{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 67,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 588.9654293060303,
      "epoch": 0.014925373134328358,
      "grad_norm": 2.6534500122070312,
      "kl": 0.0,
      "learning_rate": 4.2857142857142857e-07,
      "loss": 0.0,
      "reward": 0.6618303954601288,
      "reward_std": 0.3340754471719265,
      "rewards/accuracy_reward": 0.6618303954601288,
      "rewards/format_reward": 0.0,
      "step": 1
    },
    {
      "completion_length": 574.2835121154785,
      "epoch": 0.029850746268656716,
      "grad_norm": 1.5236740112304688,
      "kl": 0.0,
      "learning_rate": 8.571428571428571e-07,
      "loss": 0.0,
      "reward": 0.6640625298023224,
      "reward_std": 0.35328569635748863,
      "rewards/accuracy_reward": 0.6640625298023224,
      "rewards/format_reward": 0.0,
      "step": 2
    },
    {
      "completion_length": 603.0803833007812,
      "epoch": 0.04477611940298507,
      "grad_norm": 1.810470461845398,
      "kl": 0.00018996000289916992,
      "learning_rate": 1.2857142857142856e-06,
      "loss": 0.0,
      "reward": 0.6685268227010965,
      "reward_std": 0.3423070926219225,
      "rewards/accuracy_reward": 0.6685268227010965,
      "rewards/format_reward": 0.0,
      "step": 3
    },
    {
      "completion_length": 662.1841831207275,
      "epoch": 0.05970149253731343,
      "grad_norm": 2.944155693054199,
      "kl": 0.00020492076873779297,
      "learning_rate": 1.7142857142857143e-06,
      "loss": 0.0,
      "reward": 0.6506696715950966,
      "reward_std": 0.3272070288658142,
      "rewards/accuracy_reward": 0.6506696715950966,
      "rewards/format_reward": 0.0,
      "step": 4
    },
    {
      "completion_length": 611.7890968322754,
      "epoch": 0.07462686567164178,
      "grad_norm": 0.5025805234909058,
      "kl": 0.00025087594985961914,
      "learning_rate": 2.142857142857143e-06,
      "loss": 0.0,
      "reward": 0.6216518208384514,
      "reward_std": 0.2848156485706568,
      "rewards/accuracy_reward": 0.6216518208384514,
      "rewards/format_reward": 0.0,
      "step": 5
    },
    {
      "completion_length": 606.5625228881836,
      "epoch": 0.08955223880597014,
      "grad_norm": 1.21269953250885,
      "kl": 0.0005187392234802246,
      "learning_rate": 2.571428571428571e-06,
      "loss": 0.0,
      "reward": 0.683035746216774,
      "reward_std": 0.3044421002268791,
      "rewards/accuracy_reward": 0.683035746216774,
      "rewards/format_reward": 0.0,
      "step": 6
    },
    {
      "completion_length": 600.8437728881836,
      "epoch": 0.1044776119402985,
      "grad_norm": 0.5010935664176941,
      "kl": 0.0006451606750488281,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "reward": 0.7120536044239998,
      "reward_std": 0.27275579422712326,
      "rewards/accuracy_reward": 0.7120536044239998,
      "rewards/format_reward": 0.0,
      "step": 7
    },
    {
      "completion_length": 604.5390853881836,
      "epoch": 0.11940298507462686,
      "grad_norm": 1.0228732824325562,
      "kl": 0.0014821290969848633,
      "learning_rate": 2.9979443021318607e-06,
      "loss": 0.0001,
      "reward": 0.7533482499420643,
      "reward_std": 0.26047475449740887,
      "rewards/accuracy_reward": 0.7533482499420643,
      "rewards/format_reward": 0.0,
      "step": 8
    },
    {
      "completion_length": 628.8348484039307,
      "epoch": 0.13432835820895522,
      "grad_norm": 0.6632704734802246,
      "kl": 0.0016405582427978516,
      "learning_rate": 2.99178284305241e-06,
