{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 20,
  "global_step": 900,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 5e-06,
      "loss": 0.707,
      "step": 1
    },
    {
      "epoch": 0.01,
      "learning_rate": 1e-05,
      "loss": 0.7377,
      "step": 2
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.5e-05,
      "loss": 0.7171,
      "step": 3
    },
    {
      "epoch": 0.02,
      "learning_rate": 2e-05,
      "loss": 0.7164,
      "step": 4
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.5e-05,
      "loss": 0.7331,
      "step": 5
    },
    {
      "epoch": 0.03,
      "learning_rate": 3e-05,
      "loss": 0.6827,
      "step": 6
    },
    {
      "epoch": 0.03,
      "learning_rate": 3.5e-05,
      "loss": 0.6917,
      "step": 7
    },
    {
      "epoch": 0.04,
      "learning_rate": 4e-05,
      "loss": 0.7588,
      "step": 8
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.5e-05,
      "loss": 0.7018,
      "step": 9
    },
    {
      "epoch": 0.04,
      "learning_rate": 5e-05,
      "loss": 0.6927,
      "step": 10
    },
    {
      "epoch": 0.05,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.7084,
      "step": 11
    },
    {
      "epoch": 0.05,
      "learning_rate": 6e-05,
      "loss": 0.6788,
      "step": 12
    },
    {
      "epoch": 0.06,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.7172,
      "step": 13
    },
    {
      "epoch": 0.06,
      "learning_rate": 7e-05,
      "loss": 0.6258,
      "step": 14
    },
    {
      "epoch": 0.07,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.642,
      "step": 15
    },
    {
      "epoch": 0.07,
      "learning_rate": 8e-05,
      "loss": 0.7295,
      "step": 16
    },
    {
      "epoch": 0.08,
      "learning_rate": 8.5e-05,
      "loss": 0.6746,
      "step": 17
    },
    {
      "epoch": 0.08,
      "learning_rate": 9e-05,
      "loss": 0.6241,
      "step": 18
    },
    {
      "epoch": 0.08,
      "learning_rate": 9.5e-05,
      "loss": 0.6261,
      "step": 19
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001,
      "loss": 0.6297,
      "step": 20
    },
    {
      "epoch": 0.09,
      "eval_loss": 0.6454461216926575,
      "eval_runtime": 23.9286,
      "eval_samples_per_second": 3.385,
      "eval_steps_per_second": 0.878,
      "step": 20
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001,
      "loss": 0.6199,
      "step": 21
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001,
      "loss": 0.606,
      "step": 22
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001,
      "loss": 0.7053,
      "step": 23
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0001,
      "loss": 0.6349,
      "step": 24
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0001,
      "loss": 0.63,
      "step": 25
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001,
      "loss": 0.6298,
      "step": 26
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001,
      "loss": 0.6377,
      "step": 27
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001,
      "loss": 0.6523,
      "step": 28
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0001,
      "loss": 0.6683,
      "step": 29
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0001,
      "loss": 0.617,
      "step": 30
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001,
      "loss": 0.63,
      "step": 31
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001,
      "loss": 0.6244,
      "step": 32
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0001,
      "loss": 0.6353,
      "step": 33
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0001,
      "loss": 0.6369,
      "step": 34
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001,
      "loss": 0.6304,
      "step": 35
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001,
      "loss": 0.653,
      "step": 36
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001,
      "loss": 0.5831,
      "step": 37
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001,
      "loss": 0.6009,
      "step": 38
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001,
      "loss": 0.6477,
      "step": 39
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001,
      "loss": 0.5946,
      "step": 40
    },
    {
      "epoch": 0.18,
      "eval_loss": 0.616696834564209,
      "eval_runtime": 24.041,
      "eval_samples_per_second": 3.369,
      "eval_steps_per_second": 0.874,
      "step": 40
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001,
      "loss": 0.6032,
      "step": 41
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001,
      "loss": 0.6014,
      "step": 42
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001,
      "loss": 0.6541,
      "step": 43
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001,
      "loss": 0.6109,
      "step": 44
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001,
      "loss": 0.5991,
      "step": 45
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001,
      "loss": 0.6385,
      "step": 46
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001,
      "loss": 0.6642,
      "step": 47
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001,
      "loss": 0.6306,
      "step": 48
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001,
      "loss": 0.6166,
      "step": 49
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001,
      "loss": 0.6388,
      "step": 50
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001,
      "loss": 0.606,
      "step": 51
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001,
      "loss": 0.6236,
      "step": 52
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001,
      "loss": 0.6243,
      "step": 53
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001,
      "loss": 0.5816,
      "step": 54
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001,
      "loss": 0.5754,
      "step": 55
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001,
      "loss": 0.6143,
      "step": 56
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001,
      "loss": 0.6579,
      "step": 57
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001,
      "loss": 0.6179,
      "step": 58
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001,
      "loss": 0.5915,
      "step": 59
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001,
      "loss": 0.633,
      "step": 60
    },
    {
      "epoch": 0.27,
      "eval_loss": 0.6058350801467896,
