{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9893390191897654,
  "eval_steps": 100,
  "global_step": 58,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 365.3112087249756,
      "epoch": 0.017057569296375266,
      "grad_norm": 1.04221985968125,
      "kl": 0.0,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.0,
      "reward": 0.6432291865348816,
      "reward_std": 0.4505743272602558,
      "rewards/accuracy_reward": 0.11718750232830644,
      "rewards/format_reward": 0.5260416772216558,
      "step": 1
    },
    {
      "completion_length": 367.3958435058594,
      "epoch": 0.03411513859275053,
      "grad_norm": 1.1434252212815785,
      "kl": 0.0,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.0,
      "reward": 0.6783854328095913,
      "reward_std": 0.46897680312395096,
      "rewards/accuracy_reward": 0.12630208570044488,
      "rewards/format_reward": 0.5520833525806665,
      "step": 2
    },
    {
      "completion_length": 339.8724060058594,
      "epoch": 0.0511727078891258,
      "grad_norm": 0.8635963812006352,
      "kl": 0.0010623931884765625,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "reward": 0.6835937723517418,
      "reward_std": 0.43934670090675354,
      "rewards/accuracy_reward": 0.09895833639893681,
      "rewards/format_reward": 0.5846354383975267,
      "step": 3
    },
    {
      "completion_length": 149.22526454925537,
      "epoch": 0.06823027718550106,
      "grad_norm": 307.2682075780713,
      "kl": 2.980987548828125,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.1193,
      "reward": 1.010416690260172,
      "reward_std": 0.2095302422530949,
      "rewards/accuracy_reward": 0.06640625139698386,
      "rewards/format_reward": 0.9440104439854622,
      "step": 4
    },
    {
      "completion_length": 112.50521183013916,
      "epoch": 0.08528784648187633,
      "grad_norm": 507.14990688734866,
      "kl": 9.401611328125,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.3775,
      "reward": 1.007812526077032,
      "reward_std": 0.15946260537020862,
      "rewards/accuracy_reward": 0.04427083441987634,
      "rewards/format_reward": 0.9635416828095913,
      "step": 5
    },
    {
      "completion_length": 100.34375333786011,
      "epoch": 0.1023454157782516,
      "grad_norm": 5.424616910263027,
      "kl": 0.33245849609375,
      "learning_rate": 2e-05,
      "loss": 0.0133,
      "reward": 1.0052083618938923,
      "reward_std": 0.15602844418026507,
      "rewards/accuracy_reward": 0.05338541930541396,
      "rewards/format_reward": 0.9518229402601719,
      "step": 6
    },
    {
      "completion_length": 133.2343783378601,
      "epoch": 0.11940298507462686,
      "grad_norm": 1.7843437966022808,
      "kl": 0.23712158203125,
      "learning_rate": 1.9981755542233175e-05,
      "loss": 0.0095,
      "reward": 0.9570312686264515,
      "reward_std": 0.2223757691681385,
      "rewards/accuracy_reward": 0.04166666802484542,
      "rewards/format_reward": 0.9153646044433117,
      "step": 7
    },
    {
      "completion_length": 114.69271230697632,
      "epoch": 0.13646055437100213,
      "grad_norm": 0.623961957547949,
      "kl": 0.1710205078125,
      "learning_rate": 1.992708874098054e-05,
      "loss": 0.0068,
      "reward": 0.9895833544433117,
      "reward_std": 0.22396012931130826,
      "rewards/accuracy_reward": 0.054687500931322575,
      "rewards/format_reward": 0.934895858168602,
      "step": 8
    },
    {
      "completion_length": 104.41406488418579,
      "epoch": 0.1535181236673774,
      "grad_norm": 1.5259640643959018,
      "kl": 0.224365234375,
