{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5296343001261034,
  "eval_steps": 10,
  "global_step": 210,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "computed_total": 0.45715999603271484,
      "epoch": 0.0025220680958385876,
      "grad_norm": 102.097412109375,
      "kl_loss": 6.5083294309431494e-09,
      "learning_rate": 0.0,
      "loss": 4.3237,
      "sft_loss": 0.45650917291641235,
      "step": 1
    },
    {
      "computed_total": 0.555546760559082,
      "epoch": 0.005044136191677175,
      "grad_norm": 87.49225616455078,
      "kl_loss": 5.4945092919922445e-09,
      "learning_rate": 4.000000000000001e-06,
      "loss": 4.3036,
      "sft_loss": 0.5549973249435425,
      "step": 2
    },
    {
      "computed_total": 0.5587829947471619,
      "epoch": 0.007566204287515763,
      "grad_norm": 64.52155303955078,
      "kl_loss": 3.4872137177899276e-08,
      "learning_rate": 8.000000000000001e-06,
      "loss": 3.5946,
      "sft_loss": 0.5552957653999329,
      "step": 3
    },
    {
      "computed_total": 0.4053139090538025,
      "epoch": 0.01008827238335435,
      "grad_norm": 79.32300567626953,
      "kl_loss": 7.649157964806363e-07,
      "learning_rate": 1.2e-05,
      "loss": 3.3472,
      "sft_loss": 0.3288223147392273,
      "step": 4
    },
    {
      "computed_total": 0.3922015130519867,
      "epoch": 0.012610340479192938,
      "grad_norm": 49.438594818115234,
      "kl_loss": 9.65095068750088e-07,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 3.0682,
      "sft_loss": 0.29569199681282043,
      "step": 5
    },
    {
      "computed_total": 0.3126932382583618,
      "epoch": 0.015132408575031526,
      "grad_norm": 36.26791000366211,
      "kl_loss": 9.634337629904621e-07,
      "learning_rate": 2e-05,
      "loss": 2.7815,
      "sft_loss": 0.21634985506534576,
      "step": 6
    },
    {
      "computed_total": 0.2588472068309784,
      "epoch": 0.017654476670870115,
      "grad_norm": 28.21232795715332,
      "kl_loss": 7.276251494658936e-07,
      "learning_rate": 1.9948979591836737e-05,
      "loss": 2.499,
      "sft_loss": 0.18608468770980835,
      "step": 7
    },
    {
      "computed_total": 0.32910704612731934,
      "epoch": 0.0201765447667087,
      "grad_norm": 24.3712158203125,
      "kl_loss": 8.822735821922834e-07,
      "learning_rate": 1.9897959183673473e-05,
      "loss": 2.4776,
      "sft_loss": 0.24087969958782196,
      "step": 8
    },
    {
      "computed_total": 0.3658309280872345,
      "epoch": 0.02269861286254729,
      "grad_norm": 31.98260498046875,
      "kl_loss": 7.195029070317105e-07,
      "learning_rate": 1.9846938775510205e-05,
      "loss": 2.3718,
      "sft_loss": 0.2938806414604187,
      "step": 9
    },
    {
      "computed_total": 0.22178040444850922,
      "epoch": 0.025220680958385876,
      "grad_norm": 34.0799674987793,
      "kl_loss": 6.166483217384666e-07,
      "learning_rate": 1.979591836734694e-05,
      "loss": 2.4129,
      "sft_loss": 0.1601155698299408,
      "step": 10
    },
    {
      "computed_total": 2.1096696853637695,
      "epoch": 0.025220680958385876,
      "eval_loss": 1.9816868305206299,
      "eval_runtime": 67.6013,
      "eval_samples_per_second": 10.429,
      "eval_steps_per_second": 5.222,
      "kl_loss": 3.9526437944914505e-07,
      "sft_loss": 2.070143222808838,
      "step": 10
    },
    {
      "computed_total": 0.4089103043079376,
      "epoch": 0.027742749054224466,
      "grad_norm": 26.954879760742188,
      "kl_loss": 4.996533107259893e-07,
      "learning_rate": 1.9744897959183677e-05,
      "loss": 2.277,
      "sft_loss": 0.35894498229026794,
      "step": 11
    },
    {
      "computed_total": 0.23354634642601013,
      "epoch": 0.03026481715006305,
      "grad_norm": 18.464841842651367,
      "kl_loss": 5.376580816118803e-07,
      "learning_rate": 1.969387755102041e-05,
      "loss": 2.2583,
      "sft_loss": 0.17978054285049438,
      "step": 12
    },
    {
      "computed_total": 0.30147990584373474,
      "epoch": 0.03278688524590164,
      "grad_norm": 26.595727920532227,
      "kl_loss": 5.573554062721087e-07,
      "learning_rate": 1.9642857142857145e-05,
      "loss": 2.2624,
      "sft_loss": 0.24574437737464905,
      "step": 13
    },
    {
      "computed_total": 0.24872392416000366,
      "epoch": 0.03530895334174023,
      "grad_norm": 16.613079071044922,
      "kl_loss": 4.973371687810868e-07,
      "learning_rate": 1.9591836734693877e-05,
      "loss": 2.1151,
      "sft_loss": 0.19899021089076996,
      "step": 14
    },
    {
      "computed_total": 0.19776779413223267,
      "epoch": 0.03783102143757881,
      "grad_norm": 15.080662727355957,
      "kl_loss": 4.196848806259368e-07,
      "learning_rate": 1.9540816326530613e-05,
      "loss": 2.0897,
      "sft_loss": 0.15579931437969208,
      "step": 15
    },
    {
      "computed_total": 0.284699946641922,
      "epoch": 0.0403530895334174,
      "grad_norm": 11.973522186279297,
      "kl_loss": 4.3317794506947394e-07,
      "learning_rate": 1.948979591836735e-05,
      "loss": 2.216,
      "sft_loss": 0.2413821518421173,
      "step": 16
    },
    {
      "computed_total": 0.21680015325546265,
      "epoch": 0.04287515762925599,
      "grad_norm": 11.934453964233398,
      "kl_loss": 3.477235850368743e-07,
      "learning_rate": 1.9438775510204085e-05,
      "loss": 1.9235,
      "sft_loss": 0.18202780187129974,
      "step": 17
    },
    {
      "computed_total": 0.18643273413181305,
      "epoch": 0.04539722572509458,
      "grad_norm": 10.074520111083984,
      "kl_loss": 2.590618919384724e-07,
      "learning_rate": 1.9387755102040817e-05,
      "loss": 1.8522,
      "sft_loss": 0.16052654385566711,
      "step": 18
    },
    {
      "computed_total": 0.22932234406471252,
      "epoch": 0.04791929382093316,
      "grad_norm": 15.449592590332031,
      "kl_loss": 2.511018522000086e-07,
      "learning_rate": 1.9336734693877553e-05,
      "loss": 1.7982,
      "sft_loss": 0.20421215891838074,
      "step": 19
    },
    {
      "computed_total": 0.29722726345062256,
      "epoch": 0.05044136191677175,
      "grad_norm": 12.395018577575684,
      "kl_loss": 2.6658420892999857e-07,
      "learning_rate": 1.928571428571429e-05,
      "loss": 1.817,
      "sft_loss": 0.27056884765625,
      "step": 20
    },
    {
      "computed_total": 1.6116889715194702,
      "epoch": 0.05044136191677175,
      "eval_loss": 1.5718247890472412,
      "eval_runtime": 69.8078,
      "eval_samples_per_second": 10.099,
      "eval_steps_per_second": 5.057,
      "kl_loss": 3.4212027344437956e-07,
      "sft_loss": 1.577476978302002,
      "step": 20
    },
    {
      "computed_total": 0.1908552497625351,
      "epoch": 0.05296343001261034,
      "grad_norm": 23.406896591186523,
      "kl_loss": 2.7672922442434356e-07,
      "learning_rate": 1.9234693877551024e-05,
      "loss": 1.7865,
      "sft_loss": 0.163182333111763,
      "step": 21
    },
    {
      "computed_total": 0.2098204642534256,
      "epoch": 0.05548549810844893,
      "grad_norm": 12.387871742248535,
      "kl_loss": 1.9662114425500477e-07,
      "learning_rate": 1.9183673469387756e-05,
      "loss": 1.6599,
      "sft_loss": 0.19015835225582123,
      "step": 22
    },
    {
      "computed_total": 0.27859383821487427,
      "epoch": 0.058007566204287514,
      "grad_norm": 17.064393997192383,
      "kl_loss": 2.6223989380014245e-07,
      "learning_rate": 1.9132653061224492e-05,
      "loss": 1.7114,
      "sft_loss": 0.25236985087394714,
      "step": 23
    },
    {
      "computed_total": 0.2661365270614624,
      "epoch": 0.0605296343001261,
      "grad_norm": 14.536602973937988,
      "kl_loss": 3.302961033568863e-07,
      "learning_rate": 1.9081632653061225e-05,
      "loss": 1.7381,
      "sft_loss": 0.23310692608356476,
      "step": 24
    },
    {
      "computed_total": 0.21774020791053772,
      "epoch": 0.06305170239596469,
      "grad_norm": 10.612632751464844,
      "kl_loss": 2.2093004758971801e-07,
      "learning_rate": 1.903061224489796e-05,
      "loss": 1.6493,
      "sft_loss": 0.195647194981575,
      "step": 25
    },
    {
      "computed_total": 0.21532197296619415,
      "epoch": 0.06557377049180328,
      "grad_norm": 13.36475944519043,
      "kl_loss": 1.9050941091336426e-07,
      "learning_rate": 1.8979591836734696e-05,
      "loss": 1.65,
      "sft_loss": 0.19627103209495544,
      "step": 26
    },
    {
      "computed_total": 0.21690420806407928,
      "epoch": 0.06809583858764187,
      "grad_norm": 13.064225196838379,
      "kl_loss": 2.1687725393348956e-07,
      "learning_rate": 1.892857142857143e-05,
