{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 0.0,
      "loss": 1.4793,
      "step": 1
    },
    {
      "epoch": 0.13,
      "learning_rate": 6.309297535714573e-05,
      "loss": 1.2107,
      "step": 2
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001,
      "loss": 1.1527,
      "step": 3
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00012618595071429146,
      "loss": 1.0989,
      "step": 4
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001464973520717927,
      "loss": 1.0834,
      "step": 5
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00016309297535714573,
      "loss": 1.038,
      "step": 6
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017712437491614223,
      "loss": 1.0052,
      "step": 7
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001892789260714372,
      "loss": 0.8922,
      "step": 8
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0002,
      "loss": 0.8794,
      "step": 9
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002,
      "loss": 0.8602,
      "step": 10
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 0.8433,
      "step": 11
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 0.7624,
      "step": 12
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 0.7019,
      "step": 13
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0002,
      "loss": 0.5855,
      "step": 14
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0002,
      "loss": 0.4916,
      "step": 15
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.41111135482788086,
      "eval_runtime": 40.2698,
      "eval_samples_per_second": 11.224,
      "eval_steps_per_second": 0.372,
      "step": 15
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0002,
      "loss": 0.3878,
      "step": 16
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0002,
      "loss": 0.3825,
      "step": 17
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0002,
      "loss": 0.2843,
      "step": 18
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0002,
      "loss": 0.3455,
      "step": 19
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0002,
      "loss": 0.2263,
      "step": 20
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0002,
      "loss": 0.2022,
      "step": 21
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0002,
      "loss": 0.1666,
      "step": 22
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 0.1677,
      "step": 23
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 0.1666,
      "step": 24
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0002,
      "loss": 0.1099,
      "step": 25
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0002,
      "loss": 0.2226,
      "step": 26
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0002,
      "loss": 0.1854,
      "step": 27
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0002,
      "loss": 0.1483,
      "step": 28
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0002,
      "loss": 0.1586,
      "step": 29
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0002,
      "loss": 0.0688,
      "step": 30
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.06789597868919373,
      "eval_runtime": 40.3947,
      "eval_samples_per_second": 11.19,
      "eval_steps_per_second": 0.371,
      "step": 30
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0002,
      "loss": 0.0834,
      "step": 31
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0002,
      "loss": 0.032,
      "step": 32
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0002,
      "loss": 0.0581,
      "step": 33
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0002,
      "loss": 0.0244,
      "step": 34
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0002,
      "loss": 0.0342,
      "step": 35
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0002,
      "loss": 0.0185,
      "step": 36
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0002,
      "loss": 0.0608,
      "step": 37
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0002,
      "loss": 0.0638,
      "step": 38
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0002,
      "loss": 0.0322,
      "step": 39
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0002,
      "loss": 0.0498,
      "step": 40
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0002,
      "loss": 0.0195,
      "step": 41
    },
    {
      "epoch": 2.8,