      "loss": 0.0001,
      "reward": 0.6897321715950966,
      "reward_std": 0.2548013962805271,
      "rewards/accuracy_reward": 0.6897321715950966,
      "rewards/format_reward": 0.0,
      "step": 9
    },
    {
      "completion_length": 626.3750247955322,
      "epoch": 0.14925373134328357,
      "grad_norm": 1.5124034881591797,
      "kl": 0.0020029544830322266,
      "learning_rate": 2.981532510892707e-06,
      "loss": 0.0001,
      "reward": 0.7689732499420643,
      "reward_std": 0.2507007820531726,
      "rewards/accuracy_reward": 0.7689732499420643,
      "rewards/format_reward": 0.0,
      "step": 10
    },
    {
      "completion_length": 653.0457916259766,
      "epoch": 0.16417910447761194,
      "grad_norm": 0.3473568558692932,
      "kl": 0.0020618438720703125,
      "learning_rate": 2.9672214011007086e-06,
      "loss": 0.0001,
      "reward": 0.7109375335276127,
      "reward_std": 0.24070814065635204,
      "rewards/accuracy_reward": 0.7109375335276127,
      "rewards/format_reward": 0.0,
      "step": 11
    },
    {
      "completion_length": 624.7511444091797,
      "epoch": 0.1791044776119403,
      "grad_norm": 4.333103656768799,
      "kl": 0.006229400634765625,
      "learning_rate": 2.9488887394336023e-06,
      "loss": 0.0002,
      "reward": 0.6964286062866449,
      "reward_std": 0.25179733615368605,
      "rewards/accuracy_reward": 0.6964286062866449,
      "rewards/format_reward": 0.0,
      "step": 12
    },
    {
      "completion_length": 638.3526973724365,
      "epoch": 0.19402985074626866,
      "grad_norm": 0.12630324065685272,
      "kl": 0.0025043487548828125,
      "learning_rate": 2.9265847744427307e-06,
      "loss": 0.0001,
      "reward": 0.7578125335276127,
      "reward_std": 0.2057263469323516,
      "rewards/accuracy_reward": 0.7578125335276127,
      "rewards/format_reward": 0.0,
      "step": 13
    },
    {
      "completion_length": 594.9152069091797,
      "epoch": 0.208955223880597,
      "grad_norm": 0.1356700211763382,
      "kl": 0.002197265625,
      "learning_rate": 2.9003706397458025e-06,
      "loss": 0.0001,
      "reward": 0.7857143208384514,
      "reward_std": 0.16860559629276395,
      "rewards/accuracy_reward": 0.7857143208384514,
      "rewards/format_reward": 0.0,
      "step": 14
    },
    {
      "completion_length": 577.4933338165283,
      "epoch": 0.22388059701492538,
      "grad_norm": 0.12567110359668732,
      "kl": 0.0028967857360839844,
      "learning_rate": 2.8703181864639013e-06,
      "loss": 0.0001,
      "reward": 0.820312537252903,
      "reward_std": 0.17374721262603998,
      "rewards/accuracy_reward": 0.820312537252903,
      "rewards/format_reward": 0.0,
      "step": 15
    },
    {
      "completion_length": 593.5212230682373,
      "epoch": 0.23880597014925373,
      "grad_norm": 0.1694745570421219,
      "kl": 0.0028734207153320312,
      "learning_rate": 2.8365097862825516e-06,
      "loss": 0.0001,
      "reward": 0.7645089589059353,
      "reward_std": 0.2047506500966847,
      "rewards/accuracy_reward": 0.7645089589059353,
      "rewards/format_reward": 0.0,
      "step": 16
    },
    {
      "completion_length": 585.3772563934326,
      "epoch": 0.2537313432835821,
      "grad_norm": 0.23806491494178772,
      "kl": 0.0028820037841796875,
      "learning_rate": 2.7990381056766585e-06,
      "loss": 0.0001,
      "reward": 0.7645089663565159,
      "reward_std": 0.2365775415673852,
      "rewards/accuracy_reward": 0.7645089663565159,
      "rewards/format_reward": 0.0,