      "learning_rate": 1.983619906947144e-05,
      "loss": 0.009,
      "reward": 0.9765625260770321,
      "reward_std": 0.2754378484096378,
      "rewards/accuracy_reward": 0.0664062516298145,
      "rewards/format_reward": 0.9101562611758709,
      "step": 9
    },
    {
      "completion_length": 231.64844512939453,
      "epoch": 0.17057569296375266,
      "grad_norm": 40.31352492325995,
      "kl": 0.45977783203125,
      "learning_rate": 1.9709418174260523e-05,
      "loss": 0.0184,
      "reward": 0.9257812723517418,
      "reward_std": 0.39183398708701134,
      "rewards/accuracy_reward": 0.0976562516298145,
      "rewards/format_reward": 0.8281250186264515,
      "step": 10
    },
    {
      "completion_length": 613.7070560455322,
      "epoch": 0.18763326226012794,
      "grad_norm": 840.2414358802456,
      "kl": 3.21197509765625,
      "learning_rate": 1.954720866508546e-05,
      "loss": 0.1284,
      "reward": 0.5429687686264515,
      "reward_std": 0.4636515509337187,
      "rewards/accuracy_reward": 0.076822918956168,
      "rewards/format_reward": 0.46614584140479565,
      "step": 11
    },
    {
      "completion_length": 762.7265853881836,
      "epoch": 0.2046908315565032,
      "grad_norm": 10.528180991568039,
      "kl": 0.7625732421875,
      "learning_rate": 1.9350162426854152e-05,
      "loss": 0.0305,
      "reward": 0.38411459513008595,
      "reward_std": 0.3724043210968375,
      "rewards/accuracy_reward": 0.09505208616610616,
      "rewards/format_reward": 0.2890625102445483,
      "step": 12
    },
    {
      "completion_length": 382.18230056762695,
      "epoch": 0.22174840085287847,
      "grad_norm": 2.761605849015544,
      "kl": 0.384521484375,
      "learning_rate": 1.91189984599209e-05,
      "loss": 0.0154,
      "reward": 0.6484375223517418,
      "reward_std": 0.5331083796918392,
      "rewards/accuracy_reward": 0.06901041930541396,
      "rewards/format_reward": 0.5794271007180214,
      "step": 13
    },
    {
      "completion_length": 198.7330780029297,
      "epoch": 0.23880597014925373,
      "grad_norm": 0.6078465083236722,
      "kl": 0.23638916015625,
      "learning_rate": 1.8854560256532098e-05,
      "loss": 0.0095,
      "reward": 0.912760429084301,
      "reward_std": 0.4155522510409355,
      "rewards/accuracy_reward": 0.10937500302679837,
      "rewards/format_reward": 0.8033854328095913,
      "step": 14
    },
    {
      "completion_length": 136.53516006469727,
      "epoch": 0.255863539445629,
      "grad_norm": 0.8617440561534008,
      "kl": 0.21514892578125,
      "learning_rate": 1.8557812723014476e-05,
      "loss": 0.0086,
      "reward": 0.9492187723517418,
      "reward_std": 0.38575689122080803,
      "rewards/accuracy_reward": 0.09895833593327552,
      "rewards/format_reward": 0.8502604328095913,
      "step": 15
    },
    {
      "completion_length": 98.51042032241821,
      "epoch": 0.27292110874200426,
      "grad_norm": 0.6595763666425144,
      "kl": 0.2208251953125,
      "learning_rate": 1.8229838658936566e-05,
      "loss": 0.0088,
      "reward": 0.9726562686264515,
      "reward_std": 0.3207697505131364,
      "rewards/accuracy_reward": 0.08854167000390589,
      "rewards/format_reward": 0.8841146044433117,
      "step": 16
    },
    {
      "completion_length": 76.04166841506958,
      "epoch": 0.2899786780383795,
      "grad_norm": 3.048254714245794,
      "kl": 0.44549560546875,
      "learning_rate": 1.7871834806090502e-05,
      "loss": 0.0178,
      "reward": 1.0468750298023224,
      "reward_std": 0.15973232360556722,
      "rewards/accuracy_reward": 0.07552083500195295,
