      "rewards/accuracy_reward": 0.17708333861082792,
      "rewards/format_reward": 0.3229166753590107,
      "step": 64
    },
    {
      "completion_length": 3121.6041870117188,
      "entropy": 0.48046875,
      "epoch": 0.07428571428571429,
      "grad_norm": 0.053176186978816986,
      "kl": 0.00548553466796875,
      "learning_rate": 3.9609093550344907e-07,
      "loss": 0.0199,
      "reward": 0.5937500176951289,
      "reward_std": 0.32823745906352997,
      "rewards/accuracy_reward": 0.16666666697710752,
      "rewards/format_reward": 0.4270833535119891,
      "step": 65
    },
    {
      "completion_length": 2457.375,
      "entropy": 0.4365234375,
      "epoch": 0.07542857142857143,
      "grad_norm": 0.040902264416217804,
      "kl": 0.00630950927734375,
      "learning_rate": 3.8142703296283953e-07,
      "loss": -0.0028,
      "reward": 0.9166666828095913,
      "reward_std": 0.33521074056625366,
      "rewards/accuracy_reward": 0.38541667722165585,
      "rewards/format_reward": 0.5312500055879354,
      "step": 66
    },
    {
      "completion_length": 3835.1354370117188,
      "entropy": 0.4384765625,
      "epoch": 0.07657142857142857,
      "grad_norm": 0.033646635711193085,
      "kl": 0.00453948974609375,
      "learning_rate": 3.6696851061588994e-07,
      "loss": 0.0167,
      "reward": 0.1458333358168602,
      "reward_std": 0.21662194281816483,
      "rewards/accuracy_reward": 0.031250000931322575,
      "rewards/format_reward": 0.1145833358168602,
      "step": 67
    },
    {
      "completion_length": 2685.2188110351562,
      "entropy": 0.5234375,
      "epoch": 0.07771428571428571,
      "grad_norm": 0.1085897833108902,
      "kl": 0.0067596435546875,
      "learning_rate": 3.5273298394491515e-07,
      "loss": 0.0768,
      "reward": 0.7083333656191826,
      "reward_std": 0.4535161405801773,
      "rewards/accuracy_reward": 0.1770833358168602,
      "rewards/format_reward": 0.5312500074505806,
      "step": 68
    },
    {
      "completion_length": 3197.5105590820312,
      "entropy": 0.62109375,
      "epoch": 0.07885714285714286,
      "grad_norm": 0.05905470252037048,
      "kl": 0.0102996826171875,
      "learning_rate": 3.387377967463493e-07,
      "loss": 0.0289,
      "reward": 0.3854166828095913,
      "reward_std": 0.3971848413348198,
      "rewards/accuracy_reward": 0.0416666679084301,
      "rewards/format_reward": 0.3437500074505806,
      "step": 69
    },
    {
      "completion_length": 3256.760498046875,
      "entropy": 0.4345703125,
      "epoch": 0.08,
      "grad_norm": 0.05583971366286278,
      "kl": 0.004856109619140625,
      "learning_rate": 3.250000000000001e-07,
      "loss": 0.0337,
      "reward": 0.5416666697710752,
      "reward_std": 0.44431453198194504,
      "rewards/accuracy_reward": 0.15625000279396772,
      "rewards/format_reward": 0.3854166744276881,
      "step": 70
    },
    {
      "completion_length": 3021.197967529297,
      "entropy": 0.529296875,
      "epoch": 0.08114285714285714,
      "grad_norm": 0.06864861398935318,
      "kl": 0.0126190185546875,
      "learning_rate": 3.115363310950578e-07,
      "loss": 0.016,
      "reward": 0.5520833386108279,
      "reward_std": 0.2611164078116417,
      "rewards/accuracy_reward": 0.1875,
      "rewards/format_reward": 0.3645833386108279,
      "step": 71
    },
    {
      "completion_length": 3671.9168090820312,
      "entropy": 0.62890625,
