{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9931153184165233,
  "eval_steps": 20,
  "global_step": 386,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2e-05,
      "loss": 4.5594,
      "step": 1
    },
    {
      "epoch": 0.01,
      "learning_rate": 4e-05,
      "loss": 3.6496,
      "step": 2
    },
    {
      "epoch": 0.02,
      "learning_rate": 6e-05,
      "loss": 4.4924,
      "step": 3
    },
    {
      "epoch": 0.02,
      "learning_rate": 8e-05,
      "loss": 4.4171,
      "step": 4
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0001,
      "loss": 4.5836,
      "step": 5
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00012,
      "loss": 4.0585,
      "step": 6
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00014,
      "loss": 3.8486,
      "step": 7
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00016,
      "loss": 2.9259,
      "step": 8
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00018,
      "loss": 2.255,
      "step": 9
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.6949,
      "step": 10
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019999650946872738,
      "loss": 1.3034,
      "step": 11
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019998603811858571,
      "loss": 0.851,
      "step": 12
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019996858668058646,
      "loss": 0.3898,
      "step": 13
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019994415637302547,
      "loss": 0.4398,
      "step": 14
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019991274890139774,
      "loss": 0.3457,
      "step": 15
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0001998743664582786,
      "loss": 0.2632,
      "step": 16
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019982901172317031,
      "loss": 0.1759,
      "step": 17
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019977668786231534,
      "loss": 0.2608,
      "step": 18
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019971739852847514,
      "loss": 0.2244,
      "step": 19
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019965114786067516,
      "loss": 0.2069,
      "step": 20
    },
    {
      "epoch": 0.1,
      "eval_loss": NaN,
      "eval_runtime": 25.9254,
      "eval_samples_per_second": 2.97,
      "eval_steps_per_second": 0.501,
      "step": 20
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019957794048391602,
      "loss": 0.1751,
      "step": 21
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019949778150885042,
      "loss": 0.1372,
      "step": 22
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019941067653142657,
      "loss": 0.153,
      "step": 23
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019931663163249742,
      "loss": 0.2147,
      "step": 24
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0001992156533773962,
      "loss": 0.1882,
      "step": 25
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.000199107748815478,
      "loss": 0.1729,
      "step": 26
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019899292547962788,
      "loss": 0.1615,
      "step": 27
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001988711913857346,
      "loss": 0.1544,
      "step": 28
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019874255503213152,
      "loss": 0.1678,
      "step": 29
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019860702539900287,
      "loss": 0.1869,
      "step": 30
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019846461194775712,
      "loss": 0.2034,
      "step": 31
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00019831532462036636,
      "loss": 0.077,
      "step": 32
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001981591738386723,
      "loss": 0.1398,
      "step": 33
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001979961705036587,
      "loss": 0.1344,
      "step": 34
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001978263259946903,
      "loss": 0.1063,
      "step": 35
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019764965216871846,
      "loss": 0.1098,
      "step": 36
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019746616135945356,
      "loss": 0.0715,
      "step": 37
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019727586637650373,
      "loss": 0.1203,
      "step": 38
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001970787805044807,
      "loss": 0.1233,
      "step": 39
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00019687491750207254,
      "loss": 0.0986,
      "step": 40
    },
    {
      "epoch": 0.21,
      "eval_loss": NaN,
      "eval_runtime": 25.9407,
      "eval_samples_per_second": 2.968,
      "eval_steps_per_second": 0.501,
      "step": 40
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001966642916010829,
      "loss": 0.105,
      "step": 41
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00019644691750543767,
      "loss": 0.1288,
      "step": 42
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00019622281039015843,
      "loss": 0.0819,
      "step": 43
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001959919859003031,
      "loss": 0.1136,
      "step": 44
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00019575446014987363,
      "loss": 0.1384,
      "step": 45
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00019551024972069126,
      "loss": 0.1608,
      "step": 46
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00019525937166123877,
      "loss": 0.1045,
      "step": 47
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00019500184348547042,
      "loss": 0.0925,
      "step": 48
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001947376831715892,
      "loss": 0.1484,
      "step": 49
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001944669091607919,
      "loss": 0.101,
      "step": 50
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00019418954035598152,
      "loss": 0.1111,
      "step": 51