      "eval_runtime": 24.0841,
      "eval_samples_per_second": 3.363,
      "eval_steps_per_second": 0.872,
      "step": 60
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001,
      "loss": 0.5831,
      "step": 61
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001,
      "loss": 0.6112,
      "step": 62
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001,
      "loss": 0.5624,
      "step": 63
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001,
      "loss": 0.5964,
      "step": 64
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0001,
      "loss": 0.5906,
      "step": 65
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0001,
      "loss": 0.6131,
      "step": 66
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001,
      "loss": 0.5693,
      "step": 67
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001,
      "loss": 0.5907,
      "step": 68
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001,
      "loss": 0.5742,
      "step": 69
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001,
      "loss": 0.5946,
      "step": 70
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001,
      "loss": 0.6416,
      "step": 71
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001,
      "loss": 0.6034,
      "step": 72
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001,
      "loss": 0.6602,
      "step": 73
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001,
      "loss": 0.6289,
      "step": 74
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001,
      "loss": 0.6046,
      "step": 75
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001,
      "loss": 0.6433,
      "step": 76
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001,
      "loss": 0.5421,
      "step": 77
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001,
      "loss": 0.5756,
      "step": 78
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001,
      "loss": 0.6497,
      "step": 79
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0001,
      "loss": 0.5517,
      "step": 80
    },
    {
      "epoch": 0.36,
      "eval_loss": 0.5980130434036255,
      "eval_runtime": 24.0895,
      "eval_samples_per_second": 3.362,
      "eval_steps_per_second": 0.872,
      "step": 80
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0001,
      "loss": 0.5973,
      "step": 81
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0001,
      "loss": 0.5977,
      "step": 82
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0001,
      "loss": 0.5596,
      "step": 83
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0001,
      "loss": 0.6303,
      "step": 84
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0001,
      "loss": 0.6426,
      "step": 85
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0001,
      "loss": 0.6363,
      "step": 86
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001,
      "loss": 0.6225,
      "step": 87
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001,
      "loss": 0.5504,
      "step": 88
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001,
      "loss": 0.6367,
      "step": 89
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001,
      "loss": 0.5936,
      "step": 90
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001,
      "loss": 0.6353,
      "step": 91
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001,
      "loss": 0.6285,
      "step": 92
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001,
      "loss": 0.5725,
      "step": 93
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001,
      "loss": 0.5779,
      "step": 94
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001,
      "loss": 0.5878,
      "step": 95
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001,
      "loss": 0.5768,
      "step": 96
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001,
      "loss": 0.5883,
      "step": 97
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001,
      "loss": 0.635,
      "step": 98
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001,
      "loss": 0.5835,
      "step": 99
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001,
      "loss": 0.6494,
      "step": 100
    },
    {
      "epoch": 0.44,
      "eval_loss": 0.5949223041534424,
      "eval_runtime": 24.1087,
      "eval_samples_per_second": 3.36,
      "eval_steps_per_second": 0.871,
      "step": 100
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001,
      "loss": 0.5738,
      "step": 101
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001,
      "loss": 0.5712,
      "step": 102
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001,
      "loss": 0.6158,
      "step": 103
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001,
      "loss": 0.5806,
      "step": 104
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001,
      "loss": 0.5799,
      "step": 105
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001,
      "loss": 0.5511,
      "step": 106
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001,
      "loss": 0.6125,
      "step": 107
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001,
      "loss": 0.5889,
      "step": 108
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001,
      "loss": 0.5958,
      "step": 109
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001,
      "loss": 0.6386,
      "step": 110
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001,
      "loss": 0.5925,
      "step": 111
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001,
      "loss": 0.5508,
      "step": 112
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001,
      "loss": 0.6092,
      "step": 113
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001,
      "loss": 0.5813,
      "step": 114
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001,
      "loss": 0.6543,
      "step": 115
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001,
      "loss": 0.5724,
      "step": 116
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001,
      "loss": 0.5299,
      "step": 117
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001,
      "loss": 0.5967,
      "step": 118
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001,
      "loss": 0.5471,
      "step": 119
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001,
      "loss": 0.5658,
      "step": 120
    },
    {
      "epoch": 0.53,
      "eval_loss": 0.5895005464553833,
      "eval_runtime": 24.0965,
      "eval_samples_per_second": 3.361,