      "rewards/format_reward": 0.9713541865348816,
      "step": 17
    },
    {
      "completion_length": 81.40234661102295,
      "epoch": 0.3070362473347548,
      "grad_norm": 0.4832905802799025,
      "kl": 0.24029541015625,
      "learning_rate": 1.7485107481711014e-05,
      "loss": 0.0096,
      "reward": 1.0742187798023224,
      "reward_std": 0.20187148824334145,
      "rewards/accuracy_reward": 0.10026042012032121,
      "rewards/format_reward": 0.9739583507180214,
      "step": 18
    },
    {
      "completion_length": 126.81640911102295,
      "epoch": 0.32409381663113007,
      "grad_norm": 0.43834266665765215,
      "kl": 0.18353271484375,
      "learning_rate": 1.7071067811865477e-05,
      "loss": 0.0073,
      "reward": 1.0924479514360428,
      "reward_std": 0.2407330577261746,
      "rewards/accuracy_reward": 0.13151042140088975,
      "rewards/format_reward": 0.9609375186264515,
      "step": 19
    },
    {
      "completion_length": 174.13021278381348,
      "epoch": 0.3411513859275053,
      "grad_norm": 0.3691172026913083,
      "kl": 0.170654296875,
      "learning_rate": 1.6631226582407954e-05,
      "loss": 0.0068,
      "reward": 1.0807292088866234,
      "reward_std": 0.17400484485551715,
      "rewards/accuracy_reward": 0.10546875465661287,
      "rewards/format_reward": 0.9752604365348816,
      "step": 20
    },
    {
      "completion_length": 226.26042366027832,
      "epoch": 0.3582089552238806,
      "grad_norm": 0.35348063732954105,
      "kl": 0.162994384765625,
      "learning_rate": 1.6167188726285433e-05,
      "loss": 0.0065,
      "reward": 1.0768229514360428,
      "reward_std": 0.26323840813711286,
      "rewards/accuracy_reward": 0.12239583872724324,
      "rewards/format_reward": 0.9544271044433117,
      "step": 21
    },
    {
      "completion_length": 226.471360206604,
      "epoch": 0.3752665245202559,
      "grad_norm": 0.2548959416361629,
      "kl": 0.137939453125,
      "learning_rate": 1.568064746731156e-05,
      "loss": 0.0055,
      "reward": 1.0937500447034836,
      "reward_std": 0.19759728573262691,
      "rewards/accuracy_reward": 0.11588542012032121,
      "rewards/format_reward": 0.9778646007180214,
      "step": 22
    },
    {
      "completion_length": 222.7395896911621,
      "epoch": 0.39232409381663114,
      "grad_norm": 0.29372855664077896,
      "kl": 0.154876708984375,
      "learning_rate": 1.5173378141776569e-05,
      "loss": 0.0062,
      "reward": 1.1015625409781933,
      "reward_std": 0.21707096393220127,
      "rewards/accuracy_reward": 0.12890625349245965,
      "rewards/format_reward": 0.9726562686264515,
      "step": 23
    },
    {
      "completion_length": 202.705735206604,
      "epoch": 0.4093816631130064,
      "grad_norm": 0.2969744538892161,
      "kl": 0.155120849609375,
      "learning_rate": 1.4647231720437687e-05,
      "loss": 0.0062,
      "reward": 1.1184896230697632,
      "reward_std": 0.20930432621389627,
      "rewards/accuracy_reward": 0.1328125053551048,
      "rewards/format_reward": 0.9856770932674408,
      "step": 24
    },
    {
      "completion_length": 208.4622449874878,
      "epoch": 0.42643923240938164,
      "grad_norm": 0.2551264579415379,
      "kl": 0.145782470703125,
      "learning_rate": 1.410412805452757e-05,
      "loss": 0.0058,
      "reward": 1.1341146230697632,
      "reward_std": 0.20376197341829538,
      "rewards/accuracy_reward": 0.1458333374466747,
      "rewards/format_reward": 0.9882812611758709,
      "step": 25
    },
    {
      "completion_length": 194.07292079925537,
      "epoch": 0.44349680170575695,