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001939055961204478,
      "loss": 0.1366,
      "step": 52
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00019361509627651547,
      "loss": 0.0945,
      "step": 53
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00019331806110416027,
      "loss": 0.1221,
      "step": 54
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001930145113395934,
      "loss": 0.1544,
      "step": 55
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00019270446817381377,
      "loss": 0.0912,
      "step": 56
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0001923879532511287,
      "loss": 0.1422,
      "step": 57
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019206498866764288,
      "loss": 0.1247,
      "step": 58
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019173559696971594,
      "loss": 0.1055,
      "step": 59
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019139980115238827,
      "loss": 0.1101,
      "step": 60
    },
    {
      "epoch": 0.31,
      "eval_loss": NaN,
      "eval_runtime": 25.9123,
      "eval_samples_per_second": 2.972,
      "eval_steps_per_second": 0.502,
      "step": 60
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019105762465777598,
      "loss": 0.1395,
      "step": 61
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019070909137343408,
      "loss": 0.0826,
      "step": 62
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001903542256306892,
      "loss": 0.1636,
      "step": 63
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001899930522029408,
      "loss": 0.1236,
      "step": 64
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00018962559630393173,
      "loss": 0.0889,
      "step": 65
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00018925188358598813,
      "loss": 0.0861,
      "step": 66
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00018887194013822862,
      "loss": 0.1157,
      "step": 67
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00018848579248474288,
      "loss": 0.1127,
      "step": 68
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00018809346758274012,
      "loss": 0.0742,
      "step": 69
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00018769499282066717,
      "loss": 0.1302,
      "step": 70
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00018729039601629634,
      "loss": 0.1157,
      "step": 71
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00018687970541478364,
      "loss": 0.1093,
      "step": 72
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00018646294968669685,
      "loss": 0.0923,
      "step": 73
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00018604015792601396,
      "loss": 0.1522,
      "step": 74
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00018561135964809223,
      "loss": 0.1168,
      "step": 75
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001851765847876076,
      "loss": 0.1052,
      "step": 76
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00018473586369646512,
      "loss": 0.1167,
      "step": 77
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001842892271416797,
      "loss": 0.0486,
      "step": 78
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00018383670630322865,
      "loss": 0.1137,
      "step": 79
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00018337833277187472,
      "loss": 0.072,
      "step": 80
    },
    {
      "epoch": 0.41,
      "eval_loss": NaN,
      "eval_runtime": 25.9556,
      "eval_samples_per_second": 2.967,
      "eval_steps_per_second": 0.501,
      "step": 80
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00018291413854696077,
      "loss": 0.1083,
      "step": 81
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00018244415603417603,
      "loss": 0.1447,
      "step": 82
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00018196841804329368,
      "loss": 0.0932,
      "step": 83
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00018148695778588033,
      "loss": 0.123,
      "step": 84
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018099980887297781,
      "loss": 0.0981,
      "step": 85
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001805070053127563,
      "loss": 0.0944,
      "step": 86
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001800085815081406,
      "loss": 0.1077,
      "step": 87
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001795045722544083,
      "loss": 0.0989,
      "step": 88
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001789950127367606,
      "loss": 0.0765,
      "step": 89
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001784799385278661,
      "loss": 0.0971,
      "step": 90
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017795938558537753,
      "loss": 0.0822,
      "step": 91
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00017743339024942135,
      "loss": 0.1018,
      "step": 92
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00017690198924006097,
      "loss": 0.0854,
      "step": 93
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00017636521965473323,
      "loss": 0.0962,
      "step": 94
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00017582311896565856,
      "loss": 0.0785,
      "step": 95
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00017527572501722512,
      "loss": 0.0835,
      "step": 96
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00017472307602334693,
      "loss": 0.103,
      "step": 97
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00017416521056479577,
      "loss": 0.0849,
      "step": 98
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00017360216758650828,
      "loss": 0.0925,
      "step": 99
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00017303398639486695,
      "loss": 0.1258,
      "step": 100
    },
    {
      "epoch": 0.52,
      "eval_loss": NaN,
      "eval_runtime": 25.9434,
      "eval_samples_per_second": 2.968,
      "eval_steps_per_second": 0.501,