      "learning_rate": 0.0002,
      "loss": 0.025,
      "step": 42
    },
    {
      "epoch": 2.87,
      "learning_rate": 0.0002,
      "loss": 0.0303,
      "step": 43
    },
    {
      "epoch": 2.93,
      "learning_rate": 0.0002,
      "loss": 0.0368,
      "step": 44
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0002,
      "loss": 0.0309,
      "step": 45
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.01529462169855833,
      "eval_runtime": 40.4637,
      "eval_samples_per_second": 11.17,
      "eval_steps_per_second": 0.371,
      "step": 45
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0002,
      "loss": 0.0106,
      "step": 46
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0002,
      "loss": 0.0119,
      "step": 47
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0002,
      "loss": 0.0127,
      "step": 48
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0002,
      "loss": 0.0126,
      "step": 49
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0002,
      "loss": 0.0093,
      "step": 50
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0002,
      "loss": 0.0087,
      "step": 51
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0002,
      "loss": 0.0197,
      "step": 52
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0002,
      "loss": 0.0115,
      "step": 53
    },
    {
      "epoch": 3.6,
      "learning_rate": 0.0002,
      "loss": 0.0076,
      "step": 54
    },
    {
      "epoch": 3.67,
      "learning_rate": 0.0002,
      "loss": 0.0075,
      "step": 55
    },
    {
      "epoch": 3.73,
      "learning_rate": 0.0002,
      "loss": 0.0126,
      "step": 56
    },
    {
      "epoch": 3.8,
      "learning_rate": 0.0002,
      "loss": 0.0116,
      "step": 57
    },
    {
      "epoch": 3.87,
      "learning_rate": 0.0002,
      "loss": 0.0113,
      "step": 58
    },
    {
      "epoch": 3.93,
      "learning_rate": 0.0002,
      "loss": 0.0048,
      "step": 59
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.0002,
      "loss": 0.0079,
      "step": 60
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.006983411498367786,
      "eval_runtime": 40.4845,
      "eval_samples_per_second": 11.165,
      "eval_steps_per_second": 0.371,
      "step": 60
    },
    {
      "epoch": 4.07,
      "learning_rate": 0.0002,
      "loss": 0.0052,
      "step": 61
    },
    {
      "epoch": 4.13,
      "learning_rate": 0.0002,
      "loss": 0.0064,
      "step": 62
    },
    {
      "epoch": 4.2,
      "learning_rate": 0.0002,
      "loss": 0.0033,
      "step": 63
    },
    {
      "epoch": 4.27,
      "learning_rate": 0.0002,
      "loss": 0.0044,
      "step": 64
    },
    {
      "epoch": 4.33,
      "learning_rate": 0.0002,
      "loss": 0.0099,
      "step": 65
    },
    {
      "epoch": 4.4,
      "learning_rate": 0.0002,
      "loss": 0.0076,
      "step": 66
    },
    {
      "epoch": 4.47,
      "learning_rate": 0.0002,
      "loss": 0.0049,
      "step": 67
    },
    {
      "epoch": 4.53,
      "learning_rate": 0.0002,
      "loss": 0.0037,
      "step": 68
    },
    {
      "epoch": 4.6,
      "learning_rate": 0.0002,
      "loss": 0.0076,
      "step": 69
    },
    {
      "epoch": 4.67,
      "learning_rate": 0.0002,
      "loss": 0.0053,
      "step": 70
    },
    {
      "epoch": 4.73,
      "learning_rate": 0.0002,
      "loss": 0.0033,
      "step": 71
    },
    {
      "epoch": 4.8,
      "learning_rate": 0.0002,
      "loss": 0.005,
      "step": 72
    },
    {
      "epoch": 4.87,
      "learning_rate": 0.0002,
      "loss": 0.0034,
      "step": 73
    },
    {
      "epoch": 4.93,
      "learning_rate": 0.0002,
      "loss": 0.0044,
      "step": 74
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.0002,
      "loss": 0.0026,
      "step": 75
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.004540112800896168,
      "eval_runtime": 40.789,
      "eval_samples_per_second": 11.081,
      "eval_steps_per_second": 0.368,
      "step": 75
    },
    {
      "epoch": 5.07,
      "learning_rate": 0.0002,
      "loss": 0.0056,
      "step": 76
    },
    {
      "epoch": 5.13,
      "learning_rate": 0.0002,
      "loss": 0.0018,
      "step": 77
    },
    {
      "epoch": 5.2,
      "learning_rate": 0.0002,
      "loss": 0.003,
      "step": 78
    },
    {
      "epoch": 5.27,
      "learning_rate": 0.0002,
      "loss": 0.0018,
      "step": 79
    },
    {
      "epoch": 5.33,