      "grad_norm": 0.29985133201161923,
      "kl": 0.147552490234375,
      "learning_rate": 1.3546048870425356e-05,
      "loss": 0.0059,
      "reward": 1.1875000298023224,
      "reward_std": 0.22688957839272916,
      "rewards/accuracy_reward": 0.19531250500585884,
      "rewards/format_reward": 0.9921875074505806,
      "step": 26
    },
    {
      "completion_length": 199.3333396911621,
      "epoch": 0.4605543710021322,
      "grad_norm": 0.2944306259052974,
      "kl": 0.144775390625,
      "learning_rate": 1.297503053855203e-05,
      "loss": 0.0058,
      "reward": 1.1458333656191826,
      "reward_std": 0.19133792025968432,
      "rewards/accuracy_reward": 0.15494792093522847,
      "rewards/format_reward": 0.9908854253590107,
      "step": 27
    },
    {
      "completion_length": 172.79427576065063,
      "epoch": 0.47761194029850745,
      "grad_norm": 0.3362991815387882,
      "kl": 0.148773193359375,
      "learning_rate": 1.2393156642875579e-05,
      "loss": 0.006,
      "reward": 1.178385466337204,
      "reward_std": 0.22494715498760343,
      "rewards/accuracy_reward": 0.18489583837799728,
      "rewards/format_reward": 0.9934895895421505,
      "step": 28
    },
    {
      "completion_length": 176.18490028381348,
      "epoch": 0.4946695095948827,
      "grad_norm": 0.3044368690957198,
      "kl": 0.146514892578125,
      "learning_rate": 1.180255037813906e-05,
      "loss": 0.0059,
      "reward": 1.1614583730697632,
      "reward_std": 0.19968353700824082,
      "rewards/accuracy_reward": 0.16796875244472176,
      "rewards/format_reward": 0.9934895895421505,
      "step": 29
    },
    {
      "completion_length": 169.21615076065063,
      "epoch": 0.511727078891258,
      "grad_norm": 0.3050078730361449,
      "kl": 0.1395263671875,
      "learning_rate": 1.1205366802553231e-05,
      "loss": 0.0056,
      "reward": 1.1601563021540642,
      "reward_std": 0.19885429926216602,
      "rewards/accuracy_reward": 0.167968753259629,
      "rewards/format_reward": 0.9921875074505806,
      "step": 30
    },
    {
      "completion_length": 191.99479484558105,
      "epoch": 0.5287846481876333,
      "grad_norm": 0.28939261315708026,
      "kl": 0.145721435546875,
      "learning_rate": 1.0603784974222862e-05,
      "loss": 0.0058,
      "reward": 1.1354167088866234,
      "reward_std": 0.2171460180543363,
      "rewards/accuracy_reward": 0.15104167070239782,
      "rewards/format_reward": 0.9843750074505806,
      "step": 31
    },
    {
      "completion_length": 233.89974689483643,
      "epoch": 0.5458422174840085,
      "grad_norm": 0.28487537328396517,
      "kl": 0.13800048828125,
      "learning_rate": 1e-05,
      "loss": 0.0055,
      "reward": 1.123697966337204,
      "reward_std": 0.2813015836291015,
      "rewards/accuracy_reward": 0.15885417093522847,
      "rewards/format_reward": 0.9648437723517418,
      "step": 32
    },
    {
      "completion_length": 253.55078887939453,
      "epoch": 0.5628997867803838,
      "grad_norm": 0.2737431545513293,
      "kl": 0.137664794921875,
      "learning_rate": 9.39621502577714e-06,
      "loss": 0.0055,
      "reward": 1.128906287252903,
      "reward_std": 0.19965026015415788,
      "rewards/accuracy_reward": 0.1497395880287513,
      "rewards/format_reward": 0.9791666828095913,
      "step": 33
    },
    {
      "completion_length": 239.842453956604,
      "epoch": 0.579957356076759,
      "grad_norm": 0.6254452308530076,
      "kl": 0.14483642578125,
      "learning_rate": 8.79463319744677e-06,
      "loss": 0.0058,
      "reward": 1.1757812798023224,
      "reward_std": 0.26264199148863554,