      "step": 100
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00017246070665495608,
      "loss": 0.1097,
      "step": 101
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00017188236838779295,
      "loss": 0.1229,
      "step": 102
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00017129901196753363,
      "loss": 0.1071,
      "step": 103
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.0965,
      "step": 104
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00017011740791311026,
      "loss": 0.1001,
      "step": 105
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00016951924276746425,
      "loss": 0.1166,
      "step": 106
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00016891622443999965,
      "loss": 0.0838,
      "step": 107
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001683083950278031,
      "loss": 0.1061,
      "step": 108
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00016769579696382597,
      "loss": 0.051,
      "step": 109
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00016707847301392236,
      "loss": 0.1051,
      "step": 110
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001664564662738632,
      "loss": 0.0886,
      "step": 111
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00016582982016632818,
      "loss": 0.0948,
      "step": 112
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00016519857843787388,
      "loss": 0.0743,
      "step": 113
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00016456278515588024,
      "loss": 0.0836,
      "step": 114
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00016392248470547394,
      "loss": 0.0574,
      "step": 115
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00016327772178642986,
      "loss": 0.0571,
      "step": 116
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001626285414100507,
      "loss": 0.0816,
      "step": 117
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00016197498889602448,
      "loss": 0.1027,
      "step": 118
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00016131710986926108,
      "loss": 0.0773,
      "step": 119
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00016065495025670675,
      "loss": 0.0675,
      "step": 120
    },
    {
      "epoch": 0.62,
      "eval_loss": NaN,
      "eval_runtime": 25.956,
      "eval_samples_per_second": 2.967,
      "eval_steps_per_second": 0.501,
      "step": 120
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00015998855628413823,
      "loss": 0.093,
      "step": 121
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00015931797447293552,
      "loss": 0.0798,
      "step": 122
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00015864325163683431,
      "loss": 0.0861,
      "step": 123
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00015796443487865776,
      "loss": 0.0795,
      "step": 124
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00015728157158702833,
      "loss": 0.0671,
      "step": 125
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00015659470943305955,
      "loss": 0.1171,
      "step": 126
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001559038963670279,
      "loss": 0.0735,
      "step": 127
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00015520918061502569,
      "loss": 0.0688,
      "step": 128
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.000154510610675594,
      "loss": 0.0654,
      "step": 129
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00015380823531633729,
      "loss": 0.0576,
      "step": 130
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00015310210357051863,
      "loss": 0.0626,
      "step": 131
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00015239226473363687,
      "loss": 0.0937,
      "step": 132
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00015167876835998524,
      "loss": 0.0838,
      "step": 133
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00015096166425919175,
      "loss": 0.1016,
      "step": 134
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00015024100249274227,
      "loss": 0.0712,
      "step": 135
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00014951683337048537,
      "loss": 0.0473,
      "step": 136
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0001487892074471205,
      "loss": 0.0833,
      "step": 137
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00014805817551866838,
      "loss": 0.0683,
      "step": 138
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00014732378861892525,
      "loss": 0.0616,
      "step": 139
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00014658609801589982,
      "loss": 0.0728,
      "step": 140
    },
    {
      "epoch": 0.72,
      "eval_loss": NaN,
      "eval_runtime": 25.9364,
      "eval_samples_per_second": 2.969,
      "eval_steps_per_second": 0.501,
      "step": 140
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00014584515520823453,
      "loss": 0.0567,
      "step": 141
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00014510101192161018,
      "loss": 0.0842,
      "step": 142
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00014435372010513509,
      "loss": 0.0906,
      "step": 143
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001436033319277183,
      "loss": 0.0956,
      "step": 144
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001428498997744278,
      "loss": 0.0864,
      "step": 145
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001420934762428335,
      "loss": 0.0729,
      "step": 146
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00014133411413933523,
      "loss": 0.0519,
      "step": 147
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001405718664754764,
      "loss": 0.0563,
      "step": 148
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001398067864642431,
      "loss": 0.096,
      "step": 149
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00013903892751634947,