      "step": 17
    },
    {
      "completion_length": 589.7611808776855,
      "epoch": 0.26865671641791045,
      "grad_norm": 0.15571320056915283,
      "kl": 0.002490997314453125,
      "learning_rate": 2.758005851918136e-06,
      "loss": 0.0001,
      "reward": 0.7544643264263868,
      "reward_std": 0.21331929927691817,
      "rewards/accuracy_reward": 0.7544643264263868,
      "rewards/format_reward": 0.0,
      "step": 18
    },
    {
      "completion_length": 603.9732398986816,
      "epoch": 0.2835820895522388,
      "grad_norm": 0.20735572278499603,
      "kl": 0.0032434463500976562,
      "learning_rate": 2.713525491562421e-06,
      "loss": 0.0001,
      "reward": 0.7667411062866449,
      "reward_std": 0.17671918543055654,
      "rewards/accuracy_reward": 0.7667411062866449,
      "rewards/format_reward": 0.0,
      "step": 19
    },
    {
      "completion_length": 608.0290489196777,
      "epoch": 0.29850746268656714,
      "grad_norm": 0.3643544912338257,
      "kl": 0.0036559104919433594,
      "learning_rate": 2.6657189421854562e-06,
      "loss": 0.0001,
      "reward": 0.7645089607685804,
      "reward_std": 0.16924317786470056,
      "rewards/accuracy_reward": 0.7645089607685804,
      "rewards/format_reward": 0.0,
      "step": 20
    },
    {
      "completion_length": 604.2723560333252,
      "epoch": 0.31343283582089554,
      "grad_norm": 0.22008363902568817,
      "kl": 0.0029211044311523438,
      "learning_rate": 2.6147172382160914e-06,
      "loss": 0.0001,
      "reward": 0.7466518208384514,
      "reward_std": 0.14925407292321324,
      "rewards/accuracy_reward": 0.7466518208384514,
      "rewards/format_reward": 0.0,
      "step": 21
    },
    {
      "completion_length": 604.7009181976318,
      "epoch": 0.3283582089552239,
      "grad_norm": 0.5150439739227295,
      "kl": 0.0036449432373046875,
      "learning_rate": 2.5606601717798212e-06,
      "loss": 0.0001,
      "reward": 0.7466518171131611,
      "reward_std": 0.20275325747206807,
      "rewards/accuracy_reward": 0.7466518171131611,
      "rewards/format_reward": 0.0,
      "step": 22
    },
    {
      "completion_length": 539.8783779144287,
      "epoch": 0.34328358208955223,
      "grad_norm": 0.08835027366876602,
      "kl": 0.0028405189514160156,
      "learning_rate": 2.5036959095382875e-06,
      "loss": 0.0001,
      "reward": 0.8080357424914837,
      "reward_std": 0.151585946790874,
      "rewards/accuracy_reward": 0.8080357424914837,
      "rewards/format_reward": 0.0,
      "step": 23
    },
    {
      "completion_length": 617.0491275787354,
      "epoch": 0.3582089552238806,
      "grad_norm": 0.11819685250520706,
      "kl": 0.0034465789794921875,
      "learning_rate": 2.4439805865747562e-06,
      "loss": 0.0001,
      "reward": 0.7209821790456772,
      "reward_std": 0.20670274924486876,
      "rewards/accuracy_reward": 0.7209821790456772,
      "rewards/format_reward": 0.0,
      "step": 24
    },
    {
      "completion_length": 561.9810562133789,
      "epoch": 0.373134328358209,
      "grad_norm": 0.1162777692079544,
      "kl": 0.0033864974975585938,
      "learning_rate": 2.3816778784387097e-06,
      "loss": 0.0001,
      "reward": 0.8359375335276127,
      "reward_std": 0.12302456051111221,
      "rewards/accuracy_reward": 0.8359375335276127,
      "rewards/format_reward": 0.0,
      "step": 25
    },
    {
      "completion_length": 603.2611846923828,
      "epoch": 0.3880597014925373,
      "grad_norm": 0.11348129063844681,
