{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 21,
  "global_step": 135,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 9.998646205897309e-05,
      "loss": 0.6057,
      "step": 1
    },
    {
      "epoch": 0.01,
      "learning_rate": 9.994585556692624e-05,
      "loss": 0.4662,
      "step": 2
    },
    {
      "epoch": 0.02,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.3514,
      "step": 3
    },
    {
      "epoch": 0.03,
      "learning_rate": 9.978353953249022e-05,
      "loss": 0.2457,
      "step": 4
    },
    {
      "epoch": 0.04,
      "learning_rate": 9.966191788709716e-05,
      "loss": 0.2152,
      "step": 5
    },
    {
      "epoch": 0.04,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.1948,
      "step": 6
    },
    {
      "epoch": 0.05,
      "learning_rate": 9.933807660562898e-05,
      "loss": 0.177,
      "step": 7
    },
    {
      "epoch": 0.06,
      "learning_rate": 9.913603233532067e-05,
      "loss": 0.1568,
      "step": 8
    },
    {
      "epoch": 0.07,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.123,
      "step": 9
    },
    {
      "epoch": 0.07,
      "learning_rate": 9.865224352899119e-05,
      "loss": 0.1174,
      "step": 10
    },
    {
      "epoch": 0.08,
      "learning_rate": 9.837076097314319e-05,
      "loss": 0.0905,
      "step": 11
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.1101,
      "step": 12
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.77293816123866e-05,
      "loss": 0.0767,
      "step": 13
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.736983212571646e-05,
      "loss": 0.067,
      "step": 14
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0757,
      "step": 15
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.657398694630712e-05,
      "loss": 0.0656,
      "step": 16
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.613812221777212e-05,
      "loss": 0.0603,
      "step": 17
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.0526,
      "step": 18
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.519168849742604e-05,
      "loss": 0.0786,
      "step": 19
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.468163201617062e-05,
      "loss": 0.0736,
      "step": 20
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0678,
      "step": 21
    },
    {
      "epoch": 0.16,
      "eval_loss": 0.06099913641810417,
      "eval_runtime": 21.2148,
      "eval_samples_per_second": 1.414,
      "eval_steps_per_second": 0.707,
      "step": 21
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.358922068483812e-05,
      "loss": 0.0372,
      "step": 22
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.300745739476829e-05,
      "loss": 0.0709,
      "step": 23
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.0753,
      "step": 24
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.177439057064683e-05,
      "loss": 0.0436,
      "step": 25
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.112375476403312e-05,
      "loss": 0.0538,
      "step": 26
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0491,
      "step": 27
    },
    {
      "epoch": 0.21,
      "learning_rate": 8.97560398247424e-05,
      "loss": 0.0532,
      "step": 28
    },
    {
      "epoch": 0.21,
      "learning_rate": 8.903970133383297e-05,
      "loss": 0.0394,
      "step": 29
    },
    {
      "epoch": 0.22,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.0319,
      "step": 30
    },
    {
      "epoch": 0.23,
      "learning_rate": 8.754400164907497e-05,
      "loss": 0.053,
      "step": 31
    },
    {
      "epoch": 0.24,
      "learning_rate": 8.676545040299145e-05,
      "loss": 0.0409,
      "step": 32
    },
    {
      "epoch": 0.24,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.0456,
      "step": 33
    },
    {
      "epoch": 0.25,
      "learning_rate": 8.51490528712831e-05,
      "loss": 0.0354,
      "step": 34
    },
    {
      "epoch": 0.26,
      "learning_rate": 8.43120818934367e-05,
      "loss": 0.0324,
      "step": 35
    },
    {
      "epoch": 0.27,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.0279,
      "step": 36
    },
    {
      "epoch": 0.27,