      "loss": 0.0624,
      "step": 150
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.000138268343236509,
      "loss": 0.0916,
      "step": 151
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00013749508741969213,
      "loss": 0.0558,
      "step": 152
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001367192140473711,
      "loss": 0.0724,
      "step": 153
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00013594077728375128,
      "loss": 0.0798,
      "step": 154
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00013515983147199007,
      "loss": 0.0818,
      "step": 155
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00013437643113040301,
      "loss": 0.045,
      "step": 156
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001335906309486579,
      "loss": 0.0893,
      "step": 157
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001328024857839569,
      "loss": 0.0913,
      "step": 158
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00013201205065720698,
      "loss": 0.0589,
      "step": 159
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00013121938074917865,
      "loss": 0.115,
      "step": 160
    },
    {
      "epoch": 0.83,
      "eval_loss": NaN,
      "eval_runtime": 25.9367,
      "eval_samples_per_second": 2.969,
      "eval_steps_per_second": 0.501,
      "step": 160
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00013042453139665397,
      "loss": 0.0724,
      "step": 161
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00012962755808856342,
      "loss": 0.0485,
      "step": 162
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00012882851646211206,
      "loss": 0.0852,
      "step": 163
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00012802746229889563,
      "loss": 0.0483,
      "step": 164
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00012722445152100624,
      "loss": 0.1128,
      "step": 165
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00012641954018712863,
      "loss": 0.0585,
      "step": 166
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00012561278448862634,
      "loss": 0.0746,
      "step": 167
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00012480424074561933,
      "loss": 0.0728,
      "step": 168
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00012399396540305205,
      "loss": 0.0507,
      "step": 169
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00012318201502675285,
      "loss": 0.0532,
      "step": 170
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00012236844629948538,
      "loss": 0.0922,
      "step": 171
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00012155331601699136,
      "loss": 0.058,
      "step": 172
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00012073668108402565,
      "loss": 0.0576,
      "step": 173
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001199185985103836,
      "loss": 0.0635,
      "step": 174
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00011909912540692148,
      "loss": 0.0485,
      "step": 175
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00011827831898156905,
      "loss": 0.0722,
      "step": 176
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00011745623653533633,
      "loss": 0.0338,
      "step": 177
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00011663293545831302,
      "loss": 0.0511,
      "step": 178
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00011580847322566224,
      "loss": 0.0529,
      "step": 179
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00011498290739360815,
      "loss": 0.0769,
      "step": 180
    },
    {
      "epoch": 0.93,
      "eval_loss": NaN,
      "eval_runtime": 25.9307,
      "eval_samples_per_second": 2.969,
      "eval_steps_per_second": 0.501,
      "step": 180
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00011415629559541782,
      "loss": 0.0662,
      "step": 181
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001133286955373779,
      "loss": 0.0652,
      "step": 182
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00011250016499476602,
      "loss": 0.0704,
      "step": 183
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00011167076180781764,
      "loss": 0.067,
      "step": 184
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001108405438776879,
      "loss": 0.0504,
      "step": 185
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00011000956916240985,
      "loss": 0.0779,
      "step": 186
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00010917789567284804,
      "loss": 0.0576,
      "step": 187
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.000108345581468649,
      "loss": 0.0527,
      "step": 188
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00010751268465418785,
      "loss": 0.0635,
      "step": 189
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00010667926337451217,
      "loss": 0.059,
      "step": 190
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00010584537581128276,
      "loss": 0.0727,
      "step": 191
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00010501108017871192,
      "loss": 0.0365,
      "step": 192
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00010417643471949967,
      "loss": 0.0622,
      "step": 193
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00010334149770076747,
      "loss": 0.0656,
      "step": 194
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00010250632740999084,
      "loss": 0.0847,
      "step": 195
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00010167098215093009,
      "loss": 0.0576,
      "step": 196
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00010083552023956031,
      "loss": 0.0275,
      "step": 197
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0001,
      "loss": 0.0382,
      "step": 198
    },
    {
      "epoch": 1.03,
      "learning_rate": 9.916447976043972e-05,
      "loss": 0.047,
      "step": 199
    },
    {
      "epoch": 1.03,