      "loss": 1.7006,
      "sft_loss": 0.19521647691726685,
      "step": 27
    },
    {
      "computed_total": 0.19565589725971222,
      "epoch": 0.07061790668348046,
      "grad_norm": 12.320195198059082,
      "kl_loss": 2.17976335648018e-07,
      "learning_rate": 1.8877551020408164e-05,
      "loss": 1.6652,
      "sft_loss": 0.17385827004909515,
      "step": 28
    },
    {
      "computed_total": 0.27470341324806213,
      "epoch": 0.07313997477931904,
      "grad_norm": 11.224297523498535,
      "kl_loss": 2.520193618238409e-07,
      "learning_rate": 1.88265306122449e-05,
      "loss": 1.6883,
      "sft_loss": 0.24950148165225983,
      "step": 29
    },
    {
      "computed_total": 0.1886218637228012,
      "epoch": 0.07566204287515763,
      "grad_norm": 14.4179048538208,
      "kl_loss": 3.655400746538362e-07,
      "learning_rate": 1.8775510204081636e-05,
      "loss": 1.7232,
      "sft_loss": 0.1520678550004959,
      "step": 30
    },
    {
      "computed_total": 1.3417655229568481,
      "epoch": 0.07566204287515763,
      "eval_loss": 1.4369149208068848,
      "eval_runtime": 67.0852,
      "eval_samples_per_second": 10.509,
      "eval_steps_per_second": 5.262,
      "kl_loss": 2.1116991888447956e-07,
      "sft_loss": 1.3206485509872437,
      "step": 30
    },
    {
      "computed_total": 0.19783087074756622,
      "epoch": 0.07818411097099622,
      "grad_norm": 14.451069831848145,
      "kl_loss": 1.4020483263266215e-07,
      "learning_rate": 1.8724489795918368e-05,
      "loss": 1.6225,
      "sft_loss": 0.18381038308143616,
      "step": 31
    },
    {
      "computed_total": 0.2664089798927307,
      "epoch": 0.0807061790668348,
      "grad_norm": 10.6675443649292,
      "kl_loss": 2.558473113367654e-07,
      "learning_rate": 1.8673469387755104e-05,
      "loss": 1.6656,
      "sft_loss": 0.24082423746585846,
      "step": 32
    },
    {
      "computed_total": 0.16216596961021423,
      "epoch": 0.0832282471626734,
      "grad_norm": 10.89920711517334,
      "kl_loss": 3.102907157881418e-07,
      "learning_rate": 1.862244897959184e-05,
      "loss": 1.7422,
      "sft_loss": 0.13113689422607422,
      "step": 33
    },
    {
      "computed_total": 0.18052271008491516,
      "epoch": 0.08575031525851198,
      "grad_norm": 10.312054634094238,
      "kl_loss": 2.5258012215090275e-07,
      "learning_rate": 1.8571428571428575e-05,
      "loss": 1.5284,
      "sft_loss": 0.1552647054195404,
      "step": 34
    },
    {
      "computed_total": 0.13403812050819397,
      "epoch": 0.08827238335435057,
      "grad_norm": 9.937405586242676,
      "kl_loss": 1.9699928088812158e-07,
      "learning_rate": 1.8520408163265307e-05,
      "loss": 1.6276,
      "sft_loss": 0.11433819681406021,
      "step": 35
    },
    {
      "computed_total": 0.21145765483379364,
      "epoch": 0.09079445145018916,
      "grad_norm": 10.810340881347656,
      "kl_loss": 2.5251029001083225e-07,
      "learning_rate": 1.8469387755102043e-05,
      "loss": 1.5674,
      "sft_loss": 0.1862066239118576,
      "step": 36
    },
    {
      "computed_total": 0.28667497634887695,
      "epoch": 0.09331651954602774,
      "grad_norm": 13.322245597839355,
      "kl_loss": 2.524957380956039e-07,
      "learning_rate": 1.8418367346938776e-05,
      "loss": 1.6924,
      "sft_loss": 0.2614254057407379,
      "step": 37
    },
    {
      "computed_total": 0.17698872089385986,
      "epoch": 0.09583858764186633,
      "grad_norm": 12.67447566986084,
      "kl_loss": 1.8715130067903374e-07,
      "learning_rate": 1.836734693877551e-05,
      "loss": 1.6326,
      "sft_loss": 0.1582735925912857,
      "step": 38
    },
    {
      "computed_total": 0.218657985329628,
      "epoch": 0.09836065573770492,
      "grad_norm": 10.07801342010498,
      "kl_loss": 2.1152843032723467e-07,
      "learning_rate": 1.8316326530612247e-05,
      "loss": 1.5745,
      "sft_loss": 0.1975051462650299,
      "step": 39
    },
    {
      "computed_total": 0.1585373878479004,
      "epoch": 0.1008827238335435,
      "grad_norm": 10.709808349609375,
      "kl_loss": 1.5567046318665234e-07,
      "learning_rate": 1.826530612244898e-05,
      "loss": 1.4899,
      "sft_loss": 0.14297033846378326,
      "step": 40
    },
    {
      "computed_total": 1.3272018432617188,
      "epoch": 0.1008827238335435,
      "eval_loss": 1.3992165327072144,
      "eval_runtime": 67.3288,
      "eval_samples_per_second": 10.471,
      "eval_steps_per_second": 5.243,
      "kl_loss": 1.587673352787533e-07,
      "sft_loss": 1.3113250732421875,
      "step": 40
    },
    {
      "computed_total": 0.16187940537929535,
      "epoch": 0.1034047919293821,
      "grad_norm": 12.045276641845703,
      "kl_loss": 1.6272514358206536e-07,
      "learning_rate": 1.8214285714285715e-05,
      "loss": 1.4241,
      "sft_loss": 0.1456068903207779,
      "step": 41
    },
    {
      "computed_total": 0.16236107051372528,
      "epoch": 0.10592686002522068,
      "grad_norm": 9.8399658203125,
      "kl_loss": 1.7566540577718115e-07,
      "learning_rate": 1.816326530612245e-05,
      "loss": 1.557,
      "sft_loss": 0.1447945237159729,
      "step": 42
    },
    {
      "computed_total": 0.15767036378383636,
      "epoch": 0.10844892812105927,
      "grad_norm": 12.114518165588379,
      "kl_loss": 1.7348739334011043e-07,
      "learning_rate": 1.8112244897959187e-05,
      "loss": 1.583,
      "sft_loss": 0.14032162725925446,
      "step": 43
    },
    {
      "computed_total": 0.09624896198511124,
      "epoch": 0.11097099621689786,
      "grad_norm": 9.963354110717773,
      "kl_loss": 1.7239732130747143e-07,
      "learning_rate": 1.806122448979592e-05,
      "loss": 1.4214,
      "sft_loss": 0.07900922745466232,
      "step": 44
    },
    {
      "computed_total": 0.2713112533092499,
      "epoch": 0.11349306431273644,
      "grad_norm": 10.126675605773926,
      "kl_loss": 1.4448913532305596e-07,
      "learning_rate": 1.8010204081632655e-05,
      "loss": 1.5783,
      "sft_loss": 0.2568623423576355,
      "step": 45
    },
    {
      "computed_total": 0.1786537766456604,
      "epoch": 0.11601513240857503,
      "grad_norm": 9.506120681762695,
      "kl_loss": 1.1718055503706637e-07,
      "learning_rate": 1.795918367346939e-05,
      "loss": 1.5218,
      "sft_loss": 0.1669357270002365,
      "step": 46
    },
    {
      "computed_total": 0.15938213467597961,
      "epoch": 0.11853720050441362,
      "grad_norm": 11.758268356323242,
      "kl_loss": 1.7643755256813165e-07,
      "learning_rate": 1.7908163265306123e-05,
      "loss": 1.479,
      "sft_loss": 0.1417383849620819,
      "step": 47
    },
    {
      "computed_total": 0.15835891664028168,
      "epoch": 0.1210592686002522,
      "grad_norm": 10.227970123291016,
      "kl_loss": 1.3144477861715131e-07,
      "learning_rate": 1.785714285714286e-05,
      "loss": 1.5222,
      "sft_loss": 0.14521443843841553,
      "step": 48
    },
    {
      "computed_total": 0.1753772348165512,
      "epoch": 0.1235813366960908,
      "grad_norm": 9.49471378326416,
      "kl_loss": 1.2467818066852487e-07,
      "learning_rate": 1.780612244897959e-05,
      "loss": 1.4884,
      "sft_loss": 0.16290941834449768,
      "step": 49
    },
    {
      "computed_total": 0.14545822143554688,
      "epoch": 0.12610340479192939,
      "grad_norm": 9.987504005432129,
      "kl_loss": 1.3850477387222782e-07,
      "learning_rate": 1.7755102040816327e-05,
      "loss": 1.5382,
      "sft_loss": 0.13160774111747742,
      "step": 50
    },
    {
      "computed_total": 1.2876274585723877,
      "epoch": 0.12610340479192939,
      "eval_loss": 1.3761128187179565,
      "eval_runtime": 65.8814,
      "eval_samples_per_second": 10.701,
      "eval_steps_per_second": 5.358,
      "kl_loss": 1.1736652538729686e-07,
      "sft_loss": 1.275890827178955,
      "step": 50
    },
    {
      "computed_total": 0.13572491705417633,
      "epoch": 0.12862547288776796,
      "grad_norm": 8.282198905944824,
      "kl_loss": 1.2732888876598736e-07,
      "learning_rate": 1.7704081632653062e-05,
      "loss": 1.4672,
      "sft_loss": 0.12299202382564545,
      "step": 51
    },
    {
      "computed_total": 0.2022746205329895,
      "epoch": 0.13114754098360656,
      "grad_norm": 8.850266456604004,
      "kl_loss": 9.734989703247265e-08,
      "learning_rate": 1.7653061224489798e-05,
      "loss": 1.4807,
      "sft_loss": 0.19253963232040405,
      "step": 52
    },
    {
      "computed_total": 0.2428017556667328,
      "epoch": 0.13366960907944514,
      "grad_norm": 7.4850568771362305,
      "kl_loss": 1.472172925787163e-07,
      "learning_rate": 1.760204081632653e-05,