      "kl": 0.003231048583984375,
      "learning_rate": 2.316958552522541e-06,
      "loss": 0.0001,
      "reward": 0.7555803954601288,
      "reward_std": 0.19700363464653492,
      "rewards/accuracy_reward": 0.7555803954601288,
      "rewards/format_reward": 0.0,
      "step": 26
    },
    {
      "completion_length": 576.3515892028809,
      "epoch": 0.40298507462686567,
      "grad_norm": 0.10333164781332016,
      "kl": 0.003551483154296875,
      "learning_rate": 2.25e-06,
      "loss": 0.0001,
      "reward": 0.7332589626312256,
      "reward_std": 0.16596951289102435,
      "rewards/accuracy_reward": 0.7332589626312256,
      "rewards/format_reward": 0.0,
      "step": 27
    },
    {
      "completion_length": 576.9352893829346,
      "epoch": 0.417910447761194,
      "grad_norm": 0.08400996029376984,
      "kl": 0.003052234649658203,
      "learning_rate": 2.18098574960932e-06,
      "loss": 0.0001,
      "reward": 0.7656250335276127,
      "reward_std": 0.16529620438814163,
      "rewards/accuracy_reward": 0.7656250335276127,
      "rewards/format_reward": 0.0,
      "step": 28
    },
    {
      "completion_length": 572.9832820892334,
      "epoch": 0.43283582089552236,
      "grad_norm": 0.09134360402822495,
      "kl": 0.0033164024353027344,
      "learning_rate": 2.1101049646137005e-06,
      "loss": 0.0001,
      "reward": 0.7243303973227739,
      "reward_std": 0.18314252886921167,
      "rewards/accuracy_reward": 0.7243303973227739,
      "rewards/format_reward": 0.0,
      "step": 29
    },
    {
      "completion_length": 598.8917636871338,
      "epoch": 0.44776119402985076,
      "grad_norm": 0.0819033682346344,
      "kl": 0.003021240234375,
      "learning_rate": 2.03755192431795e-06,
      "loss": 0.0001,
      "reward": 0.772321455180645,
      "reward_std": 0.20655824523419142,
      "rewards/accuracy_reward": 0.772321455180645,
      "rewards/format_reward": 0.0,
      "step": 30
    },
    {
      "completion_length": 568.7176551818848,
      "epoch": 0.4626865671641791,
      "grad_norm": 0.1345580369234085,
      "kl": 0.004252433776855469,
      "learning_rate": 1.963525491562421e-06,
      "loss": 0.0002,
      "reward": 0.7767857536673546,
      "reward_std": 0.18306655017659068,
      "rewards/accuracy_reward": 0.7767857536673546,
      "rewards/format_reward": 0.0,
      "step": 31
    },
    {
      "completion_length": 579.5647583007812,
      "epoch": 0.47761194029850745,
      "grad_norm": 0.08470705151557922,
      "kl": 0.0035343170166015625,
      "learning_rate": 1.888228567653781e-06,
      "loss": 0.0001,
      "reward": 0.7957589663565159,
      "reward_std": 0.17081871908158064,
      "rewards/accuracy_reward": 0.7957589663565159,
      "rewards/format_reward": 0.0,
      "step": 32
    },
    {
      "completion_length": 609.633955001831,
      "epoch": 0.4925373134328358,
      "grad_norm": 0.07784049212932587,
      "kl": 0.0028290748596191406,
      "learning_rate": 1.8118675362266389e-06,
      "loss": 0.0001,
      "reward": 0.7578125335276127,
      "reward_std": 0.1949409432709217,
      "rewards/accuracy_reward": 0.7578125335276127,
      "rewards/format_reward": 0.0,
      "step": 33
    },
    {
      "completion_length": 622.8817157745361,
      "epoch": 0.5074626865671642,
      "grad_norm": 0.08559560030698776,
      "kl": 0.0029349327087402344,
      "learning_rate": 1.7346516975603465e-06,
      "loss": 0.0001,
      "reward": 0.7488839663565159,
      "reward_std": 0.19219930842518806,
      "rewards/accuracy_reward": 0.7488839663565159,