      "learning_rate": 9.83290178490699e-05,
      "loss": 0.0609,
      "step": 200
    },
    {
      "epoch": 1.03,
      "eval_loss": NaN,
      "eval_runtime": 25.9376,
      "eval_samples_per_second": 2.969,
      "eval_steps_per_second": 0.501,
      "step": 200
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.749367259000918e-05,
      "loss": 0.0404,
      "step": 201
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.665850229923258e-05,
      "loss": 0.036,
      "step": 202
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.582356528050034e-05,
      "loss": 0.0566,
      "step": 203
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.498891982128809e-05,
      "loss": 0.0585,
      "step": 204
    },
    {
      "epoch": 1.06,
      "learning_rate": 9.41546241887173e-05,
      "loss": 0.055,
      "step": 205
    },
    {
      "epoch": 1.06,
      "learning_rate": 9.332073662548784e-05,
      "loss": 0.0494,
      "step": 206
    },
    {
      "epoch": 1.07,
      "learning_rate": 9.248731534581218e-05,
      "loss": 0.0408,
      "step": 207
    },
    {
      "epoch": 1.07,
      "learning_rate": 9.165441853135104e-05,
      "loss": 0.0368,
      "step": 208
    },
    {
      "epoch": 1.08,
      "learning_rate": 9.082210432715197e-05,
      "loss": 0.0383,
      "step": 209
    },
    {
      "epoch": 1.08,
      "learning_rate": 8.999043083759017e-05,
      "loss": 0.0289,
      "step": 210
    },
    {
      "epoch": 1.09,
      "learning_rate": 8.915945612231214e-05,
      "loss": 0.0602,
      "step": 211
    },
    {
      "epoch": 1.09,
      "learning_rate": 8.832923819218238e-05,
      "loss": 0.0736,
      "step": 212
    },
    {
      "epoch": 1.1,
      "learning_rate": 8.7499835005234e-05,
      "loss": 0.031,
      "step": 213
    },
    {
      "epoch": 1.1,
      "learning_rate": 8.667130446262214e-05,
      "loss": 0.0456,
      "step": 214
    },
    {
      "epoch": 1.11,
      "learning_rate": 8.584370440458219e-05,
      "loss": 0.0506,
      "step": 215
    },
    {
      "epoch": 1.12,
      "learning_rate": 8.501709260639186e-05,
      "loss": 0.0347,
      "step": 216
    },
    {
      "epoch": 1.12,
      "learning_rate": 8.419152677433778e-05,
      "loss": 0.0347,
      "step": 217
    },
    {
      "epoch": 1.13,
      "learning_rate": 8.336706454168701e-05,
      "loss": 0.0639,
      "step": 218
    },
    {
      "epoch": 1.13,
      "learning_rate": 8.254376346466371e-05,
      "loss": 0.0402,
      "step": 219
    },
    {
      "epoch": 1.14,
      "learning_rate": 8.172168101843099e-05,
      "loss": 0.0881,
      "step": 220
    },
    {
      "epoch": 1.14,
      "eval_loss": NaN,
      "eval_runtime": 25.953,
      "eval_samples_per_second": 2.967,
      "eval_steps_per_second": 0.501,
      "step": 220
    },
    {
      "epoch": 1.14,
      "learning_rate": 8.090087459307854e-05,
      "loss": 0.0759,
      "step": 221
    },
    {
      "epoch": 1.15,
      "learning_rate": 8.008140148961641e-05,
      "loss": 0.0508,
      "step": 222
    },
    {
      "epoch": 1.15,
      "learning_rate": 7.926331891597436e-05,
      "loss": 0.0799,
      "step": 223
    },
    {
      "epoch": 1.16,
      "learning_rate": 7.844668398300865e-05,
      "loss": 0.0477,
      "step": 224
    },
    {
      "epoch": 1.16,
      "learning_rate": 7.763155370051465e-05,
      "loss": 0.0703,
      "step": 225
    },
    {
      "epoch": 1.17,
      "learning_rate": 7.681798497324716e-05,
      "loss": 0.0561,
      "step": 226
    },
    {
      "epoch": 1.17,
      "learning_rate": 7.600603459694799e-05,
      "loss": 0.052,
      "step": 227
    },
    {
      "epoch": 1.18,
      "learning_rate": 7.519575925438067e-05,
      "loss": 0.0425,
      "step": 228
    },
    {
      "epoch": 1.18,
      "learning_rate": 7.438721551137367e-05,
      "loss": 0.0244,
      "step": 229
    },
    {
      "epoch": 1.19,
      "learning_rate": 7.358045981287141e-05,
      "loss": 0.0342,
      "step": 230
    },
    {
      "epoch": 1.19,
      "learning_rate": 7.277554847899378e-05,
      "loss": 0.0765,
      "step": 231
    },
    {
      "epoch": 1.2,
      "learning_rate": 7.197253770110438e-05,
      "loss": 0.0461,
      "step": 232
    },
    {
      "epoch": 1.2,
      "learning_rate": 7.117148353788797e-05,
      "loss": 0.0616,
      "step": 233
    },
    {
      "epoch": 1.21,
      "learning_rate": 7.037244191143661e-05,
      "loss": 0.0647,
      "step": 234
    },
    {
      "epoch": 1.21,
      "learning_rate": 6.957546860334604e-05,
      "loss": 0.0327,
      "step": 235
    },
    {
      "epoch": 1.22,
      "learning_rate": 6.878061925082137e-05,
      "loss": 0.0329,
      "step": 236
    },
    {
      "epoch": 1.22,
      "learning_rate": 6.798794934279304e-05,
      "loss": 0.0302,
      "step": 237
    },
    {
      "epoch": 1.23,
      "learning_rate": 6.719751421604309e-05,
      "loss": 0.044,
      "step": 238
    },
    {
      "epoch": 1.23,
      "learning_rate": 6.640936905134212e-05,
      "loss": 0.039,
      "step": 239
    },
    {
      "epoch": 1.24,
      "learning_rate": 6.562356886959704e-05,
      "loss": 0.0674,
      "step": 240
    },
    {
      "epoch": 1.24,
      "eval_loss": NaN,
      "eval_runtime": 25.9254,
      "eval_samples_per_second": 2.97,
      "eval_steps_per_second": 0.501,
      "step": 240
    },
    {
      "epoch": 1.24,
      "learning_rate": 6.484016852800993e-05,
      "loss": 0.0433,
      "step": 241
    },
    {
      "epoch": 1.25,
      "learning_rate": 6.405922271624874e-05,
      "loss": 0.0658,
      "step": 242
    },
    {
      "epoch": 1.25,
      "learning_rate": 6.328078595262894e-05,
      "loss": 0.0458,
      "step": 243
    },
    {
      "epoch": 1.26,
      "learning_rate": 6.250491258030791e-05,
      "loss": 0.0362,
      "step": 244
    },
    {
      "epoch": 1.27,
      "learning_rate": 6.173165676349103e-05,
      "loss": 0.0242,
      "step": 245
    },
    {
      "epoch": 1.27,
      "learning_rate": 6.0961072483650526e-05,