      "learning_rate": 8.258286144107276e-05,
      "loss": 0.0309,
      "step": 37
    },
    {
      "epoch": 0.28,
      "learning_rate": 8.169154836993551e-05,
      "loss": 0.0374,
      "step": 38
    },
    {
      "epoch": 0.29,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.0465,
      "step": 39
    },
    {
      "epoch": 0.3,
      "learning_rate": 7.985792958513931e-05,
      "loss": 0.0213,
      "step": 40
    },
    {
      "epoch": 0.3,
      "learning_rate": 7.891661680839932e-05,
      "loss": 0.0263,
      "step": 41
    },
    {
      "epoch": 0.31,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.0177,
      "step": 42
    },
    {
      "epoch": 0.31,
      "eval_loss": 0.021131165325641632,
      "eval_runtime": 21.1467,
      "eval_samples_per_second": 1.419,
      "eval_steps_per_second": 0.709,
      "step": 42
    },
    {
      "epoch": 0.32,
      "learning_rate": 7.698753289757565e-05,
      "loss": 0.0155,
      "step": 43
    },
    {
      "epoch": 0.33,
      "learning_rate": 7.600080639646077e-05,
      "loss": 0.0158,
      "step": 44
    },
    {
      "epoch": 0.33,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.0342,
      "step": 45
    },
    {
      "epoch": 0.34,
      "learning_rate": 7.398565566251232e-05,
      "loss": 0.0175,
      "step": 46
    },
    {
      "epoch": 0.35,
      "learning_rate": 7.295832266935059e-05,
      "loss": 0.0215,
      "step": 47
    },
    {
      "epoch": 0.36,
      "learning_rate": 7.191855733945387e-05,
      "loss": 0.013,
      "step": 48
    },
    {
      "epoch": 0.36,
      "learning_rate": 7.08669227240909e-05,
      "loss": 0.0136,
      "step": 49
    },
    {
      "epoch": 0.37,
      "learning_rate": 6.980398830195785e-05,
      "loss": 0.0204,
      "step": 50
    },
    {
      "epoch": 0.38,
      "learning_rate": 6.873032967079561e-05,
      "loss": 0.0166,
      "step": 51
    },
    {
      "epoch": 0.39,
      "learning_rate": 6.764652823569344e-05,
      "loss": 0.0082,
      "step": 52
    },
    {
      "epoch": 0.39,
      "learning_rate": 6.65531708942479e-05,
      "loss": 0.011,
      "step": 53
    },
    {
      "epoch": 0.4,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0167,
      "step": 54
    },
    {
      "epoch": 0.41,
      "learning_rate": 6.434016163555452e-05,
      "loss": 0.0077,
      "step": 55
    },
    {
      "epoch": 0.41,
      "learning_rate": 6.322170810186012e-05,
      "loss": 0.0087,
      "step": 56
    },
    {
      "epoch": 0.42,
      "learning_rate": 6.209609477998338e-05,
      "loss": 0.0097,
      "step": 57
    },
    {
      "epoch": 0.43,
      "learning_rate": 6.096393120939516e-05,
      "loss": 0.0096,
      "step": 58
    },
    {
      "epoch": 0.44,
      "learning_rate": 5.982583047664151e-05,
      "loss": 0.0049,
      "step": 59
    },
    {
      "epoch": 0.44,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.0127,
      "step": 60
    },
    {
      "epoch": 0.45,
      "learning_rate": 5.753428561247416e-05,
      "loss": 0.0064,
      "step": 61
    },
    {
      "epoch": 0.46,
      "learning_rate": 5.6382082393029746e-05,
      "loss": 0.0068,
      "step": 62
    },
    {
      "epoch": 0.47,
      "learning_rate": 5.522642316338268e-05,
      "loss": 0.0121,
      "step": 63
    },
    {
      "epoch": 0.47,
      "eval_loss": 0.013300704769790173,
      "eval_runtime": 21.1421,
      "eval_samples_per_second": 1.419,
      "eval_steps_per_second": 0.709,
      "step": 63
    },
    {
      "epoch": 0.47,
      "learning_rate": 5.4067933733392915e-05,
      "loss": 0.0037,
      "step": 64
    },
    {
      "epoch": 0.48,
      "learning_rate": 5.290724144552379e-05,
      "loss": 0.0112,
      "step": 65
    },
    {
      "epoch": 0.49,
      "learning_rate": 5.174497483512506e-05,
      "loss": 0.0091,
      "step": 66
    },
    {
      "epoch": 0.5,
      "learning_rate": 5.0581763290069865e-05,
      "loss": 0.0175,
      "step": 67
    },
    {
      "epoch": 0.5,
      "learning_rate": 4.941823670993016e-05,
      "loss": 0.009,
      "step": 68
    },
    {
      "epoch": 0.51,
      "learning_rate": 4.825502516487497e-05,
      "loss": 0.0055,
      "step": 69
    },
    {