      "epoch": 0.08228571428571428,
      "grad_norm": 0.06177105754613876,
      "kl": 0.00665283203125,
      "learning_rate": 2.9836319343816397e-07,
      "loss": 0.0391,
      "reward": 0.3645833432674408,
      "reward_std": 0.5157303959131241,
      "rewards/accuracy_reward": 0.08333333674818277,
      "rewards/format_reward": 0.28125000558793545,
      "step": 72
    },
    {
      "completion_length": 3792.6563720703125,
      "entropy": 0.5341796875,
      "epoch": 0.08342857142857144,
      "grad_norm": 0.061636123806238174,
      "kl": 0.00244903564453125,
      "learning_rate": 2.854966364683872e-07,
      "loss": 0.0344,
      "reward": 0.2708333386108279,
      "reward_std": 0.4069225564599037,
      "rewards/accuracy_reward": 0.10416666977107525,
      "rewards/format_reward": 0.16666667070239782,
      "step": 73
    },
    {
      "completion_length": 3621.5313720703125,
      "entropy": 0.5068359375,
      "epoch": 0.08457142857142858,
      "grad_norm": 0.05920031666755676,
      "kl": 0.00554656982421875,
      "learning_rate": 2.729523361034538e-07,
      "loss": 0.0216,
      "reward": 0.4791666939854622,
      "reward_std": 0.4260884076356888,
      "rewards/accuracy_reward": 0.21875000186264515,
      "rewards/format_reward": 0.2604166753590107,
      "step": 74
    },
    {
      "completion_length": 3386.3751220703125,
      "entropy": 0.46826171875,
      "epoch": 0.08571428571428572,
      "grad_norm": 0.07354261726140976,
      "kl": 0.00623321533203125,
      "learning_rate": 2.6074557564105724e-07,
      "loss": 0.046,
      "reward": 0.5833333432674408,
      "reward_std": 0.47260352596640587,
      "rewards/accuracy_reward": 0.17708333488553762,
      "rewards/format_reward": 0.4062500149011612,
      "step": 75
    },
    {
      "completion_length": 3012.104248046875,
      "entropy": 0.4912109375,
      "epoch": 0.08685714285714285,
      "grad_norm": 0.08954478800296783,
      "kl": 0.00344085693359375,
      "learning_rate": 2.488912271385139e-07,
      "loss": 0.0614,
      "reward": 0.5833333432674408,
      "reward_std": 0.4225878678262234,
      "rewards/accuracy_reward": 0.08333333395421505,
      "rewards/format_reward": 0.5000000149011612,
      "step": 76
    },
    {
      "completion_length": 3350.6666870117188,
      "entropy": 0.52392578125,
      "epoch": 0.088,
      "grad_norm": 0.06333605200052261,
      "kl": 0.004314422607421875,
      "learning_rate": 2.374037332934512e-07,
      "loss": 0.0365,
      "reward": 0.5312500214204192,
      "reward_std": 0.3916928321123123,
      "rewards/accuracy_reward": 0.1041666716337204,
      "rewards/format_reward": 0.4270833497866988,
      "step": 77
    },
    {
      "completion_length": 3443.8438720703125,
      "entropy": 0.4697265625,
      "epoch": 0.08914285714285715,
      "grad_norm": 0.10582572221755981,
      "kl": 0.00374603271484375,
      "learning_rate": 2.2629708984760706e-07,
      "loss": 0.0468,
      "reward": 0.6354166865348816,
      "reward_std": 0.6427144631743431,
      "rewards/accuracy_reward": 0.2291666716337204,
      "rewards/format_reward": 0.4062500074505806,
      "step": 78
    },
    {
      "completion_length": 2630.135467529297,
      "entropy": 0.41455078125,
      "epoch": 0.09028571428571429,
      "grad_norm": 0.07537446171045303,
      "kl": 0.01548004150390625,
      "learning_rate": 2.1558482853517253e-07,