      "learning_rate": 0.0002,
      "loss": 0.0042,
      "step": 80
    },
    {
      "epoch": 5.4,
      "learning_rate": 0.0002,
      "loss": 0.0053,
      "step": 81
    },
    {
      "epoch": 5.47,
      "learning_rate": 0.0002,
      "loss": 0.0025,
      "step": 82
    },
    {
      "epoch": 5.53,
      "learning_rate": 0.0002,
      "loss": 0.0036,
      "step": 83
    },
    {
      "epoch": 5.6,
      "learning_rate": 0.0002,
      "loss": 0.0032,
      "step": 84
    },
    {
      "epoch": 5.67,
      "learning_rate": 0.0002,
      "loss": 0.0017,
      "step": 85
    },
    {
      "epoch": 5.73,
      "learning_rate": 0.0002,
      "loss": 0.0034,
      "step": 86
    },
    {
      "epoch": 5.8,
      "learning_rate": 0.0002,
      "loss": 0.0016,
      "step": 87
    },
    {
      "epoch": 5.87,
      "learning_rate": 0.0002,
      "loss": 0.0039,
      "step": 88
    },
    {
      "epoch": 5.93,
      "learning_rate": 0.0002,
      "loss": 0.0037,
      "step": 89
    },
    {
      "epoch": 6.0,
      "learning_rate": 0.0002,
      "loss": 0.0019,
      "step": 90
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.002933455863967538,
      "eval_runtime": 40.6953,
      "eval_samples_per_second": 11.107,
      "eval_steps_per_second": 0.369,
      "step": 90
    },
    {
      "epoch": 6.07,
      "learning_rate": 0.0002,
      "loss": 0.002,
      "step": 91
    },
    {
      "epoch": 6.13,
      "learning_rate": 0.0002,
      "loss": 0.0024,
      "step": 92
    },
    {
      "epoch": 6.2,
      "learning_rate": 0.0002,
      "loss": 0.0037,
      "step": 93
    },
    {
      "epoch": 6.27,
      "learning_rate": 0.0002,
      "loss": 0.0011,
      "step": 94
    },
    {
      "epoch": 6.33,
      "learning_rate": 0.0002,
      "loss": 0.0016,
      "step": 95
    },
    {
      "epoch": 6.4,
      "learning_rate": 0.0002,
      "loss": 0.0028,
      "step": 96
    },
    {
      "epoch": 6.47,
      "learning_rate": 0.0002,
      "loss": 0.0013,
      "step": 97
    },
    {
      "epoch": 6.53,
      "learning_rate": 0.0002,
      "loss": 0.0014,
      "step": 98
    },
    {
      "epoch": 6.6,
      "learning_rate": 0.0002,
      "loss": 0.0011,
      "step": 99
    },
    {
      "epoch": 6.67,
      "learning_rate": 0.0002,
      "loss": 0.0008,
      "step": 100
    },
    {
      "epoch": 6.73,
      "learning_rate": 0.0002,
      "loss": 0.0006,
      "step": 101
    },
    {
      "epoch": 6.8,
      "learning_rate": 0.0002,
      "loss": 0.0014,
      "step": 102
    },
    {
      "epoch": 6.87,
      "learning_rate": 0.0002,
      "loss": 0.0016,
      "step": 103
    },
    {
      "epoch": 6.93,
      "learning_rate": 0.0002,
      "loss": 0.0016,
      "step": 104
    },
    {
      "epoch": 7.0,
      "learning_rate": 0.0002,
      "loss": 0.001,
      "step": 105
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.0021262257359921932,
      "eval_runtime": 40.8074,
      "eval_samples_per_second": 11.076,
      "eval_steps_per_second": 0.368,
      "step": 105
    },
    {
      "epoch": 7.07,
      "learning_rate": 0.0002,
      "loss": 0.0012,
      "step": 106
    },
    {
      "epoch": 7.13,
      "learning_rate": 0.0002,
      "loss": 0.0011,
      "step": 107
    },
    {
      "epoch": 7.2,
      "learning_rate": 0.0002,
      "loss": 0.0021,
      "step": 108
    },
    {
      "epoch": 7.27,
      "learning_rate": 0.0002,
      "loss": 0.0011,
      "step": 109
    },
    {
      "epoch": 7.33,
      "learning_rate": 0.0002,
      "loss": 0.0005,
      "step": 110
    },
    {
      "epoch": 7.4,
      "learning_rate": 0.0002,
      "loss": 0.001,
      "step": 111
    },
    {
      "epoch": 7.47,
      "learning_rate": 0.0002,
      "loss": 0.0033,
      "step": 112
    },
    {
      "epoch": 7.53,
      "learning_rate": 0.0002,
      "loss": 0.0005,
      "step": 113
    },
    {
      "epoch": 7.6,
      "learning_rate": 0.0002,
      "loss": 0.0012,
      "step": 114
    },
    {
      "epoch": 7.67,
      "learning_rate": 0.0002,
      "loss": 0.0023,
      "step": 115
    },
    {
      "epoch": 7.73,
      "learning_rate": 0.0002,
      "loss": 0.0007,
      "step": 116
    },
    {
      "epoch": 7.8,
      "learning_rate": 0.0002,
      "loss": 0.0018,
      "step": 117
    },
    {
      "epoch": 7.87,
      "learning_rate": 0.0002,
      "loss": 0.0007,
      "step": 118
    },
    {