      "eval_steps_per_second": 0.871,
      "step": 120
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001,
      "loss": 0.5953,
      "step": 121
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001,
      "loss": 0.5799,
      "step": 122
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001,
      "loss": 0.6187,
      "step": 123
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001,
      "loss": 0.5454,
      "step": 124
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001,
      "loss": 0.6046,
      "step": 125
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001,
      "loss": 0.5176,
      "step": 126
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001,
      "loss": 0.5508,
      "step": 127
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001,
      "loss": 0.5521,
      "step": 128
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001,
      "loss": 0.631,
      "step": 129
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001,
      "loss": 0.5852,
      "step": 130
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001,
      "loss": 0.5581,
      "step": 131
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001,
      "loss": 0.5376,
      "step": 132
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001,
      "loss": 0.5432,
      "step": 133
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001,
      "loss": 0.6059,
      "step": 134
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001,
      "loss": 0.5585,
      "step": 135
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001,
      "loss": 0.5846,
      "step": 136
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001,
      "loss": 0.6416,
      "step": 137
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001,
      "loss": 0.608,
      "step": 138
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001,
      "loss": 0.6145,
      "step": 139
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001,
      "loss": 0.5747,
      "step": 140
    },
    {
      "epoch": 0.62,
      "eval_loss": 0.5848782658576965,
      "eval_runtime": 24.0956,
      "eval_samples_per_second": 3.362,
      "eval_steps_per_second": 0.872,
      "step": 140
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001,
      "loss": 0.5859,
      "step": 141
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001,
      "loss": 0.6319,
      "step": 142
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001,
      "loss": 0.6343,
      "step": 143
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001,
      "loss": 0.6236,
      "step": 144
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001,
      "loss": 0.5492,
      "step": 145
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0001,
      "loss": 0.6406,
      "step": 146
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0001,
      "loss": 0.6086,
      "step": 147
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001,
      "loss": 0.5028,
      "step": 148
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001,
      "loss": 0.5919,
      "step": 149
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001,
      "loss": 0.5852,
      "step": 150
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001,
      "loss": 0.5528,
      "step": 151
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0001,
      "loss": 0.5867,
      "step": 152
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0001,
      "loss": 0.6036,
      "step": 153
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0001,
      "loss": 0.6241,
      "step": 154
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0001,
      "loss": 0.5577,
      "step": 155
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0001,
      "loss": 0.5495,
      "step": 156
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001,
      "loss": 0.5771,
      "step": 157
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001,
      "loss": 0.6011,
      "step": 158
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0001,
      "loss": 0.5946,
      "step": 159
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0001,
      "loss": 0.577,
      "step": 160
    },
    {
      "epoch": 0.71,
      "eval_loss": 0.5818573832511902,
      "eval_runtime": 24.1101,
      "eval_samples_per_second": 3.36,
      "eval_steps_per_second": 0.871,
      "step": 160
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001,
      "loss": 0.6091,
      "step": 161
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001,
      "loss": 0.5716,
      "step": 162
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001,
      "loss": 0.5262,
      "step": 163
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0001,
      "loss": 0.5675,
      "step": 164
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0001,
      "loss": 0.615,
      "step": 165
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001,
      "loss": 0.5692,
      "step": 166
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001,
      "loss": 0.5257,
      "step": 167
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001,
      "loss": 0.5362,
      "step": 168
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001,
      "loss": 0.5903,
      "step": 169
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001,
      "loss": 0.5348,
      "step": 170
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001,
      "loss": 0.5976,
      "step": 171
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001,
      "loss": 0.5554,
      "step": 172
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001,
      "loss": 0.5314,
      "step": 173
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001,
      "loss": 0.6163,
      "step": 174
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 0.5713,
      "step": 175
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 0.5758,
      "step": 176
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001,
      "loss": 0.6182,
      "step": 177
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001,
      "loss": 0.6424,
      "step": 178
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001,
      "loss": 0.6469,
      "step": 179
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001,
      "loss": 0.559,
      "step": 180
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.5775895714759827,
      "eval_runtime": 24.0913,
      "eval_samples_per_second": 3.362,
      "eval_steps_per_second": 0.872,