      "rewards/format_reward": 0.0,
      "step": 34
    },
    {
      "completion_length": 612.6897583007812,
      "epoch": 0.5223880597014925,
      "grad_norm": 0.07799447327852249,
      "kl": 0.002971649169921875,
      "learning_rate": 1.6567926949014804e-06,
      "loss": 0.0001,
      "reward": 0.7533482573926449,
      "reward_std": 0.1705156215466559,
      "rewards/accuracy_reward": 0.7533482573926449,
      "rewards/format_reward": 0.0,
      "step": 35
    },
    {
      "completion_length": 575.945333480835,
      "epoch": 0.5373134328358209,
      "grad_norm": 0.11088297516107559,
      "kl": 0.00347900390625,
      "learning_rate": 1.578503934364416e-06,
      "loss": 0.0001,
      "reward": 0.753348246216774,
      "reward_std": 0.18251199880614877,
      "rewards/accuracy_reward": 0.753348246216774,
      "rewards/format_reward": 0.0,
      "step": 36
    },
    {
      "completion_length": 575.522346496582,
      "epoch": 0.5522388059701493,
      "grad_norm": 0.07977250218391418,
      "kl": 0.003589630126953125,
      "learning_rate": 1.5e-06,
      "loss": 0.0001,
      "reward": 0.768973246216774,
      "reward_std": 0.172359649091959,
      "rewards/accuracy_reward": 0.768973246216774,
      "rewards/format_reward": 0.0,
      "step": 37
    },
    {
      "completion_length": 574.8672142028809,
      "epoch": 0.5671641791044776,
      "grad_norm": 0.07976463437080383,
      "kl": 0.0032825469970703125,
      "learning_rate": 1.4214960656355842e-06,
      "loss": 0.0001,
      "reward": 0.8125000409781933,
      "reward_std": 0.12335904035717249,
      "rewards/accuracy_reward": 0.8125000409781933,
      "rewards/format_reward": 0.0,
      "step": 38
    },
    {
      "completion_length": 568.9866352081299,
      "epoch": 0.582089552238806,
      "grad_norm": 0.11306460946798325,
      "kl": 0.004673480987548828,
      "learning_rate": 1.3432073050985201e-06,
      "loss": 0.0002,
      "reward": 0.7723214589059353,
      "reward_std": 0.15484962752088904,
      "rewards/accuracy_reward": 0.7723214589059353,
      "rewards/format_reward": 0.0,
      "step": 39
    },
    {
      "completion_length": 607.1752510070801,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.1425676792860031,
      "kl": 0.0033125877380371094,
      "learning_rate": 1.2653483024396534e-06,
      "loss": 0.0001,
      "reward": 0.772321455180645,
      "reward_std": 0.17449467908591032,
      "rewards/accuracy_reward": 0.772321455180645,
      "rewards/format_reward": 0.0,
      "step": 40
    },
    {
      "completion_length": 534.8303833007812,
      "epoch": 0.6119402985074627,
      "grad_norm": 0.0919041708111763,
      "kl": 0.004061698913574219,
      "learning_rate": 1.1881324637733612e-06,
      "loss": 0.0002,
      "reward": 0.7745536006987095,
      "reward_std": 0.16172579256817698,
      "rewards/accuracy_reward": 0.7745536006987095,
      "rewards/format_reward": 0.0,
      "step": 41
    },
    {
      "completion_length": 560.0078372955322,
      "epoch": 0.6268656716417911,
      "grad_norm": 0.0930609181523323,
      "kl": 0.0036792755126953125,
      "learning_rate": 1.1117714323462188e-06,
      "loss": 0.0001,
      "reward": 0.8158482536673546,
      "reward_std": 0.170708823017776,
      "rewards/accuracy_reward": 0.8158482536673546,
      "rewards/format_reward": 0.0,
      "step": 42
    },
    {
      "completion_length": 601.4565029144287,
      "epoch": 0.6417910447761194,
      "grad_norm": 0.08595103770494461,