      "rewards/accuracy_reward": 0.195312503259629,
      "rewards/format_reward": 0.9804687574505806,
      "step": 34
    },
    {
      "completion_length": 241.17187976837158,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.3224088206959575,
      "kl": 0.14569091796875,
      "learning_rate": 8.197449621860944e-06,
      "loss": 0.0058,
      "reward": 1.1757812947034836,
      "reward_std": 0.27850970113649964,
      "rewards/accuracy_reward": 0.199218753259629,
      "rewards/format_reward": 0.9765625186264515,
      "step": 35
    },
    {
      "completion_length": 231.39062976837158,
      "epoch": 0.6140724946695096,
      "grad_norm": 0.2936768596993624,
      "kl": 0.142852783203125,
      "learning_rate": 7.606843357124426e-06,
      "loss": 0.0057,
      "reward": 1.1510417088866234,
      "reward_std": 0.26834863936528563,
      "rewards/accuracy_reward": 0.17578125349245965,
      "rewards/format_reward": 0.9752604328095913,
      "step": 36
    },
    {
      "completion_length": 200.52344417572021,
      "epoch": 0.6311300639658849,
      "grad_norm": 0.3398881371255209,
      "kl": 0.14337158203125,
      "learning_rate": 7.024969461447973e-06,
      "loss": 0.0057,
      "reward": 1.140625037252903,
      "reward_std": 0.30551271699368954,
      "rewards/accuracy_reward": 0.20312500465661287,
      "rewards/format_reward": 0.9375000223517418,
      "step": 37
    },
    {
      "completion_length": 199.080735206604,
      "epoch": 0.6481876332622601,
      "grad_norm": 0.3950051985318622,
      "kl": 0.15716552734375,
      "learning_rate": 6.453951129574644e-06,
      "loss": 0.0063,
      "reward": 1.0898437723517418,
      "reward_std": 0.3697906183078885,
      "rewards/accuracy_reward": 0.1953125053551048,
      "rewards/format_reward": 0.8945312686264515,
      "step": 38
    },
    {
      "completion_length": 176.2083396911621,
      "epoch": 0.6652452025586354,
      "grad_norm": 0.37189719351321165,
      "kl": 0.1624755859375,
      "learning_rate": 5.895871945472434e-06,
      "loss": 0.0065,
      "reward": 1.1380208805203438,
      "reward_std": 0.34118577465415,
      "rewards/accuracy_reward": 0.20703125558793545,
      "rewards/format_reward": 0.9309896044433117,
      "step": 39
    },
    {
      "completion_length": 171.10677528381348,
      "epoch": 0.6823027718550106,
      "grad_norm": 0.3867912490480322,
      "kl": 0.15789794921875,
      "learning_rate": 5.352768279562315e-06,
      "loss": 0.0063,
      "reward": 1.1210937947034836,
      "reward_std": 0.28729582112282515,
      "rewards/accuracy_reward": 0.1640625053551048,
      "rewards/format_reward": 0.9570312686264515,
      "step": 40
    },
    {
      "completion_length": 159.16016101837158,
      "epoch": 0.6993603411513859,
      "grad_norm": 0.3463972010603223,
      "kl": 0.162689208984375,
      "learning_rate": 4.826621858223431e-06,
      "loss": 0.0065,
      "reward": 1.1770833656191826,
      "reward_std": 0.2748530600219965,
      "rewards/accuracy_reward": 0.22135417209938169,
      "rewards/format_reward": 0.9557291865348816,
      "step": 41
    },
    {
      "completion_length": 155.342453956604,
      "epoch": 0.7164179104477612,
      "grad_norm": 0.31066796112570005,
      "kl": 0.16387939453125,
      "learning_rate": 4.319352532688444e-06,
      "loss": 0.0066,
      "reward": 1.1692708805203438,
      "reward_std": 0.24965191585943103,
      "rewards/accuracy_reward": 0.19010417256504297,
      "rewards/format_reward": 0.9791666828095913,
      "step": 42
    },
    {
      "completion_length": 147.61849355697632,