      "step": 180
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001,
      "loss": 0.5665,
      "step": 181
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001,
      "loss": 0.6208,
      "step": 182
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001,
      "loss": 0.5895,
      "step": 183
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001,
      "loss": 0.5629,
      "step": 184
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001,
      "loss": 0.5419,
      "step": 185
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0001,
      "loss": 0.6083,
      "step": 186
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0001,
      "loss": 0.5867,
      "step": 187
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001,
      "loss": 0.5684,
      "step": 188
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001,
      "loss": 0.5779,
      "step": 189
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001,
      "loss": 0.5727,
      "step": 190
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0001,
      "loss": 0.5735,
      "step": 191
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0001,
      "loss": 0.6164,
      "step": 192
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001,
      "loss": 0.5378,
      "step": 193
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001,
      "loss": 0.5949,
      "step": 194
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001,
      "loss": 0.533,
      "step": 195
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001,
      "loss": 0.5617,
      "step": 196
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0001,
      "loss": 0.621,
      "step": 197
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0001,
      "loss": 0.621,
      "step": 198
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0001,
      "loss": 0.5388,
      "step": 199
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001,
      "loss": 0.6011,
      "step": 200
    },
    {
      "epoch": 0.89,
      "eval_loss": 0.5767409205436707,
      "eval_runtime": 24.1004,
      "eval_samples_per_second": 3.361,
      "eval_steps_per_second": 0.871,
      "step": 200
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001,
      "loss": 0.6009,
      "step": 201
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001,
      "loss": 0.5704,
      "step": 202
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001,
      "loss": 0.6318,
      "step": 203
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0001,
      "loss": 0.5546,
      "step": 204
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0001,
      "loss": 0.627,
      "step": 205
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0001,
      "loss": 0.5927,
      "step": 206
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0001,
      "loss": 0.5916,
      "step": 207
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0001,
      "loss": 0.5127,
      "step": 208
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001,
      "loss": 0.5381,
      "step": 209
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001,
      "loss": 0.5501,
      "step": 210
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001,
      "loss": 0.5516,
      "step": 211
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001,
      "loss": 0.5264,
      "step": 212
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0001,
      "loss": 0.5556,
      "step": 213
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0001,
      "loss": 0.5596,
      "step": 214
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001,
      "loss": 0.5618,
      "step": 215
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001,
      "loss": 0.5864,
      "step": 216
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001,
      "loss": 0.6083,
      "step": 217
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001,
      "loss": 0.582,
      "step": 218
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001,
      "loss": 0.5552,
      "step": 219
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0001,
      "loss": 0.5398,
      "step": 220
    },
    {
      "epoch": 0.98,
      "eval_loss": 0.5724539756774902,
      "eval_runtime": 24.1089,
      "eval_samples_per_second": 3.36,
      "eval_steps_per_second": 0.871,
      "step": 220
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0001,
      "loss": 0.5783,
      "step": 221
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0001,
      "loss": 0.6203,
      "step": 222
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0001,
      "loss": 0.6155,
      "step": 223
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001,
      "loss": 0.5743,
      "step": 224
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001,
      "loss": 0.6386,
      "step": 225
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001,
      "loss": 0.5781,
      "step": 226
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0001,
      "loss": 0.5641,
      "step": 227
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0001,
      "loss": 0.5653,
      "step": 228
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0001,
      "loss": 0.4956,
      "step": 229
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0001,
      "loss": 0.594,
      "step": 230
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0001,
      "loss": 0.5114,
      "step": 231
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0001,
      "loss": 0.5823,
      "step": 232
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0001,
      "loss": 0.5237,
      "step": 233
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0001,
      "loss": 0.6363,
      "step": 234
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0001,
      "loss": 0.569,
      "step": 235
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0001,
      "loss": 0.5643,
      "step": 236
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0001,
      "loss": 0.5548,
      "step": 237
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0001,
      "loss": 0.531,
      "step": 238
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0001,
      "loss": 0.5728,
      "step": 239
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001,
      "loss": 0.5149,
      "step": 240
    },
    {
      "epoch": 1.07,
      "eval_loss": 0.5700140595436096,
      "eval_runtime": 24.104,
      "eval_samples_per_second": 3.36,
      "eval_steps_per_second": 0.871,