      "kl": 0.0032434463500976562,
      "learning_rate": 1.036474508437579e-06,
      "loss": 0.0001,
      "reward": 0.7522321753203869,
      "reward_std": 0.16597021045163274,
      "rewards/accuracy_reward": 0.7522321753203869,
      "rewards/format_reward": 0.0,
      "step": 43
    },
    {
      "completion_length": 553.7667694091797,
      "epoch": 0.6567164179104478,
      "grad_norm": 0.09559870511293411,
      "kl": 0.004054069519042969,
      "learning_rate": 9.624480756820497e-07,
      "loss": 0.0002,
      "reward": 0.7767857499420643,
      "reward_std": 0.14699635887518525,
      "rewards/accuracy_reward": 0.7767857499420643,
      "rewards/format_reward": 0.0,
      "step": 44
    },
    {
      "completion_length": 572.5535984039307,
      "epoch": 0.6716417910447762,
      "grad_norm": 0.07558111101388931,
      "kl": 0.0035276412963867188,
      "learning_rate": 8.898950353863e-07,
      "loss": 0.0001,
      "reward": 0.7901786155998707,
      "reward_std": 0.15063981525599957,
      "rewards/accuracy_reward": 0.7901786155998707,
      "rewards/format_reward": 0.0,
      "step": 45
    },
    {
      "completion_length": 627.7455635070801,
      "epoch": 0.6865671641791045,
      "grad_norm": 0.08073041588068008,
      "kl": 0.0030755996704101562,
      "learning_rate": 8.190142503906799e-07,
      "loss": 0.0001,
      "reward": 0.7421875447034836,
      "reward_std": 0.19877350237220526,
      "rewards/accuracy_reward": 0.7421875447034836,
      "rewards/format_reward": 0.0,
      "step": 46
    },
    {
      "completion_length": 561.7902069091797,
      "epoch": 0.7014925373134329,
      "grad_norm": 0.08770101517438889,
      "kl": 0.0040454864501953125,
      "learning_rate": 7.500000000000003e-07,
      "loss": 0.0002,
      "reward": 0.812500037252903,
      "reward_std": 0.18986603012308478,
      "rewards/accuracy_reward": 0.812500037252903,
      "rewards/format_reward": 0.0,
      "step": 47
    },
    {
      "completion_length": 582.2556076049805,
      "epoch": 0.7164179104477612,
      "grad_norm": 0.08533724397420883,
      "kl": 0.0029554367065429688,
      "learning_rate": 6.830414474774594e-07,
      "loss": 0.0001,
      "reward": 0.8058036006987095,
      "reward_std": 0.1576297921128571,
      "rewards/accuracy_reward": 0.8058036006987095,
      "rewards/format_reward": 0.0,
      "step": 48
    },
    {
      "completion_length": 613.2410888671875,
      "epoch": 0.7313432835820896,
      "grad_norm": 0.08394148200750351,
      "kl": 0.0036144256591796875,
      "learning_rate": 6.183221215612905e-07,
      "loss": 0.0001,
      "reward": 0.7667411081492901,
      "reward_std": 0.20444503612816334,
      "rewards/accuracy_reward": 0.7667411081492901,
      "rewards/format_reward": 0.0,
      "step": 49
    },
    {
      "completion_length": 596.3806076049805,
      "epoch": 0.746268656716418,
      "grad_norm": 0.09445685148239136,
      "kl": 0.0038156509399414062,
      "learning_rate": 5.560194134252441e-07,
      "loss": 0.0002,
      "reward": 0.7154018096625805,
      "reward_std": 0.16886480245739222,
      "rewards/accuracy_reward": 0.7154018096625805,
      "rewards/format_reward": 0.0,
      "step": 50
    },
    {
      "completion_length": 588.2489070892334,
      "epoch": 0.7611940298507462,
      "grad_norm": 0.1409761756658554,
      "kl": 0.0034036636352539062,
      "learning_rate": 4.963040904617131e-07,
      "loss": 0.0001,
      "reward": 0.7779018171131611,
      "reward_std": 0.15530426986515522,