      "epoch": 7.93,
      "learning_rate": 0.0002,
      "loss": 0.0012,
      "step": 119
    },
    {
      "epoch": 8.0,
      "learning_rate": 0.0002,
      "loss": 0.0016,
      "step": 120
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.0016470245318487287,
      "eval_runtime": 40.9122,
      "eval_samples_per_second": 11.048,
      "eval_steps_per_second": 0.367,
      "step": 120
    },
    {
      "epoch": 8.07,
      "learning_rate": 0.0002,
      "loss": 0.0003,
      "step": 121
    },
    {
      "epoch": 8.13,
      "learning_rate": 0.0002,
      "loss": 0.0007,
      "step": 122
    },
    {
      "epoch": 8.2,
      "learning_rate": 0.0002,
      "loss": 0.0044,
      "step": 123
    },
    {
      "epoch": 8.27,
      "learning_rate": 0.0002,
      "loss": 0.0033,
      "step": 124
    },
    {
      "epoch": 8.33,
      "learning_rate": 0.0002,
      "loss": 0.0012,
      "step": 125
    },
    {
      "epoch": 8.4,
      "learning_rate": 0.0002,
      "loss": 0.002,
      "step": 126
    },
    {
      "epoch": 8.47,
      "learning_rate": 0.0002,
      "loss": 0.0016,
      "step": 127
    },
    {
      "epoch": 8.53,
      "learning_rate": 0.0002,
      "loss": 0.0024,
      "step": 128
    },
    {
      "epoch": 8.6,
      "learning_rate": 0.0002,
      "loss": 0.0013,
      "step": 129
    },
    {
      "epoch": 8.67,
      "learning_rate": 0.0002,
      "loss": 0.0011,
      "step": 130
    },
    {
      "epoch": 8.73,
      "learning_rate": 0.0002,
      "loss": 0.0009,
      "step": 131
    },
    {
      "epoch": 8.8,
      "learning_rate": 0.0002,
      "loss": 0.0013,
      "step": 132
    },
    {
      "epoch": 8.87,
      "learning_rate": 0.0002,
      "loss": 0.0029,
      "step": 133
    },
    {
      "epoch": 8.93,
      "learning_rate": 0.0002,
      "loss": 0.0032,
      "step": 134
    },
    {
      "epoch": 9.0,
      "learning_rate": 0.0002,
      "loss": 0.002,
      "step": 135
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.0025528622791171074,
      "eval_runtime": 40.9816,
      "eval_samples_per_second": 11.029,
      "eval_steps_per_second": 0.366,
      "step": 135
    },
    {
      "epoch": 9.07,
      "learning_rate": 0.0002,
      "loss": 0.0015,
      "step": 136
    },
    {
      "epoch": 9.13,
      "learning_rate": 0.0002,
      "loss": 0.0017,
      "step": 137
    },
    {
      "epoch": 9.2,
      "learning_rate": 0.0002,
      "loss": 0.0017,
      "step": 138
    },
    {
      "epoch": 9.27,
      "learning_rate": 0.0002,
      "loss": 0.0008,
      "step": 139
    },
    {
      "epoch": 9.33,
      "learning_rate": 0.0002,
      "loss": 0.0007,
      "step": 140
    },
    {
      "epoch": 9.4,
      "learning_rate": 0.0002,
      "loss": 0.0009,
      "step": 141
    },
    {
      "epoch": 9.47,
      "learning_rate": 0.0002,
      "loss": 0.0008,
      "step": 142
    },
    {
      "epoch": 9.53,
      "learning_rate": 0.0002,
      "loss": 0.0012,
      "step": 143
    },
    {
      "epoch": 9.6,
      "learning_rate": 0.0002,
      "loss": 0.0033,
      "step": 144
    },
    {
      "epoch": 9.67,
      "learning_rate": 0.0002,
      "loss": 0.0005,
      "step": 145
    },
    {
      "epoch": 9.73,
      "learning_rate": 0.0002,
      "loss": 0.0006,
      "step": 146
    },
    {
      "epoch": 9.8,
      "learning_rate": 0.0002,
      "loss": 0.0012,
      "step": 147
    },
    {
      "epoch": 9.87,
      "learning_rate": 0.0002,
      "loss": 0.0014,
      "step": 148
    },
    {
      "epoch": 9.93,
      "learning_rate": 0.0002,
      "loss": 0.0021,
      "step": 149
    },
    {
      "epoch": 10.0,
      "learning_rate": 0.0002,
      "loss": 0.0021,
      "step": 150
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.0008361850050278008,
      "eval_runtime": 40.8576,
      "eval_samples_per_second": 11.063,
      "eval_steps_per_second": 0.367,
      "step": 150
    },
    {
      "epoch": 10.07,
      "learning_rate": 0.0002,
      "loss": 0.0002,
      "step": 151
    },
    {
      "epoch": 10.13,
      "learning_rate": 0.0002,
      "loss": 0.0011,
      "step": 152
    },
    {
      "epoch": 10.2,
      "learning_rate": 0.0002,
      "loss": 0.0003,
      "step": 153
    },
    {
      "epoch": 10.27,
      "learning_rate": 0.0002,
      "loss": 0.0012,
      "step": 154
    },
    {
      "epoch": 10.33,
      "learning_rate": 0.0002,
      "loss": 0.0006,