      "loss": 0.0668,
      "step": 246
    },
    {
      "epoch": 1.28,
      "learning_rate": 6.019321353575692e-05,
      "loss": 0.0341,
      "step": 247
    },
    {
      "epoch": 1.28,
      "learning_rate": 5.9428133524523646e-05,
      "loss": 0.0574,
      "step": 248
    },
    {
      "epoch": 1.29,
      "learning_rate": 5.866588586066481e-05,
      "loss": 0.0179,
      "step": 249
    },
    {
      "epoch": 1.29,
      "learning_rate": 5.790652375716652e-05,
      "loss": 0.032,
      "step": 250
    },
    {
      "epoch": 1.3,
      "learning_rate": 5.71501002255722e-05,
      "loss": 0.0383,
      "step": 251
    },
    {
      "epoch": 1.3,
      "learning_rate": 5.639666807228175e-05,
      "loss": 0.0473,
      "step": 252
    },
    {
      "epoch": 1.31,
      "learning_rate": 5.564627989486494e-05,
      "loss": 0.0824,
      "step": 253
    },
    {
      "epoch": 1.31,
      "learning_rate": 5.48989880783898e-05,
      "loss": 0.0464,
      "step": 254
    },
    {
      "epoch": 1.32,
      "learning_rate": 5.415484479176549e-05,
      "loss": 0.0485,
      "step": 255
    },
    {
      "epoch": 1.32,
      "learning_rate": 5.341390198410019e-05,
      "loss": 0.0304,
      "step": 256
    },
    {
      "epoch": 1.33,
      "learning_rate": 5.26762113810748e-05,
      "loss": 0.0942,
      "step": 257
    },
    {
      "epoch": 1.33,
      "learning_rate": 5.1941824481331626e-05,
      "loss": 0.0481,
      "step": 258
    },
    {
      "epoch": 1.34,
      "learning_rate": 5.1210792552879525e-05,
      "loss": 0.0447,
      "step": 259
    },
    {
      "epoch": 1.34,
      "learning_rate": 5.0483166629514654e-05,
      "loss": 0.0476,
      "step": 260
    },
    {
      "epoch": 1.34,
      "eval_loss": NaN,
      "eval_runtime": 25.9334,
      "eval_samples_per_second": 2.969,
      "eval_steps_per_second": 0.501,
      "step": 260
    },
    {
      "epoch": 1.35,
      "learning_rate": 4.9758997507257763e-05,
      "loss": 0.0506,
      "step": 261
    },
    {
      "epoch": 1.35,
      "learning_rate": 4.903833574080825e-05,
      "loss": 0.0413,
      "step": 262
    },
    {
      "epoch": 1.36,
      "learning_rate": 4.8321231640014794e-05,
      "loss": 0.0283,
      "step": 263
    },
    {
      "epoch": 1.36,
      "learning_rate": 4.760773526636315e-05,
      "loss": 0.0524,
      "step": 264
    },
    {
      "epoch": 1.37,
      "learning_rate": 4.68978964294814e-05,
      "loss": 0.0434,
      "step": 265
    },
    {
      "epoch": 1.37,
      "learning_rate": 4.6191764683662744e-05,
      "loss": 0.0435,
      "step": 266
    },
    {
      "epoch": 1.38,
      "learning_rate": 4.548938932440604e-05,
      "loss": 0.0347,
      "step": 267
    },
    {
      "epoch": 1.38,
      "learning_rate": 4.479081938497435e-05,
      "loss": 0.0611,
      "step": 268
    },
    {
      "epoch": 1.39,
      "learning_rate": 4.409610363297211e-05,
      "loss": 0.0321,
      "step": 269
    },
    {
      "epoch": 1.39,
      "learning_rate": 4.340529056694047e-05,
      "loss": 0.0447,
      "step": 270
    },
    {
      "epoch": 1.4,
      "learning_rate": 4.271842841297169e-05,
      "loss": 0.0387,
      "step": 271
    },
    {
      "epoch": 1.4,
      "learning_rate": 4.2035565121342246e-05,
      "loss": 0.0563,
      "step": 272
    },
    {
      "epoch": 1.41,
      "learning_rate": 4.135674836316569e-05,
      "loss": 0.0254,
      "step": 273
    },
    {
      "epoch": 1.41,
      "learning_rate": 4.0682025527064486e-05,
      "loss": 0.0413,
      "step": 274
    },
    {
      "epoch": 1.42,
      "learning_rate": 4.001144371586181e-05,
      "loss": 0.0509,
      "step": 275
    },
    {
      "epoch": 1.43,
      "learning_rate": 3.934504974329326e-05,
      "loss": 0.0545,
      "step": 276
    },
    {
      "epoch": 1.43,
      "learning_rate": 3.868289013073893e-05,
      "loss": 0.0435,
      "step": 277
    },
    {
      "epoch": 1.44,
      "learning_rate": 3.802501110397553e-05,
      "loss": 0.0404,
      "step": 278
    },
    {
      "epoch": 1.44,
      "learning_rate": 3.737145858994934e-05,
      "loss": 0.0505,
      "step": 279
    },
    {
      "epoch": 1.45,
      "learning_rate": 3.672227821357014e-05,
      "loss": 0.0259,
      "step": 280
    },
    {
      "epoch": 1.45,
      "eval_loss": NaN,
      "eval_runtime": 25.9372,
      "eval_samples_per_second": 2.969,
      "eval_steps_per_second": 0.501,
      "step": 280
    },
    {
      "epoch": 1.45,
      "learning_rate": 3.607751529452609e-05,
      "loss": 0.0576,
      "step": 281
    },
    {
      "epoch": 1.46,
      "learning_rate": 3.543721484411976e-05,
      "loss": 0.0383,
      "step": 282
    },
    {
      "epoch": 1.46,
      "learning_rate": 3.4801421562126144e-05,
      "loss": 0.0329,
      "step": 283
    },
    {
      "epoch": 1.47,
      "learning_rate": 3.4170179833671846e-05,
      "loss": 0.0297,
      "step": 284
    },
    {
      "epoch": 1.47,
      "learning_rate": 3.354353372613678e-05,
      "loss": 0.0412,
      "step": 285
    },
    {
      "epoch": 1.48,
      "learning_rate": 3.292152698607768e-05,
      "loss": 0.0748,
      "step": 286
    },
    {
      "epoch": 1.48,
      "learning_rate": 3.2304203036174016e-05,
      "loss": 0.0439,
      "step": 287
    },
    {
      "epoch": 1.49,
      "learning_rate": 3.169160497219692e-05,
      "loss": 0.0381,
      "step": 288
    },
    {
      "epoch": 1.49,
      "learning_rate": 3.108377556000037e-05,
      "loss": 0.0369,
      "step": 289
    },
    {
      "epoch": 1.5,
      "learning_rate": 3.0480757232535772e-05,
      "loss": 0.0489,
      "step": 290
    },
    {
      "epoch": 1.5,
      "learning_rate": 2.9882592086889736e-05,
      "loss": 0.0408,
      "step": 291
    },
    {
      "epoch": 1.51,
      "learning_rate": 2.9289321881345254e-05,
      "loss": 0.0428,
      "step": 292
    },
    {
      "epoch": 1.51,
      "learning_rate": 2.8700988032466392e-05,
      "loss": 0.0397,
      "step": 293