      "epoch": 0.7334754797441365,
      "grad_norm": 0.3883785701764709,
      "kl": 0.15966796875,
      "learning_rate": 3.832811273714569e-06,
      "loss": 0.0064,
      "reward": 1.194010466337204,
      "reward_std": 0.26059679966419935,
      "rewards/accuracy_reward": 0.20442709000781178,
      "rewards/format_reward": 0.9895833432674408,
      "step": 43
    },
    {
      "completion_length": 160.03385829925537,
      "epoch": 0.7505330490405118,
      "grad_norm": 5788.508271348362,
      "kl": 278.14984130859375,
      "learning_rate": 3.3687734175920505e-06,
      "loss": 11.1208,
      "reward": 1.1757812947034836,
      "reward_std": 0.23907144693657756,
      "rewards/accuracy_reward": 0.18880208989139646,
      "rewards/format_reward": 0.986979179084301,
      "step": 44
    },
    {
      "completion_length": 149.94141149520874,
      "epoch": 0.767590618336887,
      "grad_norm": 0.36297405142580474,
      "kl": 0.164794921875,
      "learning_rate": 2.9289321881345257e-06,
      "loss": 0.0066,
      "reward": 1.199218787252903,
      "reward_std": 0.2734901886433363,
      "rewards/accuracy_reward": 0.21744792349636555,
      "rewards/format_reward": 0.9817708469927311,
      "step": 45
    },
    {
      "completion_length": 157.68620204925537,
      "epoch": 0.7846481876332623,
      "grad_norm": 0.3108491951827798,
      "kl": 0.16436767578125,
      "learning_rate": 2.514892518288988e-06,
      "loss": 0.0066,
      "reward": 1.1744792014360428,
      "reward_std": 0.21565480902791023,
      "rewards/accuracy_reward": 0.19010417233221233,
      "rewards/format_reward": 0.9843750149011612,
      "step": 46
    },
    {
      "completion_length": 168.7890682220459,
      "epoch": 0.8017057569296375,
      "grad_norm": 1.04029324716777,
      "kl": 0.16632080078125,
      "learning_rate": 2.1281651939094996e-06,
      "loss": 0.0067,
      "reward": 1.1197917014360428,
      "reward_std": 0.20990665443241596,
      "rewards/accuracy_reward": 0.13281250500585884,
      "rewards/format_reward": 0.9869791753590107,
      "step": 47
    },
    {
      "completion_length": 167.15625381469727,
      "epoch": 0.8187633262260128,
      "grad_norm": 0.32966812871466905,
      "kl": 0.15936279296875,
      "learning_rate": 1.7701613410634367e-06,
      "loss": 0.0064,
      "reward": 1.2018229588866234,
      "reward_std": 0.2573004774749279,
      "rewards/accuracy_reward": 0.22005208767950535,
      "rewards/format_reward": 0.9817708469927311,
      "step": 48
    },
    {
      "completion_length": 171.07682704925537,
      "epoch": 0.835820895522388,
      "grad_norm": 0.32208043175249607,
      "kl": 0.151123046875,
      "learning_rate": 1.4421872769855262e-06,
      "loss": 0.006,
      "reward": 1.203125037252903,
      "reward_std": 0.26521802693605423,
      "rewards/accuracy_reward": 0.22135417209938169,
      "rewards/format_reward": 0.9817708469927311,
      "step": 49
    },
    {
      "completion_length": 166.2864637374878,
      "epoch": 0.8528784648187633,
      "grad_norm": 0.34577456114094934,
      "kl": 0.16162109375,
      "learning_rate": 1.1454397434679022e-06,
      "loss": 0.0065,
      "reward": 1.2174479588866234,
      "reward_std": 0.2847161674872041,
      "rewards/accuracy_reward": 0.23307292023673654,
      "rewards/format_reward": 0.9843750111758709,
      "step": 50
    },
    {
      "completion_length": 176.26693153381348,
      "epoch": 0.8699360341151386,
      "grad_norm": 0.32710109344570104,
      "kl": 0.151611328125,
      "learning_rate": 8.810015400790994e-07,
      "loss": 0.0061,
      "reward": 1.1367187835276127,