      "rewards/accuracy_reward": 0.7779018171131611,
      "rewards/format_reward": 0.0,
      "step": 51
    },
    {
      "completion_length": 582.168550491333,
      "epoch": 0.7761194029850746,
      "grad_norm": 0.08169642090797424,
      "kl": 0.0034346580505371094,
      "learning_rate": 4.3933982822017883e-07,
      "loss": 0.0001,
      "reward": 0.7421875409781933,
      "reward_std": 0.17119849054142833,
      "rewards/accuracy_reward": 0.7421875409781933,
      "rewards/format_reward": 0.0,
      "step": 52
    },
    {
      "completion_length": 590.315881729126,
      "epoch": 0.7910447761194029,
      "grad_norm": 0.09575485438108444,
      "kl": 0.0036096572875976562,
      "learning_rate": 3.852827617839085e-07,
      "loss": 0.0001,
      "reward": 0.776785746216774,
      "reward_std": 0.20339447353035212,
      "rewards/accuracy_reward": 0.776785746216774,
      "rewards/format_reward": 0.0,
      "step": 53
    },
    {
      "completion_length": 571.912971496582,
      "epoch": 0.8059701492537313,
      "grad_norm": 0.07775302976369858,
      "kl": 0.0041255950927734375,
      "learning_rate": 3.3428105781454364e-07,
      "loss": 0.0002,
      "reward": 0.7544643320143223,
      "reward_std": 0.14612831640988588,
      "rewards/accuracy_reward": 0.7544643320143223,
      "rewards/format_reward": 0.0,
      "step": 54
    },
    {
      "completion_length": 575.9989128112793,
      "epoch": 0.8208955223880597,
      "grad_norm": 0.13639913499355316,
      "kl": 0.0060577392578125,
      "learning_rate": 2.86474508437579e-07,
      "loss": 0.0002,
      "reward": 0.7745536118745804,
      "reward_std": 0.1799168661236763,
      "rewards/accuracy_reward": 0.7745536118745804,
      "rewards/format_reward": 0.0,
      "step": 55
    },
    {
      "completion_length": 595.2321643829346,
      "epoch": 0.835820895522388,
      "grad_norm": 0.09935541450977325,
      "kl": 0.0034551620483398438,
      "learning_rate": 2.419941480818641e-07,
      "loss": 0.0001,
      "reward": 0.7566964663565159,
      "reward_std": 0.1705572777427733,
      "rewards/accuracy_reward": 0.7566964663565159,
      "rewards/format_reward": 0.0,
      "step": 56
    },
    {
      "completion_length": 578.415205001831,
      "epoch": 0.8507462686567164,
      "grad_norm": 0.08712073415517807,
      "kl": 0.0035562515258789062,
      "learning_rate": 2.0096189432334195e-07,
      "loss": 0.0001,
      "reward": 0.7868303917348385,
      "reward_std": 0.1490265424363315,
      "rewards/accuracy_reward": 0.7868303917348385,
      "rewards/format_reward": 0.0,
      "step": 57
    },
    {
      "completion_length": 570.0915451049805,
      "epoch": 0.8656716417910447,
      "grad_norm": 0.08893033117055893,
      "kl": 0.003448486328125,
      "learning_rate": 1.634902137174483e-07,
      "loss": 0.0001,
      "reward": 0.7868303954601288,
      "reward_std": 0.13831894285976887,
      "rewards/accuracy_reward": 0.7868303954601288,
      "rewards/format_reward": 0.0,
      "step": 58
    },
    {
      "completion_length": 570.3884201049805,
      "epoch": 0.8805970149253731,
      "grad_norm": 0.0921068862080574,
      "kl": 0.0033750534057617188,
      "learning_rate": 1.2968181353609853e-07,
      "loss": 0.0001,
      "reward": 0.8102678991854191,
      "reward_std": 0.15646472247317433,
      "rewards/accuracy_reward": 0.8102678991854191,
      "rewards/format_reward": 0.0,
      "step": 59
    },
    {
      "completion_length": 562.8716716766357,
      "epoch": 0.8955223880597015,