      "loss": 1.4415,
      "sft_loss": 0.22808003425598145,
      "step": 53
    },
    {
      "computed_total": 0.22275952994823456,
      "epoch": 0.13619167717528374,
      "grad_norm": 9.871981620788574,
      "kl_loss": 1.480846094636945e-07,
      "learning_rate": 1.7551020408163266e-05,
      "loss": 1.6587,
      "sft_loss": 0.20795106887817383,
      "step": 54
    },
    {
      "computed_total": 0.1835995763540268,
      "epoch": 0.13871374527112232,
      "grad_norm": 8.844654083251953,
      "kl_loss": 1.2977692165350163e-07,
      "learning_rate": 1.7500000000000002e-05,
      "loss": 1.5753,
      "sft_loss": 0.17062188684940338,
      "step": 55
    },
    {
      "computed_total": 0.19904766976833344,
      "epoch": 0.14123581336696092,
      "grad_norm": 7.038422584533691,
      "kl_loss": 1.1099445629270122e-07,
      "learning_rate": 1.7448979591836738e-05,
      "loss": 1.4909,
      "sft_loss": 0.18794822692871094,
      "step": 56
    },
    {
      "computed_total": 0.14884871244430542,
      "epoch": 0.1437578814627995,
      "grad_norm": 7.760448455810547,
      "kl_loss": 1.0769576164193495e-07,
      "learning_rate": 1.7397959183673473e-05,
      "loss": 1.5988,
      "sft_loss": 0.13807913661003113,
      "step": 57
    },
    {
      "computed_total": 0.22467643022537231,
      "epoch": 0.14627994955863807,
      "grad_norm": 8.861936569213867,
      "kl_loss": 1.3167736767627503e-07,
      "learning_rate": 1.7346938775510206e-05,
      "loss": 1.4759,
      "sft_loss": 0.21150869131088257,
      "step": 58
    },
    {
      "computed_total": 0.1402980089187622,
      "epoch": 0.14880201765447668,
      "grad_norm": 8.389931678771973,
      "kl_loss": 1.2574119523378613e-07,
      "learning_rate": 1.729591836734694e-05,
      "loss": 1.4478,
      "sft_loss": 0.12772388756275177,
      "step": 59
    },
    {
      "computed_total": 0.2214464396238327,
      "epoch": 0.15132408575031525,
      "grad_norm": 11.03455924987793,
      "kl_loss": 1.229836925631389e-07,
      "learning_rate": 1.7244897959183674e-05,
      "loss": 1.542,
      "sft_loss": 0.20914806425571442,
      "step": 60
    },
    {
      "computed_total": 1.2934421300888062,
      "epoch": 0.15132408575031525,
      "eval_loss": 1.3563388586044312,
      "eval_runtime": 66.808,
      "eval_samples_per_second": 10.553,
      "eval_steps_per_second": 5.284,
      "kl_loss": 1.0278420603526683e-07,
      "sft_loss": 1.2831636667251587,
      "step": 60
    },
    {
      "computed_total": 0.19582681357860565,
      "epoch": 0.15384615384615385,
      "grad_norm": 8.09454345703125,
      "kl_loss": 1.0761530688796483e-07,
      "learning_rate": 1.719387755102041e-05,
      "loss": 1.4979,
      "sft_loss": 0.18506528437137604,
      "step": 61
    },
    {
      "computed_total": 0.15016137063503265,
      "epoch": 0.15636822194199243,
      "grad_norm": 6.767925262451172,
      "kl_loss": 9.817436819048453e-08,
      "learning_rate": 1.7142857142857142e-05,
      "loss": 1.3421,
      "sft_loss": 0.14034393429756165,
      "step": 62
    },
    {
      "computed_total": 0.23630164563655853,
      "epoch": 0.15889029003783103,
      "grad_norm": 6.9385552406311035,
      "kl_loss": 1.2005209271137574e-07,
      "learning_rate": 1.7091836734693878e-05,
      "loss": 1.4917,
      "sft_loss": 0.224296435713768,
      "step": 63
    },
    {
      "computed_total": 0.23200978338718414,
      "epoch": 0.1614123581336696,
      "grad_norm": 7.448540210723877,
      "kl_loss": 8.879172952447334e-08,
      "learning_rate": 1.7040816326530613e-05,
      "loss": 1.4776,
      "sft_loss": 0.22313061356544495,
      "step": 64
    },
    {
      "computed_total": 0.16305182874202728,
      "epoch": 0.16393442622950818,
      "grad_norm": 7.1528401374816895,
      "kl_loss": 9.930875677355289e-08,
      "learning_rate": 1.698979591836735e-05,
      "loss": 1.3943,
      "sft_loss": 0.1531209498643875,
      "step": 65
    },
    {
      "computed_total": 0.14623767137527466,
      "epoch": 0.1664564943253468,
      "grad_norm": 7.3966474533081055,
      "kl_loss": 7.404720037129664e-08,
      "learning_rate": 1.6938775510204085e-05,
      "loss": 1.5426,
      "sft_loss": 0.1388329565525055,
      "step": 66
    },
    {
      "computed_total": 0.14071843028068542,
      "epoch": 0.16897856242118536,
      "grad_norm": 7.82870626449585,
      "kl_loss": 9.07349146928027e-08,
      "learning_rate": 1.6887755102040817e-05,
      "loss": 1.4386,
      "sft_loss": 0.13164493441581726,
      "step": 67
    },
    {
      "computed_total": 0.14310967922210693,
      "epoch": 0.17150063051702397,
      "grad_norm": 6.757770538330078,
      "kl_loss": 7.598376328132872e-08,
      "learning_rate": 1.6836734693877553e-05,
      "loss": 1.3732,
      "sft_loss": 0.13551130890846252,
      "step": 68
    },
    {
      "computed_total": 0.14132973551750183,
      "epoch": 0.17402269861286254,
      "grad_norm": 8.344841957092285,
      "kl_loss": 1.0905234404390285e-07,
      "learning_rate": 1.678571428571429e-05,
      "loss": 1.4286,
      "sft_loss": 0.13042449951171875,
      "step": 69
    },
    {
      "computed_total": 0.13716241717338562,
      "epoch": 0.17654476670870115,
      "grad_norm": 7.953212738037109,
      "kl_loss": 7.554955061550572e-08,
      "learning_rate": 1.673469387755102e-05,
      "loss": 1.3187,
      "sft_loss": 0.12960746884346008,
      "step": 70
    },
    {
      "computed_total": 1.2660552263259888,
      "epoch": 0.17654476670870115,
      "eval_loss": 1.348509669303894,
      "eval_runtime": 66.6809,
      "eval_samples_per_second": 10.573,
      "eval_steps_per_second": 5.294,
      "kl_loss": 5.885167198016461e-08,
      "sft_loss": 1.2601701021194458,
      "step": 70
    },
    {
      "computed_total": 0.21419619023799896,
      "epoch": 0.17906683480453972,
      "grad_norm": 7.022055625915527,
      "kl_loss": 5.5535913645599067e-08,
      "learning_rate": 1.6683673469387757e-05,
      "loss": 1.3962,
      "sft_loss": 0.2086426019668579,
      "step": 71
    },
    {
      "computed_total": 0.1429305225610733,
      "epoch": 0.18158890290037832,
      "grad_norm": 6.416485786437988,
      "kl_loss": 5.6032408934925115e-08,
      "learning_rate": 1.6632653061224492e-05,
      "loss": 1.4386,
      "sft_loss": 0.13732728362083435,
      "step": 72
    },
    {
      "computed_total": 0.1595781445503235,
      "epoch": 0.1841109709962169,
      "grad_norm": 6.244477272033691,
      "kl_loss": 1.0954471463264781e-07,
      "learning_rate": 1.6581632653061225e-05,
      "loss": 1.4295,
      "sft_loss": 0.14862367510795593,
      "step": 73
    },
    {
      "computed_total": 0.1830446571111679,
      "epoch": 0.18663303909205547,
      "grad_norm": 6.70966100692749,
      "kl_loss": 1.011455665889116e-07,
      "learning_rate": 1.653061224489796e-05,
      "loss": 1.5022,
      "sft_loss": 0.17293010652065277,
      "step": 74
    },
    {
      "computed_total": 0.13835778832435608,
      "epoch": 0.18915510718789408,
      "grad_norm": 7.2905378341674805,
      "kl_loss": 1.1240518205113403e-07,
      "learning_rate": 1.6479591836734696e-05,
      "loss": 1.4917,
      "sft_loss": 0.12711727619171143,
      "step": 75
    },
    {
      "computed_total": 0.17067813873291016,
      "epoch": 0.19167717528373265,
      "grad_norm": 6.712996482849121,
      "kl_loss": 7.433935422795912e-08,
      "learning_rate": 1.642857142857143e-05,
      "loss": 1.3446,
      "sft_loss": 0.16324420273303986,
      "step": 76
    },
    {
      "computed_total": 0.1600562483072281,
      "epoch": 0.19419924337957126,
      "grad_norm": 6.522639274597168,
      "kl_loss": 6.566633459215154e-08,
      "learning_rate": 1.6377551020408164e-05,
      "loss": 1.2858,
      "sft_loss": 0.1534896194934845,
      "step": 77
    },
    {
      "computed_total": 0.15458209812641144,
      "epoch": 0.19672131147540983,
      "grad_norm": 6.488280773162842,
      "kl_loss": 8.56703010754245e-08,
      "learning_rate": 1.63265306122449e-05,
      "loss": 1.3759,
      "sft_loss": 0.14601506292819977,
      "step": 78
    },
    {
      "computed_total": 0.14832951128482819,
      "epoch": 0.19924337957124844,
      "grad_norm": 7.430500507354736,
      "kl_loss": 1.0060924182653253e-07,
      "learning_rate": 1.6275510204081636e-05,
      "loss": 1.1762,
      "sft_loss": 0.1382685899734497,
      "step": 79
    },
    {
      "computed_total": 0.1753160059452057,
      "epoch": 0.201765447667087,
      "grad_norm": 8.094470024108887,
      "kl_loss": 8.996708089625827e-08,
      "learning_rate": 1.6224489795918368e-05,