      "reward_std": 0.2029442568309605,
      "rewards/accuracy_reward": 0.14713542093522847,
      "rewards/format_reward": 0.9895833395421505,
      "step": 51
    },
    {
      "completion_length": 181.84505653381348,
      "epoch": 0.8869936034115139,
      "grad_norm": 0.3181109150342723,
      "kl": 0.177215576171875,
      "learning_rate": 6.498375731458529e-07,
      "loss": 0.0071,
      "reward": 1.1692708618938923,
      "reward_std": 0.21676216088235378,
      "rewards/accuracy_reward": 0.18229167233221233,
      "rewards/format_reward": 0.9869791753590107,
      "step": 52
    },
    {
      "completion_length": 171.81901359558105,
      "epoch": 0.9040511727078892,
      "grad_norm": 0.3007348356231504,
      "kl": 0.15460205078125,
      "learning_rate": 4.5279133491454406e-07,
      "loss": 0.0062,
      "reward": 1.169270858168602,
      "reward_std": 0.26831901678815484,
      "rewards/accuracy_reward": 0.18880208872724324,
      "rewards/format_reward": 0.9804687686264515,
      "step": 53
    },
    {
      "completion_length": 165.6862015724182,
      "epoch": 0.9211087420042644,
      "grad_norm": 0.3465232635146208,
      "kl": 0.160491943359375,
      "learning_rate": 2.905818257394799e-07,
      "loss": 0.0064,
      "reward": 1.2343750223517418,
      "reward_std": 0.2902528368867934,
      "rewards/accuracy_reward": 0.2513020889600739,
      "rewards/format_reward": 0.983072929084301,
      "step": 54
    },
    {
      "completion_length": 185.47396278381348,
      "epoch": 0.9381663113006397,
      "grad_norm": 0.339986245997186,
      "kl": 0.151763916015625,
      "learning_rate": 1.6380093052856482e-07,
      "loss": 0.0061,
      "reward": 1.1757812909781933,
      "reward_std": 0.2616687403060496,
      "rewards/accuracy_reward": 0.1953125058207661,
      "rewards/format_reward": 0.9804687611758709,
      "step": 55
    },
    {
      "completion_length": 168.7773494720459,
      "epoch": 0.9552238805970149,
      "grad_norm": 0.33393014556768225,
      "kl": 0.152862548828125,
      "learning_rate": 7.291125901946027e-08,
      "loss": 0.0061,
      "reward": 1.2343750447034836,
      "reward_std": 0.2642471818253398,
      "rewards/accuracy_reward": 0.24479167396202683,
      "rewards/format_reward": 0.9895833432674408,
      "step": 56
    },
    {
      "completion_length": 175.51172637939453,
      "epoch": 0.9722814498933902,
      "grad_norm": 0.30356714080443054,
      "kl": 0.1541748046875,
      "learning_rate": 1.824445776682504e-08,
      "loss": 0.0062,
      "reward": 1.2239583656191826,
      "reward_std": 0.2539575123228133,
      "rewards/accuracy_reward": 0.24739584140479565,
      "rewards/format_reward": 0.9765625149011612,
      "step": 57
    },
    {
      "completion_length": 187.67057609558105,
      "epoch": 0.9893390191897654,
      "grad_norm": 0.3138605458712969,
      "kl": 0.152740478515625,
      "learning_rate": 0.0,
      "loss": 0.0061,
      "reward": 1.208333358168602,
      "reward_std": 0.24711204366758466,
      "rewards/accuracy_reward": 0.22786458861082792,
      "rewards/format_reward": 0.9804687649011612,
      "step": 58
    },
    {
      "epoch": 0.9893390191897654,
      "step": 58,
      "total_flos": 0.0,
      "train_loss": 0.20938398721150356,
      "train_runtime": 4954.7093,
      "train_samples_per_second": 1.514,
      "train_steps_per_second": 0.012
    }
  ],
  "logging_steps": 1,
  "max_steps": 58,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "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": 16,
  "trial_name": null,
  "trial_params": null
}