    },
    {
      "epoch": 1.52,
      "learning_rate": 2.8117631612207084e-05,
      "loss": 0.06,
      "step": 294
    },
    {
      "epoch": 1.52,
      "learning_rate": 2.7539293345043926e-05,
      "loss": 0.0506,
      "step": 295
    },
    {
      "epoch": 1.53,
      "learning_rate": 2.6966013605133088e-05,
      "loss": 0.0551,
      "step": 296
    },
    {
      "epoch": 1.53,
      "learning_rate": 2.639783241349174e-05,
      "loss": 0.06,
      "step": 297
    },
    {
      "epoch": 1.54,
      "learning_rate": 2.5834789435204243e-05,
      "loss": 0.0419,
      "step": 298
    },
    {
      "epoch": 1.54,
      "learning_rate": 2.527692397665311e-05,
      "loss": 0.0284,
      "step": 299
    },
    {
      "epoch": 1.55,
      "learning_rate": 2.4724274982774865e-05,
      "loss": 0.0534,
      "step": 300
    },
    {
      "epoch": 1.55,
      "eval_loss": NaN,
      "eval_runtime": 25.9307,
      "eval_samples_per_second": 2.969,
      "eval_steps_per_second": 0.501,
      "step": 300
    },
    {
      "epoch": 1.55,
      "learning_rate": 2.417688103434146e-05,
      "loss": 0.0285,
      "step": 301
    },
    {
      "epoch": 1.56,
      "learning_rate": 2.3634780345266806e-05,
      "loss": 0.0133,
      "step": 302
    },
    {
      "epoch": 1.56,
      "learning_rate": 2.3098010759939036e-05,
      "loss": 0.0376,
      "step": 303
    },
    {
      "epoch": 1.57,
      "learning_rate": 2.2566609750578673e-05,
      "loss": 0.0269,
      "step": 304
    },
    {
      "epoch": 1.57,
      "learning_rate": 2.2040614414622484e-05,
      "loss": 0.0314,
      "step": 305
    },
    {
      "epoch": 1.58,
      "learning_rate": 2.1520061472133902e-05,
      "loss": 0.0275,
      "step": 306
    },
    {
      "epoch": 1.59,
      "learning_rate": 2.100498726323943e-05,
      "loss": 0.0477,
      "step": 307
    },
    {
      "epoch": 1.59,
      "learning_rate": 2.04954277455917e-05,
      "loss": 0.0432,
      "step": 308
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.999141849185938e-05,
      "loss": 0.0682,
      "step": 309
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.9492994687243714e-05,
      "loss": 0.0383,
      "step": 310
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.9000191127022237e-05,
      "loss": 0.0256,
      "step": 311
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.851304221411967e-05,
      "loss": 0.0313,
      "step": 312
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.8031581956706346e-05,
      "loss": 0.0635,
      "step": 313
    },
    {
      "epoch": 1.62,
      "learning_rate": 1.7555843965823992e-05,
      "loss": 0.0472,
      "step": 314
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.7085861453039243e-05,
      "loss": 0.038,
      "step": 315
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.6621667228125302e-05,
      "loss": 0.0363,
      "step": 316
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.616329369677134e-05,
      "loss": 0.077,
      "step": 317
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.57107728583203e-05,
      "loss": 0.0436,
      "step": 318
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.5264136303534894e-05,
      "loss": 0.0345,
      "step": 319
    },
    {
      "epoch": 1.65,
      "learning_rate": 1.4823415212392377e-05,
      "loss": 0.0449,
      "step": 320
    },
    {
      "epoch": 1.65,
      "eval_loss": NaN,
      "eval_runtime": 25.9493,
      "eval_samples_per_second": 2.967,
      "eval_steps_per_second": 0.501,
      "step": 320
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.4388640351907801e-05,
      "loss": 0.0572,
      "step": 321
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.3959842073986085e-05,
      "loss": 0.0398,
      "step": 322
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.3537050313303179e-05,
      "loss": 0.0373,
      "step": 323
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.3120294585216353e-05,
      "loss": 0.0194,
      "step": 324
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.2709603983703678e-05,
      "loss": 0.0489,
      "step": 325
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.230500717933285e-05,
      "loss": 0.0385,
      "step": 326
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.1906532417259886e-05,
      "loss": 0.0782,
      "step": 327
    },
    {
      "epoch": 1.69,
      "learning_rate": 1.1514207515257147e-05,
      "loss": 0.0355,
      "step": 328
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.11280598617714e-05,
      "loss": 0.0239,
      "step": 329
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.0748116414011888e-05,
      "loss": 0.03,
      "step": 330
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.0374403696068291e-05,
      "loss": 0.0356,
      "step": 331
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.0006947797059219e-05,
      "loss": 0.0803,
      "step": 332
    },
    {
      "epoch": 1.72,
      "learning_rate": 9.64577436931081e-06,
      "loss": 0.0358,
      "step": 333
    },
    {
      "epoch": 1.72,
      "learning_rate": 9.29090862656593e-06,
      "loss": 0.0155,
      "step": 334
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.942375342224041e-06,
      "loss": 0.0522,
      "step": 335
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.600198847611729e-06,
      "loss": 0.0484,
      "step": 336
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.264403030284074e-06,
      "loss": 0.0417,
      "step": 337
    },
    {
      "epoch": 1.75,
      "learning_rate": 7.935011332357112e-06,
      "loss": 0.038,
      "step": 338
    },
    {
      "epoch": 1.75,
      "learning_rate": 7.612046748871327e-06,
      "loss": 0.0379,
      "step": 339
    },
    {
      "epoch": 1.76,
      "learning_rate": 7.295531826186264e-06,
      "loss": 0.0325,
      "step": 340
    },
    {
      "epoch": 1.76,
      "eval_loss": NaN,
      "eval_runtime": 25.9429,
      "eval_samples_per_second": 2.968,
      "eval_steps_per_second": 0.501,