      "loss": 0.0285,
      "reward": 0.8750000298023224,
      "reward_std": 0.44084493070840836,
      "rewards/accuracy_reward": 0.27083334140479565,
      "rewards/format_reward": 0.604166679084301,
      "step": 79
    },
    {
      "completion_length": 3601.322998046875,
      "entropy": 0.57275390625,
      "epoch": 0.09142857142857143,
      "grad_norm": 0.050721827894449234,
      "kl": 0.0056438446044921875,
      "learning_rate": 2.0528000059645995e-07,
      "loss": 0.0131,
      "reward": 0.5104166772216558,
      "reward_std": 0.27086786925792694,
      "rewards/accuracy_reward": 0.1770833432674408,
      "rewards/format_reward": 0.33333333395421505,
      "step": 80
    },
    {
      "completion_length": 3389.9063110351562,
      "entropy": 0.716796875,
      "epoch": 0.09257142857142857,
      "grad_norm": 0.04999749734997749,
      "kl": 0.01094818115234375,
      "learning_rate": 1.9539516087697517e-07,
      "loss": 0.011,
      "reward": 0.39583334885537624,
      "reward_std": 0.30708885192871094,
      "rewards/accuracy_reward": 0.11458333674818277,
      "rewards/format_reward": 0.2812500027939677,
      "step": 81
    },
    {
      "completion_length": 3012.3959350585938,
      "entropy": 0.51806640625,
      "epoch": 0.09371428571428571,
      "grad_norm": 0.054486002773046494,
      "kl": 0.009601593017578125,
      "learning_rate": 1.8594235253127372e-07,
      "loss": 0.0318,
      "reward": 0.572916679084301,
      "reward_std": 0.30159565061330795,
      "rewards/accuracy_reward": 0.18750000558793545,
      "rewards/format_reward": 0.385416679084301,
      "step": 82
    },
    {
      "completion_length": 3209.9271850585938,
      "entropy": 0.5732421875,
      "epoch": 0.09485714285714286,
      "grad_norm": 0.04813052713871002,
      "kl": 0.00592041015625,
      "learning_rate": 1.7693309235023127e-07,
      "loss": 0.012,
      "reward": 0.41666667349636555,
      "reward_std": 0.3812928982079029,
      "rewards/accuracy_reward": 0.1354166679084301,
      "rewards/format_reward": 0.2812500009313226,
      "step": 83
    },
    {
      "completion_length": 3261.291748046875,
      "entropy": 0.521484375,
      "epoch": 0.096,
      "grad_norm": 0.07685024291276932,
      "kl": 0.0028629302978515625,
      "learning_rate": 1.6837835672960831e-07,
      "loss": 0.0347,
      "reward": 0.8125,
      "reward_std": 0.452178031206131,
      "rewards/accuracy_reward": 0.34375,
      "rewards/format_reward": 0.46875,
      "step": 84
    },
    {
      "completion_length": 3507.2916870117188,
      "entropy": 0.45458984375,
      "epoch": 0.09714285714285714,
      "grad_norm": 0.0731167271733284,
      "kl": 0.004367828369140625,
      "learning_rate": 1.6028856829700258e-07,
      "loss": 0.0261,
      "reward": 0.677083358168602,
      "reward_std": 0.6612033396959305,
      "rewards/accuracy_reward": 0.2500000102445483,
      "rewards/format_reward": 0.42708334885537624,
      "step": 85
    },
    {
      "completion_length": 3195.3646850585938,
      "entropy": 0.5634765625,
      "epoch": 0.09828571428571428,
      "grad_norm": 0.06479182839393616,
      "kl": 0.00772857666015625,
      "learning_rate": 1.5267358321348285e-07,
      "loss": 0.028,
      "reward": 0.5729166865348816,
      "reward_std": 0.24059326201677322,
      "rewards/accuracy_reward": 0.1354166716337204,
      "rewards/format_reward": 0.4375000074505806,