      "step": 155
    },
    {
      "epoch": 10.4,
      "learning_rate": 0.0002,
      "loss": 0.0017,
      "step": 156
    },
    {
      "epoch": 10.47,
      "learning_rate": 0.0002,
      "loss": 0.0002,
      "step": 157
    },
    {
      "epoch": 10.53,
      "learning_rate": 0.0002,
      "loss": 0.0013,
      "step": 158
    },
    {
      "epoch": 10.6,
      "learning_rate": 0.0002,
      "loss": 0.0002,
      "step": 159
    },
    {
      "epoch": 10.67,
      "learning_rate": 0.0002,
      "loss": 0.0003,
      "step": 160
    },
    {
      "epoch": 10.73,
      "learning_rate": 0.0002,
      "loss": 0.0005,
      "step": 161
    },
    {
      "epoch": 10.8,
      "learning_rate": 0.0002,
      "loss": 0.0002,
      "step": 162
    },
    {
      "epoch": 10.87,
      "learning_rate": 0.0002,
      "loss": 0.0003,
      "step": 163
    },
    {
      "epoch": 10.93,
      "learning_rate": 0.0002,
      "loss": 0.001,
      "step": 164
    },
    {
      "epoch": 11.0,
      "learning_rate": 0.0002,
      "loss": 0.0002,
      "step": 165
    },
    {
      "epoch": 11.0,
      "eval_loss": 0.0006615883321501315,
      "eval_runtime": 40.7704,
      "eval_samples_per_second": 11.086,
      "eval_steps_per_second": 0.368,
      "step": 165
    },
    {
      "epoch": 11.07,
      "learning_rate": 0.0002,
      "loss": 0.0001,
      "step": 166
    },
    {
      "epoch": 11.13,
      "learning_rate": 0.0002,
      "loss": 0.0003,
      "step": 167
    },
    {
      "epoch": 11.2,
      "learning_rate": 0.0002,
      "loss": 0.0003,
      "step": 168
    },
    {
      "epoch": 11.27,
      "learning_rate": 0.0002,
      "loss": 0.0004,
      "step": 169
    },
    {
      "epoch": 11.33,
      "learning_rate": 0.0002,
      "loss": 0.001,
      "step": 170
    },
    {
      "epoch": 11.4,
      "learning_rate": 0.0002,
      "loss": 0.0006,
      "step": 171
    },
    {
      "epoch": 11.47,
      "learning_rate": 0.0002,
      "loss": 0.0001,
      "step": 172
    },
    {
      "epoch": 11.53,
      "learning_rate": 0.0002,
      "loss": 0.0024,
      "step": 173
    },
    {
      "epoch": 11.6,
      "learning_rate": 0.0002,
      "loss": 0.0003,
      "step": 174
    },
    {
      "epoch": 11.67,
      "learning_rate": 0.0002,
      "loss": 0.0006,
      "step": 175
    },
    {
      "epoch": 11.73,
      "learning_rate": 0.0002,
      "loss": 0.0006,
      "step": 176
    },
    {
      "epoch": 11.8,
      "learning_rate": 0.0002,
      "loss": 0.003,
      "step": 177
    },
    {
      "epoch": 11.87,
      "learning_rate": 0.0002,
      "loss": 0.0004,
      "step": 178
    },
    {
      "epoch": 11.93,
      "learning_rate": 0.0002,
      "loss": 0.0016,
      "step": 179
    },
    {
      "epoch": 12.0,
      "learning_rate": 0.0002,
      "loss": 0.0001,
      "step": 180
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.002664078725501895,
      "eval_runtime": 40.8886,
      "eval_samples_per_second": 11.054,
      "eval_steps_per_second": 0.367,
      "step": 180
    },
    {
      "epoch": 12.07,
      "learning_rate": 0.0002,
      "loss": 0.006,
      "step": 181
    },
    {
      "epoch": 12.13,
      "learning_rate": 0.0002,
      "loss": 0.0021,
      "step": 182
    },
    {
      "epoch": 12.2,
      "learning_rate": 0.0002,
      "loss": 0.0019,
      "step": 183
    },
    {
      "epoch": 12.27,
      "learning_rate": 0.0002,
      "loss": 0.0007,
      "step": 184
    },
    {
      "epoch": 12.33,
      "learning_rate": 0.0002,
      "loss": 0.0024,
      "step": 185
    },
    {
      "epoch": 12.4,
      "learning_rate": 0.0002,
      "loss": 0.0002,
      "step": 186
    },
    {
      "epoch": 12.47,
      "learning_rate": 0.0002,
      "loss": 0.0008,
      "step": 187
    },
    {
      "epoch": 12.53,
      "learning_rate": 0.0002,
      "loss": 0.0005,
      "step": 188
    },
    {
      "epoch": 12.6,
      "learning_rate": 0.0002,
      "loss": 0.0012,
      "step": 189
    },
    {
      "epoch": 12.67,
      "learning_rate": 0.0002,
      "loss": 0.0009,
      "step": 190
    },
    {
      "epoch": 12.73,
      "learning_rate": 0.0002,
      "loss": 0.0011,
      "step": 191
    },
    {
      "epoch": 12.8,
      "learning_rate": 0.0002,
      "loss": 0.0046,
      "step": 192
    },
    {
      "epoch": 12.87,
      "learning_rate": 0.0002,
      "loss": 0.0026,
      "step": 193
    },
    {