      "loss": 1.5172,
      "sft_loss": 0.16631929576396942,
      "step": 80
    },
    {
      "computed_total": 1.2994036674499512,
      "epoch": 0.201765447667087,
      "eval_loss": 1.349037766456604,
      "eval_runtime": 64.527,
      "eval_samples_per_second": 10.926,
      "eval_steps_per_second": 5.471,
      "kl_loss": 7.120447520492235e-08,
      "sft_loss": 1.2922831773757935,
      "step": 80
    },
    {
      "computed_total": 0.1462080031633377,
      "epoch": 0.2042875157629256,
      "grad_norm": 7.84616756439209,
      "kl_loss": 9.43423970056756e-08,
      "learning_rate": 1.6173469387755104e-05,
      "loss": 1.3986,
      "sft_loss": 0.13677376508712769,
      "step": 81
    },
    {
      "computed_total": 0.20616410672664642,
      "epoch": 0.2068095838587642,
      "grad_norm": 7.06512975692749,
      "kl_loss": 7.940719370935767e-08,
      "learning_rate": 1.612244897959184e-05,
      "loss": 1.5103,
      "sft_loss": 0.19822338223457336,
      "step": 82
    },
    {
      "computed_total": 0.2241806834936142,
      "epoch": 0.20933165195460277,
      "grad_norm": 6.878040313720703,
      "kl_loss": 1.1082394024697351e-07,
      "learning_rate": 1.6071428571428572e-05,
      "loss": 1.4422,
      "sft_loss": 0.21309828758239746,
      "step": 83
    },
    {
      "computed_total": 0.19030708074569702,
      "epoch": 0.21185372005044137,
      "grad_norm": 6.8972344398498535,
      "kl_loss": 7.487692244012578e-08,
      "learning_rate": 1.6020408163265308e-05,
      "loss": 1.4802,
      "sft_loss": 0.18281938135623932,
      "step": 84
    },
    {
      "computed_total": 0.18855753540992737,
      "epoch": 0.21437578814627994,
      "grad_norm": 6.072068214416504,
      "kl_loss": 6.224269810672922e-08,
      "learning_rate": 1.596938775510204e-05,
      "loss": 1.3092,
      "sft_loss": 0.18233326077461243,
      "step": 85
    },
    {
      "computed_total": 0.17298777401447296,
      "epoch": 0.21689785624211855,
      "grad_norm": 6.168684005737305,
      "kl_loss": 7.623211928375895e-08,
      "learning_rate": 1.5918367346938776e-05,
      "loss": 1.4069,
      "sft_loss": 0.1653645634651184,
      "step": 86
    },
    {
      "computed_total": 0.19384431838989258,
      "epoch": 0.21941992433795712,
      "grad_norm": 6.2668352127075195,
      "kl_loss": 6.887150760803706e-08,
      "learning_rate": 1.586734693877551e-05,
      "loss": 1.4861,
      "sft_loss": 0.18695716559886932,
      "step": 87
    },
    {
      "computed_total": 0.23912306129932404,
      "epoch": 0.22194199243379573,
      "grad_norm": 5.836297035217285,
      "kl_loss": 7.468336349347737e-08,
      "learning_rate": 1.5816326530612247e-05,
      "loss": 1.4451,
      "sft_loss": 0.23165471851825714,
      "step": 88
    },
    {
      "computed_total": 0.1360023021697998,
      "epoch": 0.2244640605296343,
      "grad_norm": 7.015989303588867,
      "kl_loss": 6.417322140350734e-08,
      "learning_rate": 1.576530612244898e-05,
      "loss": 1.3913,
      "sft_loss": 0.12958498299121857,
      "step": 89
    },
    {
      "computed_total": 0.12024354189634323,
      "epoch": 0.22698612862547288,
      "grad_norm": 6.264023780822754,
      "kl_loss": 7.851058825281143e-08,
      "learning_rate": 1.5714285714285715e-05,
      "loss": 1.3727,
      "sft_loss": 0.11239248514175415,
      "step": 90
    },
    {
      "computed_total": 1.277982473373413,
      "epoch": 0.22698612862547288,
      "eval_loss": 1.3299487829208374,
      "eval_runtime": 63.6992,
      "eval_samples_per_second": 11.068,
      "eval_steps_per_second": 5.542,
      "kl_loss": 4.547840504187661e-08,
      "sft_loss": 1.2734346389770508,
      "step": 90
    },
    {
      "computed_total": 0.12751387059688568,
      "epoch": 0.22950819672131148,
      "grad_norm": 5.627036094665527,
      "kl_loss": 4.57232651740469e-08,
      "learning_rate": 1.566326530612245e-05,
      "loss": 1.4379,
      "sft_loss": 0.12294154614210129,
      "step": 91
    },
    {
      "computed_total": 0.1301025003194809,
      "epoch": 0.23203026481715006,
      "grad_norm": 5.573334217071533,
      "kl_loss": 5.7947175946537754e-08,
      "learning_rate": 1.5612244897959187e-05,
      "loss": 1.5407,
      "sft_loss": 0.124307781457901,
      "step": 92
    },
    {
      "computed_total": 0.23854486644268036,
      "epoch": 0.23455233291298866,
      "grad_norm": 7.010274887084961,
      "kl_loss": 9.491332519928619e-08,
      "learning_rate": 1.556122448979592e-05,
      "loss": 1.4609,
      "sft_loss": 0.2290535271167755,
      "step": 93
    },
    {
      "computed_total": 0.1872793436050415,
      "epoch": 0.23707440100882723,
      "grad_norm": 5.828718185424805,
      "kl_loss": 5.511600420504692e-08,
      "learning_rate": 1.5510204081632655e-05,
      "loss": 1.3066,
      "sft_loss": 0.18176774680614471,
      "step": 94
    },
    {
      "computed_total": 0.16530220210552216,
      "epoch": 0.23959646910466584,
      "grad_norm": 5.710712909698486,
      "kl_loss": 5.7540098907793436e-08,
      "learning_rate": 1.545918367346939e-05,
      "loss": 1.3414,
      "sft_loss": 0.15954819321632385,
      "step": 95
    },
    {
      "computed_total": 0.19449111819267273,
      "epoch": 0.2421185372005044,
      "grad_norm": 6.434009075164795,
      "kl_loss": 6.818410014375331e-08,
      "learning_rate": 1.5408163265306123e-05,
      "loss": 1.5427,
      "sft_loss": 0.1876727044582367,
      "step": 96
    },
    {
      "computed_total": 0.21699348092079163,
      "epoch": 0.244640605296343,
      "grad_norm": 6.784626007080078,
      "kl_loss": 9.980713855384238e-08,
      "learning_rate": 1.535714285714286e-05,
      "loss": 1.4034,
      "sft_loss": 0.20701277256011963,
      "step": 97
    },
    {
      "computed_total": 0.16156645119190216,
      "epoch": 0.2471626733921816,
      "grad_norm": 6.383281707763672,
      "kl_loss": 9.229388098219715e-08,
      "learning_rate": 1.530612244897959e-05,
      "loss": 1.4013,
      "sft_loss": 0.15233705937862396,
      "step": 98
    },
    {
      "computed_total": 0.18982835114002228,
      "epoch": 0.24968474148802017,
      "grad_norm": 6.980619430541992,
      "kl_loss": 7.594896089813119e-08,
      "learning_rate": 1.5255102040816327e-05,
      "loss": 1.3794,
      "sft_loss": 0.18223345279693604,
      "step": 99
    },
    {
      "computed_total": 0.14153845608234406,
      "epoch": 0.25220680958385877,
      "grad_norm": 6.785498142242432,
      "kl_loss": 9.108703125093598e-08,
      "learning_rate": 1.5204081632653063e-05,
      "loss": 1.3444,
      "sft_loss": 0.13242974877357483,
      "step": 100
    },
    {
      "computed_total": 1.2749768495559692,
      "epoch": 0.25220680958385877,
      "eval_loss": 1.3281408548355103,
      "eval_runtime": 64.7152,
      "eval_samples_per_second": 10.894,
      "eval_steps_per_second": 5.455,
      "kl_loss": 5.828524507478505e-08,
      "sft_loss": 1.269148349761963,
      "step": 100
    },
    {
      "computed_total": 0.2398085743188858,
      "epoch": 0.2547288776796974,
      "grad_norm": 5.916060447692871,
      "kl_loss": 5.360384491837067e-08,
      "learning_rate": 1.5153061224489798e-05,
      "loss": 1.3837,
      "sft_loss": 0.23444819450378418,
      "step": 101
    },
    {
      "computed_total": 0.27853187918663025,
      "epoch": 0.2572509457755359,
      "grad_norm": 5.655714511871338,
      "kl_loss": 5.615338949382931e-08,
      "learning_rate": 1.510204081632653e-05,
      "loss": 1.5446,
      "sft_loss": 0.2729165256023407,
      "step": 102
    },
    {
      "computed_total": 0.13499850034713745,
      "epoch": 0.2597730138713745,
      "grad_norm": 5.75067138671875,
      "kl_loss": 1.0047263998558265e-07,
      "learning_rate": 1.5051020408163266e-05,
      "loss": 1.433,
      "sft_loss": 0.12495124340057373,
      "step": 103
    },
    {
      "computed_total": 0.19640223681926727,
      "epoch": 0.26229508196721313,
      "grad_norm": 5.683864593505859,
      "kl_loss": 6.692740583957857e-08,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 1.307,
      "sft_loss": 0.18970949947834015,
      "step": 104
    },
    {
      "computed_total": 0.1632794886827469,
      "epoch": 0.2648171500630517,
      "grad_norm": 8.37120532989502,
      "kl_loss": 5.480029940940767e-08,
      "learning_rate": 1.4948979591836736e-05,
      "loss": 1.4752,
      "sft_loss": 0.15779945254325867,
      "step": 105
    },
    {
      "computed_total": 0.125886470079422,
      "epoch": 0.2673392181588903,
      "grad_norm": 6.49238395690918,
      "kl_loss": 6.983302114349499e-08,
      "learning_rate": 1.4897959183673472e-05,