      "epoch": 0.52,
      "learning_rate": 4.709275855447621e-05,
      "loss": 0.0276,
      "step": 70
    },
    {
      "epoch": 0.53,
      "learning_rate": 4.593206626660709e-05,
      "loss": 0.013,
      "step": 71
    },
    {
      "epoch": 0.53,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.0102,
      "step": 72
    },
    {
      "epoch": 0.54,
      "learning_rate": 4.361791760697027e-05,
      "loss": 0.0095,
      "step": 73
    },
    {
      "epoch": 0.55,
      "learning_rate": 4.246571438752585e-05,
      "loss": 0.0077,
      "step": 74
    },
    {
      "epoch": 0.56,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.006,
      "step": 75
    },
    {
      "epoch": 0.56,
      "learning_rate": 4.017416952335849e-05,
      "loss": 0.0054,
      "step": 76
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.903606879060483e-05,
      "loss": 0.0114,
      "step": 77
    },
    {
      "epoch": 0.58,
      "learning_rate": 3.790390522001662e-05,
      "loss": 0.0044,
      "step": 78
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.67782918981399e-05,
      "loss": 0.0072,
      "step": 79
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.5659838364445505e-05,
      "loss": 0.0047,
      "step": 80
    },
    {
      "epoch": 0.6,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.0058,
      "step": 81
    },
    {
      "epoch": 0.61,
      "learning_rate": 3.34468291057521e-05,
      "loss": 0.0095,
      "step": 82
    },
    {
      "epoch": 0.61,
      "learning_rate": 3.235347176430656e-05,
      "loss": 0.0008,
      "step": 83
    },
    {
      "epoch": 0.62,
      "learning_rate": 3.12696703292044e-05,
      "loss": 0.0079,
      "step": 84
    },
    {
      "epoch": 0.62,
      "eval_loss": 0.006422529462724924,
      "eval_runtime": 21.1401,
      "eval_samples_per_second": 1.419,
      "eval_steps_per_second": 0.71,
      "step": 84
    },
    {
      "epoch": 0.63,
      "learning_rate": 3.019601169804216e-05,
      "loss": 0.009,
      "step": 85
    },
    {
      "epoch": 0.64,
      "learning_rate": 2.9133077275909108e-05,
      "loss": 0.006,
      "step": 86
    },
    {
      "epoch": 0.64,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 0.0089,
      "step": 87
    },
    {
      "epoch": 0.65,
      "learning_rate": 2.7041677330649407e-05,
      "loss": 0.0053,
      "step": 88
    },
    {
      "epoch": 0.66,
      "learning_rate": 2.6014344337487707e-05,
      "loss": 0.0047,
      "step": 89
    },
    {
      "epoch": 0.67,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.0054,
      "step": 90
    },
    {
      "epoch": 0.67,
      "learning_rate": 2.399919360353923e-05,
      "loss": 0.0014,
      "step": 91
    },
    {
      "epoch": 0.68,
      "learning_rate": 2.3012467102424373e-05,
      "loss": 0.0077,
      "step": 92
    },
    {
      "epoch": 0.69,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 0.0116,
      "step": 93
    },
    {
      "epoch": 0.7,
      "learning_rate": 2.1083383191600674e-05,
      "loss": 0.0017,
      "step": 94
    },
    {
      "epoch": 0.7,
      "learning_rate": 2.0142070414860704e-05,
      "loss": 0.0073,
      "step": 95
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 0.0056,
      "step": 96
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.8308451630064484e-05,
      "loss": 0.0025,
      "step": 97
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.7417138558927244e-05,
      "loss": 0.0029,
      "step": 98
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 0.0026,
      "step": 99
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5687918106563326e-05,
      "loss": 0.0034,
      "step": 100
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.4850947128716913e-05,
      "loss": 0.0007,
      "step": 101
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 0.0022,
      "step": 102
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.3234549597008571e-05,
      "loss": 0.005,
      "step": 103
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.245599835092504e-05,
      "loss": 0.0043,
      "step": 104
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 0.0024,
      "step": 105
    },
    {