      "epoch": 12.93,
      "learning_rate": 0.0002,
      "loss": 0.0019,
      "step": 194
    },
    {
      "epoch": 13.0,
      "learning_rate": 0.0002,
      "loss": 0.0003,
      "step": 195
    },
    {
      "epoch": 13.0,
      "eval_loss": 0.0012445432366803288,
      "eval_runtime": 40.8791,
      "eval_samples_per_second": 11.057,
      "eval_steps_per_second": 0.367,
      "step": 195
    },
    {
      "epoch": 13.07,
      "learning_rate": 0.0002,
      "loss": 0.0009,
      "step": 196
    },
    {
      "epoch": 13.13,
      "learning_rate": 0.0002,
      "loss": 0.0027,
      "step": 197
    },
    {
      "epoch": 13.2,
      "learning_rate": 0.0002,
      "loss": 0.0016,
      "step": 198
    },
    {
      "epoch": 13.27,
      "learning_rate": 0.0002,
      "loss": 0.0014,
      "step": 199
    },
    {
      "epoch": 13.33,
      "learning_rate": 0.0002,
      "loss": 0.0011,
      "step": 200
    },
    {
      "epoch": 13.4,
      "learning_rate": 0.0002,
      "loss": 0.0008,
      "step": 201
    },
    {
      "epoch": 13.47,
      "learning_rate": 0.0002,
      "loss": 0.0031,
      "step": 202
    },
    {
      "epoch": 13.53,
      "learning_rate": 0.0002,
      "loss": 0.0006,
      "step": 203
    },
    {
      "epoch": 13.6,
      "learning_rate": 0.0002,
      "loss": 0.0011,
      "step": 204
    },
    {
      "epoch": 13.67,
      "learning_rate": 0.0002,
      "loss": 0.0005,
      "step": 205
    },
    {
      "epoch": 13.73,
      "learning_rate": 0.0002,
      "loss": 0.002,
      "step": 206
    },
    {
      "epoch": 13.8,
      "learning_rate": 0.0002,
      "loss": 0.0023,
      "step": 207
    },
    {
      "epoch": 13.87,
      "learning_rate": 0.0002,
      "loss": 0.0006,
      "step": 208
    },
    {
      "epoch": 13.93,
      "learning_rate": 0.0002,
      "loss": 0.0027,
      "step": 209
    },
    {
      "epoch": 14.0,
      "learning_rate": 0.0002,
      "loss": 0.0017,
      "step": 210
    },
    {
      "epoch": 14.0,
      "eval_loss": 0.000952349801082164,
      "eval_runtime": 40.9836,
      "eval_samples_per_second": 11.029,
      "eval_steps_per_second": 0.366,
      "step": 210
    },
    {
      "epoch": 14.07,
      "learning_rate": 0.0002,
      "loss": 0.0027,
      "step": 211
    },
    {
      "epoch": 14.13,
      "learning_rate": 0.0002,
      "loss": 0.0003,
      "step": 212
    },
    {
      "epoch": 14.2,
      "learning_rate": 0.0002,
      "loss": 0.0012,
      "step": 213
    },
    {
      "epoch": 14.27,
      "learning_rate": 0.0002,
      "loss": 0.0009,
      "step": 214
    },
    {
      "epoch": 14.33,
      "learning_rate": 0.0002,
      "loss": 0.0009,
      "step": 215
    },
    {
      "epoch": 14.4,
      "learning_rate": 0.0002,
      "loss": 0.0026,
      "step": 216
    },
    {
      "epoch": 14.47,
      "learning_rate": 0.0002,
      "loss": 0.0039,
      "step": 217
    },
    {
      "epoch": 14.53,
      "learning_rate": 0.0002,
      "loss": 0.0046,
      "step": 218
    },
    {
      "epoch": 14.6,
      "learning_rate": 0.0002,
      "loss": 0.001,
      "step": 219
    },
    {
      "epoch": 14.67,
      "learning_rate": 0.0002,
      "loss": 0.0033,
      "step": 220
    },
    {
      "epoch": 14.73,
      "learning_rate": 0.0002,
      "loss": 0.0007,
      "step": 221
    },
    {
      "epoch": 14.8,
      "learning_rate": 0.0002,
      "loss": 0.0019,
      "step": 222
    },
    {
      "epoch": 14.87,
      "learning_rate": 0.0002,
      "loss": 0.003,
      "step": 223
    },
    {
      "epoch": 14.93,
      "learning_rate": 0.0002,
      "loss": 0.0025,
      "step": 224
    },
    {
      "epoch": 15.0,
      "learning_rate": 0.0002,
      "loss": 0.005,
      "step": 225
    },
    {
      "epoch": 15.0,
      "eval_loss": 0.002726432401686907,
      "eval_runtime": 41.0215,
      "eval_samples_per_second": 11.019,
      "eval_steps_per_second": 0.366,
      "step": 225
    },
    {
      "epoch": 15.07,
      "learning_rate": 0.0002,
      "loss": 0.0023,
      "step": 226
    },
    {
      "epoch": 15.13,
      "learning_rate": 0.0002,
      "loss": 0.0017,
      "step": 227
    },
    {
      "epoch": 15.2,
      "learning_rate": 0.0002,
      "loss": 0.0017,
      "step": 228
    },
    {
      "epoch": 15.27,
      "learning_rate": 0.0002,
      "loss": 0.0021,
      "step": 229
    },
    {
      "epoch": 15.33,
      "learning_rate": 0.0002,