      "loss": 1.3938,
      "sft_loss": 0.11890317499637604,
      "step": 106
    },
    {
      "computed_total": 0.15055638551712036,
      "epoch": 0.2698612862547289,
      "grad_norm": 8.551501274108887,
      "kl_loss": 6.082338899204842e-08,
      "learning_rate": 1.4846938775510204e-05,
      "loss": 1.5212,
      "sft_loss": 0.14447404444217682,
      "step": 107
    },
    {
      "computed_total": 0.17404527962207794,
      "epoch": 0.2723833543505675,
      "grad_norm": 6.229053974151611,
      "kl_loss": 8.89570088702385e-08,
      "learning_rate": 1.479591836734694e-05,
      "loss": 1.3903,
      "sft_loss": 0.16514958441257477,
      "step": 108
    },
    {
      "computed_total": 0.14084871113300323,
      "epoch": 0.27490542244640603,
      "grad_norm": 6.743444919586182,
      "kl_loss": 9.151927571338092e-08,
      "learning_rate": 1.4744897959183676e-05,
      "loss": 1.4246,
      "sft_loss": 0.13169679045677185,
      "step": 109
    },
    {
      "computed_total": 0.18657232820987701,
      "epoch": 0.27742749054224464,
      "grad_norm": 9.816254615783691,
      "kl_loss": 9.022709690498232e-08,
      "learning_rate": 1.469387755102041e-05,
      "loss": 1.4679,
      "sft_loss": 0.17754961550235748,
      "step": 110
    },
    {
      "computed_total": 1.2606725692749023,
      "epoch": 0.27742749054224464,
      "eval_loss": 1.327353596687317,
      "eval_runtime": 67.5363,
      "eval_samples_per_second": 10.439,
      "eval_steps_per_second": 5.227,
      "kl_loss": 5.828660931683771e-08,
      "sft_loss": 1.2548439502716064,
      "step": 110
    },
    {
      "computed_total": 0.22784219682216644,
      "epoch": 0.27994955863808324,
      "grad_norm": 6.819587230682373,
      "kl_loss": 5.9556931830684334e-08,
      "learning_rate": 1.4642857142857144e-05,
      "loss": 1.428,
      "sft_loss": 0.2218865007162094,
      "step": 111
    },
    {
      "computed_total": 0.18110135197639465,
      "epoch": 0.28247162673392184,
      "grad_norm": 6.500746250152588,
      "kl_loss": 6.125114992983072e-08,
      "learning_rate": 1.4591836734693878e-05,
      "loss": 1.4575,
      "sft_loss": 0.17497624456882477,
      "step": 112
    },
    {
      "computed_total": 0.19457776844501495,
      "epoch": 0.2849936948297604,
      "grad_norm": 6.06559419631958,
      "kl_loss": 6.390897766550552e-08,
      "learning_rate": 1.4540816326530614e-05,
      "loss": 1.3074,
      "sft_loss": 0.1881868690252304,
      "step": 113
    },
    {
      "computed_total": 0.16224755346775055,
      "epoch": 0.287515762925599,
      "grad_norm": 6.051407337188721,
      "kl_loss": 6.450272138636137e-08,
      "learning_rate": 1.448979591836735e-05,
      "loss": 1.3463,
      "sft_loss": 0.15579728782176971,
      "step": 114
    },
    {
      "computed_total": 0.15151885151863098,
      "epoch": 0.2900378310214376,
      "grad_norm": 6.495911121368408,
      "kl_loss": 5.75203635833077e-08,
      "learning_rate": 1.4438775510204083e-05,
      "loss": 1.1866,
      "sft_loss": 0.14576680958271027,
      "step": 115
    },
    {
      "computed_total": 0.21337838470935822,
      "epoch": 0.29255989911727615,
      "grad_norm": 5.487979412078857,
      "kl_loss": 5.020171300884613e-08,
      "learning_rate": 1.4387755102040817e-05,
      "loss": 1.3196,
      "sft_loss": 0.20835821330547333,
      "step": 116
    },
    {
      "computed_total": 0.23495040833950043,
      "epoch": 0.29508196721311475,
      "grad_norm": 6.188425540924072,
      "kl_loss": 7.129592205501467e-08,
      "learning_rate": 1.4336734693877551e-05,
      "loss": 1.4065,
      "sft_loss": 0.22782081365585327,
      "step": 117
    },
    {
      "computed_total": 0.173566997051239,
      "epoch": 0.29760403530895335,
      "grad_norm": 6.5124664306640625,
      "kl_loss": 8.479923963022884e-08,
      "learning_rate": 1.4285714285714287e-05,
      "loss": 1.4831,
      "sft_loss": 0.16508707404136658,
      "step": 118
    },
    {
      "computed_total": 0.1915544718503952,
      "epoch": 0.30012610340479196,
      "grad_norm": 6.951809883117676,
      "kl_loss": 8.485299929361645e-08,
      "learning_rate": 1.4234693877551023e-05,
      "loss": 1.4373,
      "sft_loss": 0.1830691695213318,
      "step": 119
    },
    {
      "computed_total": 0.13525913655757904,
      "epoch": 0.3026481715006305,
      "grad_norm": 5.700254440307617,
      "kl_loss": 7.165923676666353e-08,
      "learning_rate": 1.4183673469387755e-05,
      "loss": 1.248,
      "sft_loss": 0.12809321284294128,
      "step": 120
    },
    {
      "computed_total": 1.2876116037368774,
      "epoch": 0.3026481715006305,
      "eval_loss": 1.3242225646972656,
      "eval_runtime": 65.3078,
      "eval_samples_per_second": 10.795,
      "eval_steps_per_second": 5.405,
      "kl_loss": 5.639139644131319e-08,
      "sft_loss": 1.2819724082946777,
      "step": 120
    },
    {
      "computed_total": 0.24368716776371002,
      "epoch": 0.3051702395964691,
      "grad_norm": 8.6900053024292,
      "kl_loss": 4.4171105884061035e-08,
      "learning_rate": 1.4132653061224491e-05,
      "loss": 1.3817,
      "sft_loss": 0.23927006125450134,
      "step": 121
    },
    {
      "computed_total": 0.21491077542304993,
      "epoch": 0.3076923076923077,
      "grad_norm": 5.918478488922119,
      "kl_loss": 4.5203840670637874e-08,
      "learning_rate": 1.4081632653061225e-05,
      "loss": 1.417,
      "sft_loss": 0.2103903889656067,
      "step": 122
    },
    {
      "computed_total": 0.16387128829956055,
      "epoch": 0.31021437578814626,
      "grad_norm": 6.375618934631348,
      "kl_loss": 7.955883063459623e-08,
      "learning_rate": 1.403061224489796e-05,
      "loss": 1.3237,
      "sft_loss": 0.15591540932655334,
      "step": 123
    },
    {
      "computed_total": 0.13963785767555237,
      "epoch": 0.31273644388398486,
      "grad_norm": 6.16308069229126,
      "kl_loss": 6.622069292916422e-08,
      "learning_rate": 1.3979591836734696e-05,
      "loss": 1.3588,
      "sft_loss": 0.13301578164100647,
      "step": 124
    },
    {
      "computed_total": 0.13709679245948792,
      "epoch": 0.31525851197982346,
      "grad_norm": 6.024322032928467,
      "kl_loss": 6.11066113265224e-08,
      "learning_rate": 1.3928571428571429e-05,
      "loss": 1.3791,
      "sft_loss": 0.13098612427711487,
      "step": 125
    },
    {
      "computed_total": 0.2479337900876999,
      "epoch": 0.31778058007566207,
      "grad_norm": 5.9967942237854,
      "kl_loss": 5.687433457524094e-08,
      "learning_rate": 1.3877551020408165e-05,
      "loss": 1.4524,
      "sft_loss": 0.2422463595867157,
      "step": 126
    },
    {
      "computed_total": 0.17359566688537598,
      "epoch": 0.3203026481715006,
      "grad_norm": 6.545812606811523,
      "kl_loss": 5.99686273972111e-08,
      "learning_rate": 1.38265306122449e-05,
      "loss": 1.4399,
      "sft_loss": 0.16759879887104034,
      "step": 127
    },
    {
      "computed_total": 0.20949114859104156,
      "epoch": 0.3228247162673392,
      "grad_norm": 5.510209560394287,
      "kl_loss": 5.430339200529488e-08,
      "learning_rate": 1.3775510204081634e-05,
      "loss": 1.248,
      "sft_loss": 0.20406080782413483,
      "step": 128
    },
    {
      "computed_total": 0.1810821294784546,
      "epoch": 0.3253467843631778,
      "grad_norm": 5.705124378204346,
      "kl_loss": 4.393492147869438e-08,
      "learning_rate": 1.3724489795918368e-05,
      "loss": 1.3882,
      "sft_loss": 0.17668864130973816,
      "step": 129
    },
    {
      "computed_total": 0.18383508920669556,
      "epoch": 0.32786885245901637,
      "grad_norm": 6.114168643951416,
      "kl_loss": 5.20567979833686e-08,
      "learning_rate": 1.3673469387755102e-05,
      "loss": 1.3768,
      "sft_loss": 0.17862941324710846,
      "step": 130
    },
    {
      "computed_total": 1.2415862083435059,
      "epoch": 0.32786885245901637,
      "eval_loss": 1.3175008296966553,
      "eval_runtime": 66.8218,
      "eval_samples_per_second": 10.55,
      "eval_steps_per_second": 5.283,
      "kl_loss": 4.615326076873316e-08,
      "sft_loss": 1.2369709014892578,
      "step": 130
    },
    {
      "computed_total": 0.15210777521133423,
      "epoch": 0.33039092055485497,
      "grad_norm": 5.750833988189697,
      "kl_loss": 3.750525578993802e-08,
      "learning_rate": 1.3622448979591838e-05,
      "loss": 1.3064,
      "sft_loss": 0.14835724234580994,
      "step": 131
    },
    {
      "computed_total": 0.2978253960609436,
      "epoch": 0.3329129886506936,
      "grad_norm": 5.927938461303711,
      "kl_loss": 5.971872241161691e-08,