      "epoch": 0.78,
      "eval_loss": 0.0036511807702481747,
      "eval_runtime": 21.148,
      "eval_samples_per_second": 1.419,
      "eval_steps_per_second": 0.709,
      "step": 105
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.096029866616704e-05,
      "loss": 0.0027,
      "step": 106
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.0243960175257606e-05,
      "loss": 0.006,
      "step": 107
    },
    {
      "epoch": 0.8,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0029,
      "step": 108
    },
    {
      "epoch": 0.81,
      "learning_rate": 8.876245235966885e-06,
      "loss": 0.0016,
      "step": 109
    },
    {
      "epoch": 0.81,
      "learning_rate": 8.225609429353187e-06,
      "loss": 0.0044,
      "step": 110
    },
    {
      "epoch": 0.82,
      "learning_rate": 7.597595192178702e-06,
      "loss": 0.002,
      "step": 111
    },
    {
      "epoch": 0.83,
      "learning_rate": 6.992542605231739e-06,
      "loss": 0.0054,
      "step": 112
    },
    {
      "epoch": 0.84,
      "learning_rate": 6.410779315161886e-06,
      "loss": 0.0081,
      "step": 113
    },
    {
      "epoch": 0.84,
      "learning_rate": 5.852620357053651e-06,
      "loss": 0.0078,
      "step": 114
    },
    {
      "epoch": 0.85,
      "learning_rate": 5.318367983829392e-06,
      "loss": 0.0013,
      "step": 115
    },
    {
      "epoch": 0.86,
      "learning_rate": 4.8083115025739756e-06,
      "loss": 0.0027,
      "step": 116
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.322727117869951e-06,
      "loss": 0.005,
      "step": 117
    },
    {
      "epoch": 0.87,
      "learning_rate": 3.861877782227885e-06,
      "loss": 0.0018,
      "step": 118
    },
    {
      "epoch": 0.88,
      "learning_rate": 3.426013053692878e-06,
      "loss": 0.0019,
      "step": 119
    },
    {
      "epoch": 0.89,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.0038,
      "step": 120
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.63016787428354e-06,
      "loss": 0.0005,
      "step": 121
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.2706183876134045e-06,
      "loss": 0.0067,
      "step": 122
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 0.0069,
      "step": 123
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.6292390268568104e-06,
      "loss": 0.0041,
      "step": 124
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3477564710088098e-06,
      "loss": 0.0055,
      "step": 125
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 0.0015,
      "step": 126
    },
    {
      "epoch": 0.93,
      "eval_loss": 0.0034059311728924513,
      "eval_runtime": 21.1425,
      "eval_samples_per_second": 1.419,
      "eval_steps_per_second": 0.709,
      "step": 126
    },
    {
      "epoch": 0.94,
      "learning_rate": 8.639676646793382e-07,
      "loss": 0.0045,
      "step": 127
    },
    {
      "epoch": 0.95,
      "learning_rate": 6.61923394371039e-07,
      "loss": 0.0035,
      "step": 128
    },
    {
      "epoch": 0.96,
      "learning_rate": 4.865965629214819e-07,
      "loss": 0.0009,
      "step": 129
    },
    {
      "epoch": 0.96,
      "learning_rate": 3.380821129028489e-07,
      "loss": 0.0048,
      "step": 130
    },
    {
      "epoch": 0.97,
      "learning_rate": 2.1646046750978254e-07,
      "loss": 0.001,
      "step": 131
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 0.0015,
      "step": 132
    },
    {
      "epoch": 0.99,
      "learning_rate": 5.4144433073771707e-08,
      "loss": 0.006,
      "step": 133
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3537941026914303e-08,
      "loss": 0.0023,
      "step": 134
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0,
      "loss": 0.0054,
      "step": 135
    },
    {
      "epoch": 1.0,
      "step": 135,
      "total_flos": 9.492248232198144e+16,
      "train_loss": 0.03722052312180124,
      "train_runtime": 798.3612,
      "train_samples_per_second": 0.338,
      "train_steps_per_second": 0.169
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 135,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10000,
  "total_flos": 9.492248232198144e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}