      "loss": 0.0056,
      "step": 230
    },
    {
      "epoch": 15.4,
      "learning_rate": 0.0002,
      "loss": 0.0028,
      "step": 231
    },
    {
      "epoch": 15.47,
      "learning_rate": 0.0002,
      "loss": 0.0017,
      "step": 232
    },
    {
      "epoch": 15.53,
      "learning_rate": 0.0002,
      "loss": 0.001,
      "step": 233
    },
    {
      "epoch": 15.6,
      "learning_rate": 0.0002,
      "loss": 0.002,
      "step": 234
    },
    {
      "epoch": 15.67,
      "learning_rate": 0.0002,
      "loss": 0.0025,
      "step": 235
    },
    {
      "epoch": 15.73,
      "learning_rate": 0.0002,
      "loss": 0.0012,
      "step": 236
    },
    {
      "epoch": 15.8,
      "learning_rate": 0.0002,
      "loss": 0.0029,
      "step": 237
    },
    {
      "epoch": 15.87,
      "learning_rate": 0.0002,
      "loss": 0.0052,
      "step": 238
    },
    {
      "epoch": 15.93,
      "learning_rate": 0.0002,
      "loss": 0.0018,
      "step": 239
    },
    {
      "epoch": 16.0,
      "learning_rate": 0.0002,
      "loss": 0.0059,
      "step": 240
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.00219073542393744,
      "eval_runtime": 41.3556,
      "eval_samples_per_second": 10.93,
      "eval_steps_per_second": 0.363,
      "step": 240
    },
    {
      "epoch": 16.07,
      "learning_rate": 0.0002,
      "loss": 0.0018,
      "step": 241
    },
    {
      "epoch": 16.13,
      "learning_rate": 0.0002,
      "loss": 0.0022,
      "step": 242
    },
    {
      "epoch": 16.2,
      "learning_rate": 0.0002,
      "loss": 0.0013,
      "step": 243
    },
    {
      "epoch": 16.27,
      "learning_rate": 0.0002,
      "loss": 0.0027,
      "step": 244
    },
    {
      "epoch": 16.33,
      "learning_rate": 0.0002,
      "loss": 0.0072,
      "step": 245
    },
    {
      "epoch": 16.4,
      "learning_rate": 0.0002,
      "loss": 0.0022,
      "step": 246
    },
    {
      "epoch": 16.47,
      "learning_rate": 0.0002,
      "loss": 0.0018,
      "step": 247
    },
    {
      "epoch": 16.53,
      "learning_rate": 0.0002,
      "loss": 0.003,
      "step": 248
    },
    {
      "epoch": 16.6,
      "learning_rate": 0.0002,
      "loss": 0.0017,
      "step": 249
    },
    {
      "epoch": 16.67,
      "learning_rate": 0.0002,
      "loss": 0.0016,
      "step": 250
    },
    {
      "epoch": 16.73,
      "learning_rate": 0.0002,
      "loss": 0.0038,
      "step": 251
    },
    {
      "epoch": 16.8,
      "learning_rate": 0.0002,
      "loss": 0.002,
      "step": 252
    },
    {
      "epoch": 16.87,
      "learning_rate": 0.0002,
      "loss": 0.0013,
      "step": 253
    },
    {
      "epoch": 16.93,
      "learning_rate": 0.0002,
      "loss": 0.0036,
      "step": 254
    },
    {
      "epoch": 17.0,
      "learning_rate": 0.0002,
      "loss": 0.0047,
      "step": 255
    },
    {
      "epoch": 17.0,
      "eval_loss": 0.0022247186861932278,
      "eval_runtime": 40.9693,
      "eval_samples_per_second": 11.033,
      "eval_steps_per_second": 0.366,
      "step": 255
    },
    {
      "epoch": 17.07,
      "learning_rate": 0.0002,
      "loss": 0.0002,
      "step": 256
    },
    {
      "epoch": 17.13,
      "learning_rate": 0.0002,
      "loss": 0.0018,
      "step": 257
    },
    {
      "epoch": 17.2,
      "learning_rate": 0.0002,
      "loss": 0.0017,
      "step": 258
    },
    {
      "epoch": 17.27,
      "learning_rate": 0.0002,
      "loss": 0.0024,
      "step": 259
    },
    {
      "epoch": 17.33,
      "learning_rate": 0.0002,
      "loss": 0.0005,
      "step": 260
    },
    {
      "epoch": 17.4,
      "learning_rate": 0.0002,
      "loss": 0.0014,
      "step": 261
    },
    {
      "epoch": 17.47,
      "learning_rate": 0.0002,
      "loss": 0.0025,
      "step": 262
    },
    {
      "epoch": 17.53,
      "learning_rate": 0.0002,
      "loss": 0.0049,
      "step": 263
    },
    {
      "epoch": 17.6,
      "learning_rate": 0.0002,
      "loss": 0.003,
      "step": 264
    },
    {
      "epoch": 17.67,
      "learning_rate": 0.0002,
      "loss": 0.0026,
      "step": 265
    },
    {
      "epoch": 17.73,
      "learning_rate": 0.0002,
      "loss": 0.0013,
      "step": 266
    },
    {
      "epoch": 17.8,
      "learning_rate": 0.0002,
      "loss": 0.0024,
      "step": 267
    },
    {
      "epoch": 17.87,
      "learning_rate": 0.0002,
      "loss": 0.0007,
      "step": 268