      "learning_rate": 1.3571428571428574e-05,
      "loss": 1.5619,
      "sft_loss": 0.29185351729393005,
      "step": 132
    },
    {
      "computed_total": 0.19227644801139832,
      "epoch": 0.3354350567465322,
      "grad_norm": 6.050528049468994,
      "kl_loss": 6.213254266640433e-08,
      "learning_rate": 1.3520408163265306e-05,
      "loss": 1.3891,
      "sft_loss": 0.18606320023536682,
      "step": 133
    },
    {
      "computed_total": 0.2411646544933319,
      "epoch": 0.3379571248423707,
      "grad_norm": 6.082936763763428,
      "kl_loss": 4.6511246409863816e-08,
      "learning_rate": 1.3469387755102042e-05,
      "loss": 1.3559,
      "sft_loss": 0.23651352524757385,
      "step": 134
    },
    {
      "computed_total": 0.2286994606256485,
      "epoch": 0.34047919293820933,
      "grad_norm": 5.977089881896973,
      "kl_loss": 4.714263468486024e-08,
      "learning_rate": 1.3418367346938776e-05,
      "loss": 1.3913,
      "sft_loss": 0.2239851951599121,
      "step": 135
    },
    {
      "computed_total": 0.1456630676984787,
      "epoch": 0.34300126103404793,
      "grad_norm": 5.774937629699707,
      "kl_loss": 4.883188609028366e-08,
      "learning_rate": 1.3367346938775512e-05,
      "loss": 1.4413,
      "sft_loss": 0.14077988266944885,
      "step": 136
    },
    {
      "computed_total": 0.1698574274778366,
      "epoch": 0.3455233291298865,
      "grad_norm": 5.268606662750244,
      "kl_loss": 3.678103510651454e-08,
      "learning_rate": 1.3316326530612247e-05,
      "loss": 1.3221,
      "sft_loss": 0.1661793291568756,
      "step": 137
    },
    {
      "computed_total": 0.1976916640996933,
      "epoch": 0.3480453972257251,
      "grad_norm": 5.664859294891357,
      "kl_loss": 6.647618988608883e-08,
      "learning_rate": 1.326530612244898e-05,
      "loss": 1.363,
      "sft_loss": 0.19104404747486115,
      "step": 138
    },
    {
      "computed_total": 0.1511334627866745,
      "epoch": 0.3505674653215637,
      "grad_norm": 7.167004585266113,
      "kl_loss": 9.495975206164076e-08,
      "learning_rate": 1.3214285714285716e-05,
      "loss": 1.4917,
      "sft_loss": 0.14163748919963837,
      "step": 139
    },
    {
      "computed_total": 0.26482662558555603,
      "epoch": 0.3530895334174023,
      "grad_norm": 6.801730155944824,
      "kl_loss": 7.647604860494539e-08,
      "learning_rate": 1.316326530612245e-05,
      "loss": 1.4963,
      "sft_loss": 0.25717902183532715,
      "step": 140
    },
    {
      "computed_total": 1.2509357929229736,
      "epoch": 0.3530895334174023,
      "eval_loss": 1.312376856803894,
      "eval_runtime": 66.3405,
      "eval_samples_per_second": 10.627,
      "eval_steps_per_second": 5.321,
      "kl_loss": 4.728717328816856e-08,
      "sft_loss": 1.2462071180343628,
      "step": 140
    },
    {
      "computed_total": 0.15087392926216125,
      "epoch": 0.35561160151324084,
      "grad_norm": 5.64943790435791,
      "kl_loss": 4.7736318009583556e-08,
      "learning_rate": 1.3112244897959185e-05,
      "loss": 1.2534,
      "sft_loss": 0.14610029757022858,
      "step": 141
    },
    {
      "computed_total": 0.1842363029718399,
      "epoch": 0.35813366960907944,
      "grad_norm": 5.523940086364746,
      "kl_loss": 6.048590250884445e-08,
      "learning_rate": 1.3061224489795918e-05,
      "loss": 1.3035,
      "sft_loss": 0.17818771302700043,
      "step": 142
    },
    {
      "computed_total": 0.17870008945465088,
      "epoch": 0.36065573770491804,
      "grad_norm": 5.714141368865967,
      "kl_loss": 6.908807392846938e-08,
      "learning_rate": 1.3010204081632653e-05,
      "loss": 1.4674,
      "sft_loss": 0.17179128527641296,
      "step": 143
    },
    {
      "computed_total": 0.12025298178195953,
      "epoch": 0.36317780580075665,
      "grad_norm": 5.176926612854004,
      "kl_loss": 5.138770475809906e-08,
      "learning_rate": 1.2959183673469389e-05,
      "loss": 1.2982,
      "sft_loss": 0.11511421203613281,
      "step": 144
    },
    {
      "computed_total": 0.11642109602689743,
      "epoch": 0.3656998738965952,
      "grad_norm": 5.232058525085449,
      "kl_loss": 5.1927578681443265e-08,
      "learning_rate": 1.2908163265306123e-05,
      "loss": 1.3548,
      "sft_loss": 0.1112283393740654,
      "step": 145
    },
    {
      "computed_total": 0.17399540543556213,
      "epoch": 0.3682219419924338,
      "grad_norm": 5.944302082061768,
      "kl_loss": 7.11170713429965e-08,
      "learning_rate": 1.2857142857142859e-05,
      "loss": 1.4696,
      "sft_loss": 0.1668836921453476,
      "step": 146
    },
    {
      "computed_total": 0.24251551926136017,
      "epoch": 0.3707440100882724,
      "grad_norm": 5.792961597442627,
      "kl_loss": 5.667594393798936e-08,
      "learning_rate": 1.2806122448979591e-05,
      "loss": 1.3838,
      "sft_loss": 0.236847922205925,
      "step": 147
    },
    {
      "computed_total": 0.12372773885726929,
      "epoch": 0.37326607818411095,
      "grad_norm": 6.338442325592041,
      "kl_loss": 5.697808092008927e-08,
      "learning_rate": 1.2755102040816327e-05,
      "loss": 1.4152,
      "sft_loss": 0.11802992969751358,
      "step": 148
    },
    {
      "computed_total": 0.17351199686527252,
      "epoch": 0.37578814627994955,
      "grad_norm": 6.114325523376465,
      "kl_loss": 4.277083220927125e-08,
      "learning_rate": 1.2704081632653063e-05,
      "loss": 1.4443,
      "sft_loss": 0.16923491656780243,
      "step": 149
    },
    {
      "computed_total": 0.1712173968553543,
      "epoch": 0.37831021437578816,
      "grad_norm": 5.3177995681762695,
      "kl_loss": 4.446590651241422e-08,
      "learning_rate": 1.2653061224489798e-05,
      "loss": 1.3345,
      "sft_loss": 0.166770800948143,
      "step": 150
    },
    {
      "computed_total": 1.2424803972244263,
      "epoch": 0.37831021437578816,
      "eval_loss": 1.3096885681152344,
      "eval_runtime": 66.587,
      "eval_samples_per_second": 10.588,
      "eval_steps_per_second": 5.301,
      "kl_loss": 3.8276976255247064e-08,
      "sft_loss": 1.2386527061462402,
      "step": 150
    },
    {
      "computed_total": 0.16291235387325287,
      "epoch": 0.38083228247162676,
      "grad_norm": 5.252249240875244,
      "kl_loss": 3.6955761117951624e-08,
      "learning_rate": 1.260204081632653e-05,
      "loss": 1.2757,
      "sft_loss": 0.1592167764902115,
      "step": 151
    },
    {
      "computed_total": 0.16880770027637482,
      "epoch": 0.3833543505674653,
      "grad_norm": 5.493607044219971,
      "kl_loss": 3.392970882032387e-08,
      "learning_rate": 1.2551020408163267e-05,
      "loss": 1.4651,
      "sft_loss": 0.1654147356748581,
      "step": 152
    },
    {
      "computed_total": 0.21267661452293396,
      "epoch": 0.3858764186633039,
      "grad_norm": 5.093979835510254,
      "kl_loss": 5.242512202130456e-08,
      "learning_rate": 1.25e-05,
      "loss": 1.2663,
      "sft_loss": 0.20743410289287567,
      "step": 153
    },
    {
      "computed_total": 0.18249347805976868,
      "epoch": 0.3883984867591425,
      "grad_norm": 5.7369184494018555,
      "kl_loss": 5.691558158105181e-08,
      "learning_rate": 1.2448979591836736e-05,
      "loss": 1.4205,
      "sft_loss": 0.1768019199371338,
      "step": 154
    },
    {
      "computed_total": 0.1492944061756134,
      "epoch": 0.39092055485498106,
      "grad_norm": 5.527451038360596,
      "kl_loss": 4.2787782206232805e-08,
      "learning_rate": 1.2397959183673472e-05,
      "loss": 1.4068,
      "sft_loss": 0.1450156271457672,
      "step": 155
    },
    {
      "computed_total": 0.11073505133390427,
      "epoch": 0.39344262295081966,
      "grad_norm": 5.060709476470947,
      "kl_loss": 3.849062224503541e-08,
      "learning_rate": 1.2346938775510204e-05,
      "loss": 1.2852,
      "sft_loss": 0.10688599199056625,
      "step": 156
    },
    {
      "computed_total": 0.28565481305122375,
      "epoch": 0.39596469104665827,
      "grad_norm": 5.988109588623047,
      "kl_loss": 4.468333614227049e-08,
      "learning_rate": 1.229591836734694e-05,
      "loss": 1.4913,
      "sft_loss": 0.2811864912509918,
      "step": 157
    },
    {
      "computed_total": 0.13953863084316254,
      "epoch": 0.39848675914249687,
      "grad_norm": 5.936272621154785,
      "kl_loss": 4.650754092949683e-08,
      "learning_rate": 1.2244897959183674e-05,
      "loss": 1.3665,
      "sft_loss": 0.13488787412643433,
      "step": 158
    },
    {
      "computed_total": 0.11373602598905563,
      "epoch": 0.4010088272383354,
      "grad_norm": 5.1962971687316895,
      "kl_loss": 5.032283567629747e-08,
      "learning_rate": 1.219387755102041e-05,