      "step": 86
    },
    {
      "completion_length": 3176.0833740234375,
      "entropy": 0.646484375,
      "epoch": 0.09942857142857142,
      "grad_norm": 0.10074935853481293,
      "kl": 0.0096588134765625,
      "learning_rate": 1.4554267916537495e-07,
      "loss": 0.0461,
      "reward": 0.7291666939854622,
      "reward_std": 0.45753754302859306,
      "rewards/accuracy_reward": 0.1979166716337204,
      "rewards/format_reward": 0.5312500149011612,
      "step": 87
    },
    {
      "completion_length": 3095.9584350585938,
      "entropy": 0.5673828125,
      "epoch": 0.10057142857142858,
      "grad_norm": 0.1463245153427124,
      "kl": 0.02208709716796875,
      "learning_rate": 1.3890454406082956e-07,
      "loss": 0.0721,
      "reward": 0.833333358168602,
      "reward_std": 0.7277083396911621,
      "rewards/accuracy_reward": 0.3437500149011612,
      "rewards/format_reward": 0.4895833507180214,
      "step": 88
    },
    {
      "completion_length": 3526.2605590820312,
      "entropy": 0.51171875,
      "epoch": 0.10171428571428572,
      "grad_norm": 0.058177653700113297,
      "kl": 0.00699615478515625,
      "learning_rate": 1.3276726544494571e-07,
      "loss": 0.024,
      "reward": 0.4375000111758709,
      "reward_std": 0.34694401547312737,
      "rewards/accuracy_reward": 0.14583334140479565,
      "rewards/format_reward": 0.29166666977107525,
      "step": 89
    },
    {
      "completion_length": 2816.729248046875,
      "entropy": 0.724609375,
      "epoch": 0.10285714285714286,
      "grad_norm": 0.0412394180893898,
      "kl": 0.01495361328125,
      "learning_rate": 1.2713832064634125e-07,
      "loss": 0.0118,
      "reward": 0.53125,
      "reward_std": 0.19076444953680038,
      "rewards/accuracy_reward": 0.041666666977107525,
      "rewards/format_reward": 0.4895833432674408,
      "step": 90
    },
    {
      "completion_length": 3363.0521240234375,
      "entropy": 0.548828125,
      "epoch": 0.104,
      "grad_norm": 0.09054948389530182,
      "kl": 0.0079193115234375,
      "learning_rate": 1.220245676671809e-07,
      "loss": 0.0385,
      "reward": 0.6458333358168602,
      "reward_std": 0.48818454146385193,
      "rewards/accuracy_reward": 0.2291666679084301,
      "rewards/format_reward": 0.4166666679084301,
      "step": 91
    },
    {
      "completion_length": 3039.9896850585938,
      "entropy": 0.52685546875,
      "epoch": 0.10514285714285715,
      "grad_norm": 0.3311564326286316,
      "kl": 0.111328125,
      "learning_rate": 1.1743223682775649e-07,
      "loss": 0.0405,
      "reward": 0.7187500298023224,
      "reward_std": 0.5042702481150627,
      "rewards/accuracy_reward": 0.17708333861082792,
      "rewards/format_reward": 0.541666679084301,
      "step": 92
    },
    {
      "completion_length": 3752.4166870117188,
      "entropy": 0.68115234375,
      "epoch": 0.10628571428571429,
      "grad_norm": 0.07829831540584564,
      "kl": 0.0079345703125,
      "learning_rate": 1.1336692317580158e-07,
      "loss": 0.0213,
      "reward": 0.07291666697710752,
      "reward_std": 0.10136350989341736,
      "rewards/accuracy_reward": 0.0,
      "rewards/format_reward": 0.07291666697710752,
      "step": 93
    },
    {
      "completion_length": 3379.666748046875,
      "entropy": 0.6376953125,
      "epoch": 0.10742857142857143,
      "grad_norm": 0.0632358267903328,
      "kl": 0.009815216064453125,