    },
    {
      "epoch": 17.93,
      "learning_rate": 0.0002,
      "loss": 0.0012,
      "step": 269
    },
    {
      "epoch": 18.0,
      "learning_rate": 0.0002,
      "loss": 0.0009,
      "step": 270
    },
    {
      "epoch": 18.0,
      "eval_loss": 0.0014401717344298959,
      "eval_runtime": 41.0768,
      "eval_samples_per_second": 11.004,
      "eval_steps_per_second": 0.365,
      "step": 270
    },
    {
      "epoch": 18.07,
      "learning_rate": 0.0002,
      "loss": 0.0018,
      "step": 271
    },
    {
      "epoch": 18.13,
      "learning_rate": 0.0002,
      "loss": 0.002,
      "step": 272
    },
    {
      "epoch": 18.2,
      "learning_rate": 0.0002,
      "loss": 0.0032,
      "step": 273
    },
    {
      "epoch": 18.27,
      "learning_rate": 0.0002,
      "loss": 0.0014,
      "step": 274
    },
    {
      "epoch": 18.33,
      "learning_rate": 0.0002,
      "loss": 0.0012,
      "step": 275
    },
    {
      "epoch": 18.4,
      "learning_rate": 0.0002,
      "loss": 0.0013,
      "step": 276
    },
    {
      "epoch": 18.47,
      "learning_rate": 0.0002,
      "loss": 0.0004,
      "step": 277
    },
    {
      "epoch": 18.53,
      "learning_rate": 0.0002,
      "loss": 0.0025,
      "step": 278
    },
    {
      "epoch": 18.6,
      "learning_rate": 0.0002,
      "loss": 0.0037,
      "step": 279
    },
    {
      "epoch": 18.67,
      "learning_rate": 0.0002,
      "loss": 0.0023,
      "step": 280
    },
    {
      "epoch": 18.73,
      "learning_rate": 0.0002,
      "loss": 0.0012,
      "step": 281
    },
    {
      "epoch": 18.8,
      "learning_rate": 0.0002,
      "loss": 0.0031,
      "step": 282
    },
    {
      "epoch": 18.87,
      "learning_rate": 0.0002,
      "loss": 0.0015,
      "step": 283
    },
    {
      "epoch": 18.93,
      "learning_rate": 0.0002,
      "loss": 0.0024,
      "step": 284
    },
    {
      "epoch": 19.0,
      "learning_rate": 0.0002,
      "loss": 0.0046,
      "step": 285
    },
    {
      "epoch": 19.0,
      "eval_loss": 0.0023026454728096724,
      "eval_runtime": 40.8292,
      "eval_samples_per_second": 11.071,
      "eval_steps_per_second": 0.367,
      "step": 285
    },
    {
      "epoch": 19.07,
      "learning_rate": 0.0002,
      "loss": 0.0007,
      "step": 286
    },
    {
      "epoch": 19.13,
      "learning_rate": 0.0002,
      "loss": 0.0019,
      "step": 287
    },
    {
      "epoch": 19.2,
      "learning_rate": 0.0002,
      "loss": 0.0042,
      "step": 288
    },
    {
      "epoch": 19.27,
      "learning_rate": 0.0002,
      "loss": 0.0016,
      "step": 289
    },
    {
      "epoch": 19.33,
      "learning_rate": 0.0002,
      "loss": 0.0022,
      "step": 290
    },
    {
      "epoch": 19.4,
      "learning_rate": 0.0002,
      "loss": 0.0018,
      "step": 291
    },
    {
      "epoch": 19.47,
      "learning_rate": 0.0002,
      "loss": 0.002,
      "step": 292
    },
    {
      "epoch": 19.53,
      "learning_rate": 0.0002,
      "loss": 0.0022,
      "step": 293
    },
    {
      "epoch": 19.6,
      "learning_rate": 0.0002,
      "loss": 0.0015,
      "step": 294
    },
    {
      "epoch": 19.67,
      "learning_rate": 0.0002,
      "loss": 0.0024,
      "step": 295
    },
    {
      "epoch": 19.73,
      "learning_rate": 0.0002,
      "loss": 0.0058,
      "step": 296
    },
    {
      "epoch": 19.8,
      "learning_rate": 0.0002,
      "loss": 0.008,
      "step": 297
    },
    {
      "epoch": 19.87,
      "learning_rate": 0.0002,
      "loss": 0.0054,
      "step": 298
    },
    {
      "epoch": 19.93,
      "learning_rate": 0.0002,
      "loss": 0.0032,
      "step": 299
    },
    {
      "epoch": 20.0,
      "learning_rate": 0.0002,
      "loss": 0.006,
      "step": 300
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.003988527227193117,
      "eval_runtime": 40.9101,
      "eval_samples_per_second": 11.049,
      "eval_steps_per_second": 0.367,
      "step": 300
    },
    {
      "epoch": 20.0,
      "step": 300,
      "total_flos": 1.1369758627263939e+18,
      "train_loss": 0.061939337516426655,
      "train_runtime": 6353.7763,
      "train_samples_per_second": 1.423,
      "train_steps_per_second": 0.047
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 300,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 50000,
  "total_flos": 1.1369758627263939e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}