      "loss": 1.2644,
      "sft_loss": 0.10870373994112015,
      "step": 159
    },
    {
      "computed_total": 0.14209257066249847,
      "epoch": 0.403530895334174,
      "grad_norm": 6.162265300750732,
      "kl_loss": 6.199127255968051e-08,
      "learning_rate": 1.2142857142857142e-05,
      "loss": 1.3934,
      "sft_loss": 0.13589344918727875,
      "step": 160
    },
    {
      "computed_total": 1.2494064569473267,
      "epoch": 0.403530895334174,
      "eval_loss": 1.3051645755767822,
      "eval_runtime": 64.4178,
      "eval_samples_per_second": 10.944,
      "eval_steps_per_second": 5.48,
      "kl_loss": 3.006507398595204e-08,
      "sft_loss": 1.246399998664856,
      "step": 160
    },
    {
      "computed_total": 0.1575666218996048,
      "epoch": 0.4060529634300126,
      "grad_norm": 5.189947605133057,
      "kl_loss": 3.18427701984092e-08,
      "learning_rate": 1.2091836734693878e-05,
      "loss": 1.464,
      "sft_loss": 0.1543823480606079,
      "step": 161
    },
    {
      "computed_total": 0.18502387404441833,
      "epoch": 0.4085750315258512,
      "grad_norm": 5.290241241455078,
      "kl_loss": 4.9009557301360474e-08,
      "learning_rate": 1.2040816326530614e-05,
      "loss": 1.3788,
      "sft_loss": 0.18012291193008423,
      "step": 162
    },
    {
      "computed_total": 0.1514197289943695,
      "epoch": 0.4110970996216898,
      "grad_norm": 5.770331382751465,
      "kl_loss": 6.018641585114892e-08,
      "learning_rate": 1.1989795918367348e-05,
      "loss": 1.4546,
      "sft_loss": 0.14540109038352966,
      "step": 163
    },
    {
      "computed_total": 0.14091259241104126,
      "epoch": 0.4136191677175284,
      "grad_norm": 5.589346885681152,
      "kl_loss": 5.507525457915108e-08,
      "learning_rate": 1.1938775510204084e-05,
      "loss": 1.3665,
      "sft_loss": 0.1354050636291504,
      "step": 164
    },
    {
      "computed_total": 0.16737474501132965,
      "epoch": 0.416141235813367,
      "grad_norm": 5.2255659103393555,
      "kl_loss": 4.1372025805230805e-08,
      "learning_rate": 1.1887755102040816e-05,
      "loss": 1.3646,
      "sft_loss": 0.1632375419139862,
      "step": 165
    },
    {
      "computed_total": 0.13658396899700165,
      "epoch": 0.41866330390920553,
      "grad_norm": 5.557989597320557,
      "kl_loss": 4.627627347986163e-08,
      "learning_rate": 1.1836734693877552e-05,
      "loss": 1.3747,
      "sft_loss": 0.1319563388824463,
      "step": 166
    },
    {
      "computed_total": 0.19387367367744446,
      "epoch": 0.42118537200504413,
      "grad_norm": 5.382153511047363,
      "kl_loss": 3.4211062427402794e-08,
      "learning_rate": 1.1785714285714287e-05,
      "loss": 1.3553,
      "sft_loss": 0.19045256078243256,
      "step": 167
    },
    {
      "computed_total": 0.17456893622875214,
      "epoch": 0.42370744010088274,
      "grad_norm": 4.9972243309021,
      "kl_loss": 4.348507687268466e-08,
      "learning_rate": 1.1734693877551021e-05,
      "loss": 1.323,
      "sft_loss": 0.17022043466567993,
      "step": 168
    },
    {
      "computed_total": 0.17107342183589935,
      "epoch": 0.4262295081967213,
      "grad_norm": 5.4530205726623535,
      "kl_loss": 5.122308976979184e-08,
      "learning_rate": 1.1683673469387755e-05,
      "loss": 1.4871,
      "sft_loss": 0.16595111787319183,
      "step": 169
    },
    {
      "computed_total": 0.190860316157341,
      "epoch": 0.4287515762925599,
      "grad_norm": 5.652188777923584,
      "kl_loss": 5.993562268713504e-08,
      "learning_rate": 1.1632653061224491e-05,
      "loss": 1.3863,
      "sft_loss": 0.1848667562007904,
      "step": 170
    },
    {
      "computed_total": 1.2313315868377686,
      "epoch": 0.4287515762925599,
      "eval_loss": 1.301425814628601,
      "eval_runtime": 66.926,
      "eval_samples_per_second": 10.534,
      "eval_steps_per_second": 5.274,
      "kl_loss": 3.545708437968642e-08,
      "sft_loss": 1.2277858257293701,
      "step": 170
    },
    {
      "computed_total": 0.15655340254306793,
      "epoch": 0.4312736443883985,
      "grad_norm": 5.242849826812744,
      "kl_loss": 4.240297357682721e-08,
      "learning_rate": 1.1581632653061225e-05,
      "loss": 1.2935,
      "sft_loss": 0.15231309831142426,
      "step": 171
    },
    {
      "computed_total": 0.12881718575954437,
      "epoch": 0.4337957124842371,
      "grad_norm": 5.205791473388672,
      "kl_loss": 3.2137183580971396e-08,
      "learning_rate": 1.1530612244897961e-05,
      "loss": 1.3333,
      "sft_loss": 0.12560346722602844,
      "step": 172
    },
    {
      "computed_total": 0.15559355914592743,
      "epoch": 0.43631778058007564,
      "grad_norm": 4.979446887969971,
      "kl_loss": 4.915695583918023e-08,
      "learning_rate": 1.1479591836734697e-05,
      "loss": 1.51,
      "sft_loss": 0.1506778597831726,
      "step": 173
    },
    {
      "computed_total": 0.19263696670532227,
      "epoch": 0.43883984867591425,
      "grad_norm": 5.016871929168701,
      "kl_loss": 3.290235639497041e-08,
      "learning_rate": 1.1428571428571429e-05,
      "loss": 1.3166,
      "sft_loss": 0.18934673070907593,
      "step": 174
    },
    {
      "computed_total": 0.18717873096466064,
      "epoch": 0.44136191677175285,
      "grad_norm": 5.055232048034668,
      "kl_loss": 3.505217094357249e-08,
      "learning_rate": 1.1377551020408165e-05,
      "loss": 1.3125,
      "sft_loss": 0.18367351591587067,
      "step": 175
    },
    {
      "computed_total": 0.1265634000301361,
      "epoch": 0.44388398486759145,
      "grad_norm": 5.522885799407959,
      "kl_loss": 3.791987879253611e-08,
      "learning_rate": 1.1326530612244899e-05,
      "loss": 1.447,
      "sft_loss": 0.12277141958475113,
      "step": 176
    },
    {
      "computed_total": 0.262478232383728,
      "epoch": 0.44640605296343,
      "grad_norm": 4.968825817108154,
      "kl_loss": 2.4255626840385958e-08,
      "learning_rate": 1.1275510204081635e-05,
      "loss": 1.3434,
      "sft_loss": 0.2600526809692383,
      "step": 177
    },
    {
      "computed_total": 0.11854858696460724,
      "epoch": 0.4489281210592686,
      "grad_norm": 4.760946273803711,
      "kl_loss": 2.9037074966709042e-08,
      "learning_rate": 1.1224489795918367e-05,
      "loss": 1.2781,
      "sft_loss": 0.11564487963914871,
      "step": 178
    },
    {
      "computed_total": 0.1870357096195221,
      "epoch": 0.4514501891551072,
      "grad_norm": 4.921207904815674,
      "kl_loss": 2.6307930056646e-08,
      "learning_rate": 1.1173469387755103e-05,
      "loss": 1.325,
      "sft_loss": 0.1844049096107483,
      "step": 179
    },
    {
      "computed_total": 0.1689537614583969,
      "epoch": 0.45397225725094575,
      "grad_norm": 5.677329063415527,
      "kl_loss": 4.060397529315196e-08,
      "learning_rate": 1.1122448979591838e-05,
      "loss": 1.5654,
      "sft_loss": 0.16489335894584656,
      "step": 180
    },
    {
      "computed_total": 1.2363641262054443,
      "epoch": 0.45397225725094575,
      "eval_loss": 1.2977334260940552,
      "eval_runtime": 64.1626,
      "eval_samples_per_second": 10.988,
      "eval_steps_per_second": 5.502,
      "kl_loss": 2.7487111253776675e-08,
      "sft_loss": 1.2336153984069824,
      "step": 180
    },
    {
      "computed_total": 0.15740957856178284,
      "epoch": 0.45649432534678436,
      "grad_norm": 5.158702850341797,
      "kl_loss": 2.6572712030770163e-08,
      "learning_rate": 1.1071428571428572e-05,
      "loss": 1.4782,
      "sft_loss": 0.15475231409072876,
      "step": 181
    },
    {
      "computed_total": 0.14208067953586578,
      "epoch": 0.45901639344262296,
      "grad_norm": 4.95545768737793,
      "kl_loss": 2.9629616093984623e-08,
      "learning_rate": 1.1020408163265306e-05,
      "loss": 1.3376,
      "sft_loss": 0.13911771774291992,
      "step": 182
    },
    {
      "computed_total": 0.1579395830631256,
      "epoch": 0.46153846153846156,
      "grad_norm": 4.77717399597168,
      "kl_loss": 4.216128601797209e-08,
      "learning_rate": 1.096938775510204e-05,
      "loss": 1.3987,
      "sft_loss": 0.15372344851493835,
      "step": 183
    },
    {
      "computed_total": 0.1506652683019638,
      "epoch": 0.4640605296343001,
      "grad_norm": 5.211386680603027,
      "kl_loss": 5.038693373649039e-08,
      "learning_rate": 1.0918367346938776e-05,
      "loss": 1.4013,
      "sft_loss": 0.14562657475471497,
      "step": 184
    },
    {
      "computed_total": 0.16317421197891235,
      "epoch": 0.4665825977301387,
      "grad_norm": 4.944554328918457,
      "kl_loss": 3.7608401726174634e-08,