      "grad_norm": 0.06974907219409943,
      "kl": 0.0035638809204101562,
      "learning_rate": 9.962936025419756e-08,
      "loss": 0.0001,
      "reward": 0.7857143245637417,
      "reward_std": 0.1675086230970919,
      "rewards/accuracy_reward": 0.7857143245637417,
      "rewards/format_reward": 0.0,
      "step": 60
    },
    {
      "completion_length": 584.5279273986816,
      "epoch": 0.9104477611940298,
      "grad_norm": 0.13824282586574554,
      "kl": 0.003696441650390625,
      "learning_rate": 7.341522555726971e-08,
      "loss": 0.0001,
      "reward": 0.7511161006987095,
      "reward_std": 0.19415182201191783,
      "rewards/accuracy_reward": 0.7511161006987095,
      "rewards/format_reward": 0.0,
      "step": 61
    },
    {
      "completion_length": 549.1629695892334,
      "epoch": 0.9253731343283582,
      "grad_norm": 0.10054602473974228,
      "kl": 0.004033088684082031,
      "learning_rate": 5.11112605663977e-08,
      "loss": 0.0002,
      "reward": 0.8381696827709675,
      "reward_std": 0.1603721366263926,
      "rewards/accuracy_reward": 0.8381696827709675,
      "rewards/format_reward": 0.0,
      "step": 62
    },
    {
      "completion_length": 593.0893096923828,
      "epoch": 0.9402985074626866,
      "grad_norm": 0.08826031535863876,
      "kl": 0.0032329559326171875,
      "learning_rate": 3.277859889929147e-08,
      "loss": 0.0001,
      "reward": 0.812500037252903,
      "reward_std": 0.1807763264514506,
      "rewards/accuracy_reward": 0.812500037252903,
      "rewards/format_reward": 0.0,
      "step": 63
    },
    {
      "completion_length": 549.9207801818848,
      "epoch": 0.9552238805970149,
      "grad_norm": 0.0768972784280777,
      "kl": 0.0036096572875976562,
      "learning_rate": 1.846748910729351e-08,
      "loss": 0.0001,
      "reward": 0.8225446753203869,
      "reward_std": 0.12298994930461049,
      "rewards/accuracy_reward": 0.8225446753203869,
      "rewards/format_reward": 0.0,
      "step": 64
    },
    {
      "completion_length": 544.910737991333,
      "epoch": 0.9701492537313433,
      "grad_norm": 0.07084512710571289,
      "kl": 0.0035963058471679688,
      "learning_rate": 8.217156947590065e-09,
      "loss": 0.0001,
      "reward": 0.8169643096625805,
      "reward_std": 0.12426057457923889,
      "rewards/accuracy_reward": 0.8169643096625805,
      "rewards/format_reward": 0.0,
      "step": 65
    },
    {
      "completion_length": 587.8024768829346,
      "epoch": 0.9850746268656716,
      "grad_norm": 0.08717910200357437,
      "kl": 0.003803253173828125,
      "learning_rate": 2.0556978681392503e-09,
      "loss": 0.0002,
      "reward": 0.7444196753203869,
      "reward_std": 0.17243521055206656,
      "rewards/accuracy_reward": 0.7444196753203869,
      "rewards/format_reward": 0.0,
      "step": 66
    },
    {
      "completion_length": 574.7916831970215,
      "epoch": 1.0,
      "grad_norm": 0.0701616182923317,
      "kl": 0.0031533241271972656,
      "learning_rate": 0.0,
      "loss": 0.0001,
      "reward": 0.7291666679084301,
      "reward_std": 0.0578637570142746,
      "rewards/accuracy_reward": 0.7291666679084301,
      "rewards/format_reward": 0.0,
      "step": 67
    },
    {
      "epoch": 1.0,
      "step": 67,
      "total_flos": 0.0,
      "train_loss": 0.00012216681254848273,
      "train_runtime": 12383.6526,
      "train_samples_per_second": 0.606,
      "train_steps_per_second": 0.005
    }
  ],
  "logging_steps": 1,
  "max_steps": 67,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}