      "step": 340
    },
    {
      "epoch": 1.76,
      "learning_rate": 6.985488660406625e-06,
      "loss": 0.0148,
      "step": 341
    },
    {
      "epoch": 1.77,
      "learning_rate": 6.681938895839746e-06,
      "loss": 0.0515,
      "step": 342
    },
    {
      "epoch": 1.77,
      "learning_rate": 6.3849037234845584e-06,
      "loss": 0.0432,
      "step": 343
    },
    {
      "epoch": 1.78,
      "learning_rate": 6.094403879552213e-06,
      "loss": 0.0269,
      "step": 344
    },
    {
      "epoch": 1.78,
      "learning_rate": 5.810459644018506e-06,
      "loss": 0.0349,
      "step": 345
    },
    {
      "epoch": 1.79,
      "learning_rate": 5.533090839208133e-06,
      "loss": 0.0098,
      "step": 346
    },
    {
      "epoch": 1.79,
      "learning_rate": 5.262316828410807e-06,
      "loss": 0.0267,
      "step": 347
    },
    {
      "epoch": 1.8,
      "learning_rate": 4.998156514529595e-06,
      "loss": 0.028,
      "step": 348
    },
    {
      "epoch": 1.8,
      "learning_rate": 4.740628338761255e-06,
      "loss": 0.0534,
      "step": 349
    },
    {
      "epoch": 1.81,
      "learning_rate": 4.489750279308757e-06,
      "loss": 0.0559,
      "step": 350
    },
    {
      "epoch": 1.81,
      "learning_rate": 4.24553985012639e-06,
      "loss": 0.0442,
      "step": 351
    },
    {
      "epoch": 1.82,
      "learning_rate": 4.008014099696922e-06,
      "loss": 0.035,
      "step": 352
    },
    {
      "epoch": 1.82,
      "learning_rate": 3.7771896098415695e-06,
      "loss": 0.019,
      "step": 353
    },
    {
      "epoch": 1.83,
      "learning_rate": 3.5530824945623542e-06,
      "loss": 0.0218,
      "step": 354
    },
    {
      "epoch": 1.83,
      "learning_rate": 3.3357083989171256e-06,
      "loss": 0.0524,
      "step": 355
    },
    {
      "epoch": 1.84,
      "learning_rate": 3.1250824979274675e-06,
      "loss": 0.0502,
      "step": 356
    },
    {
      "epoch": 1.84,
      "learning_rate": 2.921219495519301e-06,
      "loss": 0.0321,
      "step": 357
    },
    {
      "epoch": 1.85,
      "learning_rate": 2.7241336234962944e-06,
      "loss": 0.0173,
      "step": 358
    },
    {
      "epoch": 1.85,
      "learning_rate": 2.5338386405464377e-06,
      "loss": 0.0196,
      "step": 359
    },
    {
      "epoch": 1.86,
      "learning_rate": 2.3503478312815298e-06,
      "loss": 0.03,
      "step": 360
    },
    {
      "epoch": 1.86,
      "eval_loss": NaN,
      "eval_runtime": 25.944,
      "eval_samples_per_second": 2.968,
      "eval_steps_per_second": 0.501,
      "step": 360
    },
    {
      "epoch": 1.86,
      "learning_rate": 2.1736740053097338e-06,
      "loss": 0.0453,
      "step": 361
    },
    {
      "epoch": 1.87,
      "learning_rate": 2.003829496341325e-06,
      "loss": 0.0287,
      "step": 362
    },
    {
      "epoch": 1.87,
      "learning_rate": 1.8408261613277022e-06,
      "loss": 0.05,
      "step": 363
    },
    {
      "epoch": 1.88,
      "learning_rate": 1.684675379633649e-06,
      "loss": 0.0202,
      "step": 364
    },
    {
      "epoch": 1.88,
      "learning_rate": 1.5353880522429032e-06,
      "loss": 0.0461,
      "step": 365
    },
    {
      "epoch": 1.89,
      "learning_rate": 1.3929746009971433e-06,
      "loss": 0.0287,
      "step": 366
    },
    {
      "epoch": 1.9,
      "learning_rate": 1.257444967868482e-06,
      "loss": 0.0451,
      "step": 367
    },
    {
      "epoch": 1.9,
      "learning_rate": 1.1288086142653864e-06,
      "loss": 0.0338,
      "step": 368
    },
    {
      "epoch": 1.91,
      "learning_rate": 1.0070745203721532e-06,
      "loss": 0.0201,
      "step": 369
    },
    {
      "epoch": 1.91,
      "learning_rate": 8.922511845219971e-07,
      "loss": 0.0266,
      "step": 370
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.843466226038155e-07,
      "loss": 0.0353,
      "step": 371
    },
    {
      "epoch": 1.92,
      "learning_rate": 6.833683675025904e-07,
      "loss": 0.0359,
      "step": 372
    },
    {
      "epoch": 1.93,
      "learning_rate": 5.893234685734439e-07,
      "loss": 0.0168,
      "step": 373
    },
    {
      "epoch": 1.93,
      "learning_rate": 5.022184911495864e-07,
      "loss": 0.0396,
      "step": 374
    },
    {
      "epoch": 1.94,
      "learning_rate": 4.220595160840057e-07,
      "loss": 0.0716,
      "step": 375
    },
    {
      "epoch": 1.94,
      "learning_rate": 3.488521393248401e-07,
      "loss": 0.0455,
      "step": 376
    },
    {
      "epoch": 1.95,
      "learning_rate": 2.826014715248904e-07,
      "loss": 0.071,
      "step": 377
    },
    {
      "epoch": 1.95,
      "learning_rate": 2.2331213768468363e-07,
      "loss": 0.039,
      "step": 378
    },
    {
      "epoch": 1.96,
      "learning_rate": 1.7098827682970886e-07,
      "loss": 0.0416,
      "step": 379
    },
    {
      "epoch": 1.96,
      "learning_rate": 1.2563354172142606e-07,
      "loss": 0.0416,
      "step": 380
    },
    {
      "epoch": 1.96,
      "eval_loss": NaN,
      "eval_runtime": 25.9223,
      "eval_samples_per_second": 2.97,
      "eval_steps_per_second": 0.501,
      "step": 380
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.725109860224789e-08,
      "loss": 0.0327,
      "step": 381
    },
    {
      "epoch": 1.97,
      "learning_rate": 5.584362697453882e-08,
      "loss": 0.0516,
      "step": 382
    },
    {
      "epoch": 1.98,
      "learning_rate": 3.141331941354242e-08,
      "loss": 0.0253,
      "step": 383
    },
    {
      "epoch": 1.98,
      "learning_rate": 1.3961881414292778e-08,
      "loss": 0.0276,
      "step": 384
    },
    {
      "epoch": 1.99,
      "learning_rate": 3.4905312726207783e-09,
      "loss": 0.0121,
      "step": 385
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.0,
      "loss": 0.0193,
      "step": 386
    }
  ],
  "logging_steps": 1,
  "max_steps": 386,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 1.123898485583433e+19,
  "trial_name": null,
  "trial_params": null
}