      "learning_rate": 1.0867346938775512e-05,
      "loss": 1.1998,
      "sft_loss": 0.15941336750984192,
      "step": 185
    },
    {
      "computed_total": 0.18855802714824677,
      "epoch": 0.4691046658259773,
      "grad_norm": 4.752446174621582,
      "kl_loss": 3.388852221064553e-08,
      "learning_rate": 1.0816326530612246e-05,
      "loss": 1.3246,
      "sft_loss": 0.18516917526721954,
      "step": 186
    },
    {
      "computed_total": 0.13987620174884796,
      "epoch": 0.47162673392181587,
      "grad_norm": 5.217368125915527,
      "kl_loss": 3.243337687308667e-08,
      "learning_rate": 1.076530612244898e-05,
      "loss": 1.3806,
      "sft_loss": 0.13663285970687866,
      "step": 187
    },
    {
      "computed_total": 0.14583595097064972,
      "epoch": 0.47414880201765447,
      "grad_norm": 4.883551120758057,
      "kl_loss": 3.8305525862369905e-08,
      "learning_rate": 1.0714285714285714e-05,
      "loss": 1.3337,
      "sft_loss": 0.14200539886951447,
      "step": 188
    },
    {
      "computed_total": 0.12065444886684418,
      "epoch": 0.4766708701134931,
      "grad_norm": 5.246138095855713,
      "kl_loss": 3.744847987263711e-08,
      "learning_rate": 1.066326530612245e-05,
      "loss": 1.1985,
      "sft_loss": 0.11690960079431534,
      "step": 189
    },
    {
      "computed_total": 0.16517040133476257,
      "epoch": 0.4791929382093317,
      "grad_norm": 5.633914470672607,
      "kl_loss": 3.465543585434716e-08,
      "learning_rate": 1.0612244897959186e-05,
      "loss": 1.2648,
      "sft_loss": 0.16170485317707062,
      "step": 190
    },
    {
      "computed_total": 1.2361537218093872,
      "epoch": 0.4791929382093317,
      "eval_loss": 1.2966482639312744,
      "eval_runtime": 65.136,
      "eval_samples_per_second": 10.824,
      "eval_steps_per_second": 5.419,
      "kl_loss": 2.7988660278310817e-08,
      "sft_loss": 1.2333548069000244,
      "step": 190
    },
    {
      "computed_total": 0.25373631715774536,
      "epoch": 0.4817150063051702,
      "grad_norm": 5.030147075653076,
      "kl_loss": 3.2297595708996596e-08,
      "learning_rate": 1.0561224489795918e-05,
      "loss": 1.4551,
      "sft_loss": 0.25050655007362366,
      "step": 191
    },
    {
      "computed_total": 0.1556681990623474,
      "epoch": 0.4842370744010088,
      "grad_norm": 4.94256591796875,
      "kl_loss": 3.0316428478727175e-08,
      "learning_rate": 1.0510204081632654e-05,
      "loss": 1.4168,
      "sft_loss": 0.1526365578174591,
      "step": 192
    },
    {
      "computed_total": 0.11549299210309982,
      "epoch": 0.48675914249684743,
      "grad_norm": 4.747858047485352,
      "kl_loss": 2.870864079795865e-08,
      "learning_rate": 1.045918367346939e-05,
      "loss": 1.199,
      "sft_loss": 0.11262212693691254,
      "step": 193
    },
    {
      "computed_total": 0.17795605957508087,
      "epoch": 0.489281210592686,
      "grad_norm": 5.250139236450195,
      "kl_loss": 2.986351432809897e-08,
      "learning_rate": 1.0408163265306123e-05,
      "loss": 1.394,
      "sft_loss": 0.17496970295906067,
      "step": 194
    },
    {
      "computed_total": 0.19819606840610504,
      "epoch": 0.4918032786885246,
      "grad_norm": 5.39154052734375,
      "kl_loss": 2.9720043315251132e-08,
      "learning_rate": 1.0357142857142859e-05,
      "loss": 1.5046,
      "sft_loss": 0.19522406160831451,
      "step": 195
    },
    {
      "computed_total": 0.21292340755462646,
      "epoch": 0.4943253467843632,
      "grad_norm": 4.84930419921875,
      "kl_loss": 2.377551133747602e-08,
      "learning_rate": 1.0306122448979591e-05,
      "loss": 1.5064,
      "sft_loss": 0.2105458527803421,
      "step": 196
    },
    {
      "computed_total": 0.15452894568443298,
      "epoch": 0.4968474148802018,
      "grad_norm": 4.649036407470703,
      "kl_loss": 2.7660080448299595e-08,
      "learning_rate": 1.0255102040816327e-05,
      "loss": 1.3328,
      "sft_loss": 0.1517629325389862,
      "step": 197
    },
    {
      "computed_total": 0.1479904055595398,
      "epoch": 0.49936948297604034,
      "grad_norm": 4.551985263824463,
      "kl_loss": 2.352908445857338e-08,
      "learning_rate": 1.0204081632653063e-05,
      "loss": 1.2328,
      "sft_loss": 0.14563749730587006,
      "step": 198
    },
    {
      "computed_total": 0.12488767504692078,
      "epoch": 0.501891551071879,
      "grad_norm": 5.3588433265686035,
      "kl_loss": 4.177105594749264e-08,
      "learning_rate": 1.0153061224489797e-05,
      "loss": 1.3333,
      "sft_loss": 0.12071056663990021,
      "step": 199
    },
    {
      "computed_total": 0.21304959058761597,
      "epoch": 0.5044136191677175,
      "grad_norm": 5.3644914627075195,
      "kl_loss": 2.755204775439779e-08,
      "learning_rate": 1.0102040816326531e-05,
      "loss": 1.3657,
      "sft_loss": 0.21029438078403473,
      "step": 200
    },
    {
      "computed_total": 1.2131683826446533,
      "epoch": 0.5044136191677175,
      "eval_loss": 1.2966324090957642,
      "eval_runtime": 66.0512,
      "eval_samples_per_second": 10.674,
      "eval_steps_per_second": 5.344,
      "kl_loss": 1.9633105807770335e-08,
      "sft_loss": 1.2112051248550415,
      "step": 200
    },
    {
      "computed_total": 0.17476873099803925,
      "epoch": 0.5069356872635561,
      "grad_norm": 4.610480785369873,
      "kl_loss": 2.109863217469865e-08,
      "learning_rate": 1.0051020408163265e-05,
      "loss": 1.2396,
      "sft_loss": 0.17265886068344116,
      "step": 201
    },
    {
      "computed_total": 0.16051669418811798,
      "epoch": 0.5094577553593947,
      "grad_norm": 4.916041851043701,
      "kl_loss": 2.2446158709499286e-08,
      "learning_rate": 1e-05,
      "loss": 1.4855,
      "sft_loss": 0.15827207267284393,
      "step": 202
    },
    {
      "computed_total": 0.10832415521144867,
      "epoch": 0.5119798234552333,
      "grad_norm": 4.801663398742676,
      "kl_loss": 2.7506091626605667e-08,
      "learning_rate": 9.948979591836737e-06,
      "loss": 1.3307,
      "sft_loss": 0.10557354241609573,
      "step": 203
    },
    {
      "computed_total": 0.2282698005437851,
      "epoch": 0.5145018915510718,
      "grad_norm": 5.357565402984619,
      "kl_loss": 4.084584759311838e-08,
      "learning_rate": 9.89795918367347e-06,
      "loss": 1.4057,
      "sft_loss": 0.22418521344661713,
      "step": 204
    },
    {
      "computed_total": 0.1381111890077591,
      "epoch": 0.5170239596469105,
      "grad_norm": 5.372463703155518,
      "kl_loss": 3.638923473658906e-08,
      "learning_rate": 9.846938775510205e-06,
      "loss": 1.4017,
      "sft_loss": 0.13447226583957672,
      "step": 205
    },
    {
      "computed_total": 0.23932000994682312,
      "epoch": 0.519546027742749,
      "grad_norm": 4.798728942871094,
      "kl_loss": 3.008491589184814e-08,
      "learning_rate": 9.795918367346939e-06,
      "loss": 1.4547,
      "sft_loss": 0.23631152510643005,
      "step": 206
    },
    {
      "computed_total": 0.24363160133361816,
      "epoch": 0.5220680958385876,
      "grad_norm": 4.651880741119385,
      "kl_loss": 3.3694899315150906e-08,
      "learning_rate": 9.744897959183674e-06,
      "loss": 1.3315,
      "sft_loss": 0.24026210606098175,
      "step": 207
    },
    {
      "computed_total": 0.12392779439687729,
      "epoch": 0.5245901639344263,
      "grad_norm": 4.969662666320801,
      "kl_loss": 3.430271178217481e-08,
      "learning_rate": 9.693877551020408e-06,
      "loss": 1.263,
      "sft_loss": 0.12049752473831177,
      "step": 208
    },
    {
      "computed_total": 0.18426865339279175,
      "epoch": 0.5271122320302648,
      "grad_norm": 4.773207664489746,
      "kl_loss": 2.9775220511396583e-08,
      "learning_rate": 9.642857142857144e-06,
      "loss": 1.2383,
      "sft_loss": 0.1812911331653595,
      "step": 209
    },
    {
      "computed_total": 0.16482222080230713,
      "epoch": 0.5296343001261034,
      "grad_norm": 5.222415447235107,
      "kl_loss": 3.173001417167143e-08,
      "learning_rate": 9.591836734693878e-06,
      "loss": 1.4537,
      "sft_loss": 0.1616492122411728,
      "step": 210
    },
    {
      "computed_total": 1.2207623720169067,
      "epoch": 0.5296343001261034,
      "eval_loss": 1.2934982776641846,
      "eval_runtime": 65.6715,
      "eval_samples_per_second": 10.735,
      "eval_steps_per_second": 5.375,
      "kl_loss": 2.6054731705471568e-08,
      "sft_loss": 1.2181569337844849,
      "step": 210
    }
  ],
  "logging_steps": 1,
  "max_steps": 397,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.87384409875456e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}