      "learning_rate": 1.0983357966978745e-07,
      "loss": 0.0332,
      "reward": 0.4687500186264515,
      "reward_std": 0.3699027970433235,
      "rewards/accuracy_reward": 0.1354166716337204,
      "rewards/format_reward": 0.3333333395421505,
      "step": 94
    },
    {
      "completion_length": 3759.7500610351562,
      "entropy": 0.50048828125,
      "epoch": 0.10857142857142857,
      "grad_norm": 0.06647934019565582,
      "kl": 0.0034885406494140625,
      "learning_rate": 1.068365111445064e-07,
      "loss": 0.0212,
      "reward": 0.4270833386108279,
      "reward_std": 0.5315546244382858,
      "rewards/accuracy_reward": 0.13541667442768812,
      "rewards/format_reward": 0.2916666744276881,
      "step": 95
    },
    {
      "completion_length": 3273.229248046875,
      "entropy": 0.5556640625,
      "epoch": 0.10971428571428571,
      "grad_norm": 0.08489354699850082,
      "kl": 0.008378982543945312,
      "learning_rate": 1.0437936906629334e-07,
      "loss": 0.0477,
      "reward": 0.6458333507180214,
      "reward_std": 0.48296865820884705,
      "rewards/accuracy_reward": 0.2604166716337204,
      "rewards/format_reward": 0.385416679084301,
      "step": 96
    },
    {
      "completion_length": 3500.1563110351562,
      "entropy": 0.546875,
      "epoch": 0.11085714285714286,
      "grad_norm": 0.03905348479747772,
      "kl": 0.0056133270263671875,
      "learning_rate": 1.0246514708427701e-07,
      "loss": 0.0213,
      "reward": 0.531250013038516,
      "reward_std": 0.2828656807541847,
      "rewards/accuracy_reward": 0.22916667722165585,
      "rewards/format_reward": 0.3020833386108279,
      "step": 97
    },
    {
      "completion_length": 3151.5625610351562,
      "entropy": 0.49755859375,
      "epoch": 0.112,
      "grad_norm": 0.06831816583871841,
      "kl": 0.00501251220703125,
      "learning_rate": 1.0109617738307911e-07,
      "loss": 0.0345,
      "reward": 0.7187500298023224,
      "reward_std": 0.5354187488555908,
      "rewards/accuracy_reward": 0.18750001024454832,
      "rewards/format_reward": 0.5312500149011612,
      "step": 98
    },
    {
      "completion_length": 3112.8646545410156,
      "entropy": 0.4921875,
      "epoch": 0.11314285714285714,
      "grad_norm": 0.05772259458899498,
      "kl": 0.0065765380859375,
      "learning_rate": 1.002741278414069e-07,
      "loss": 0.0306,
      "reward": 0.5312500055879354,
      "reward_std": 0.38146835565567017,
      "rewards/accuracy_reward": 0.1979166716337204,
      "rewards/format_reward": 0.33333334140479565,
      "step": 99
    },
    {
      "completion_length": 3131.6146240234375,
      "entropy": 0.49755859375,
      "epoch": 0.11428571428571428,
      "grad_norm": 0.10680815577507019,
      "kl": 0.00878143310546875,
      "learning_rate": 1e-07,
      "loss": 0.0572,
      "reward": 0.7395833358168602,
      "reward_std": 0.4930148720741272,
      "rewards/accuracy_reward": 0.2708333358168602,
      "rewards/format_reward": 0.4687500074505806,
      "step": 100
    },
    {
      "epoch": 0.11428571428571428,
      "step": 100,
      "total_flos": 0.0,
      "train_loss": 0.03577750400174409,
      "train_runtime": 8397.3301,
      "train_samples_per_second": 1.143,
      "train_steps_per_second": 0.012
    }
  ],
  "logging_steps": 1,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 6,
  "trial_name": null,
  "trial_params": null
}