      "step": 240
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001,
      "loss": 0.4867,
      "step": 241
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0001,
      "loss": 0.5643,
      "step": 242
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0001,
      "loss": 0.5037,
      "step": 243
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0001,
      "loss": 0.5028,
      "step": 244
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001,
      "loss": 0.5113,
      "step": 245
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001,
      "loss": 0.5141,
      "step": 246
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001,
      "loss": 0.51,
      "step": 247
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001,
      "loss": 0.563,
      "step": 248
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001,
      "loss": 0.5623,
      "step": 249
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001,
      "loss": 0.4754,
      "step": 250
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001,
      "loss": 0.4944,
      "step": 251
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001,
      "loss": 0.5417,
      "step": 252
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001,
      "loss": 0.5171,
      "step": 253
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0001,
      "loss": 0.569,
      "step": 254
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0001,
      "loss": 0.4875,
      "step": 255
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001,
      "loss": 0.5378,
      "step": 256
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001,
      "loss": 0.5453,
      "step": 257
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0001,
      "loss": 0.5139,
      "step": 258
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0001,
      "loss": 0.5286,
      "step": 259
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0001,
      "loss": 0.5104,
      "step": 260
    },
    {
      "epoch": 1.16,
      "eval_loss": 0.5690858364105225,
      "eval_runtime": 24.1042,
      "eval_samples_per_second": 3.36,
      "eval_steps_per_second": 0.871,
      "step": 260
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0001,
      "loss": 0.5725,
      "step": 261
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0001,
      "loss": 0.5731,
      "step": 262
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001,
      "loss": 0.5165,
      "step": 263
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001,
      "loss": 0.521,
      "step": 264
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0001,
      "loss": 0.5396,
      "step": 265
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0001,
      "loss": 0.5059,
      "step": 266
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0001,
      "loss": 0.4909,
      "step": 267
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0001,
      "loss": 0.5054,
      "step": 268
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001,
      "loss": 0.5542,
      "step": 269
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001,
      "loss": 0.5238,
      "step": 270
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001,
      "loss": 0.4888,
      "step": 271
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0001,
      "loss": 0.6221,
      "step": 272
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0001,
      "loss": 0.4888,
      "step": 273
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001,
      "loss": 0.5156,
      "step": 274
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001,
      "loss": 0.5625,
      "step": 275
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001,
      "loss": 0.5728,
      "step": 276
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001,
      "loss": 0.5524,
      "step": 277
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001,
      "loss": 0.5336,
      "step": 278
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001,
      "loss": 0.5212,
      "step": 279
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001,
      "loss": 0.5319,
      "step": 280
    },
    {
      "epoch": 1.24,
      "eval_loss": 0.5698450207710266,
      "eval_runtime": 24.1024,
      "eval_samples_per_second": 3.361,
      "eval_steps_per_second": 0.871,
      "step": 280
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0001,
      "loss": 0.593,
      "step": 281
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0001,
      "loss": 0.503,
      "step": 282
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001,
      "loss": 0.5921,
      "step": 283
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001,
      "loss": 0.6451,
      "step": 284
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001,
      "loss": 0.6535,
      "step": 285
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001,
      "loss": 0.5404,
      "step": 286
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001,
      "loss": 0.4991,
      "step": 287
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001,
      "loss": 0.5003,
      "step": 288
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001,
      "loss": 0.563,
      "step": 289
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0001,
      "loss": 0.6135,
      "step": 290
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0001,
      "loss": 0.49,
      "step": 291
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001,
      "loss": 0.5275,
      "step": 292
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001,
      "loss": 0.599,
      "step": 293
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0001,
      "loss": 0.5244,
      "step": 294
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0001,
      "loss": 0.4751,
      "step": 295
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0001,
      "loss": 0.4696,
      "step": 296
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0001,
      "loss": 0.5434,
      "step": 297
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0001,
      "loss": 0.5122,
      "step": 298
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0001,
      "loss": 0.6085,
      "step": 299
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0001,
      "loss": 0.5736,
      "step": 300
    },
    {
      "epoch": 1.33,
      "eval_loss": 0.5659383535385132,
      "eval_runtime": 24.1127,
      "eval_samples_per_second": 3.359,
      "eval_steps_per_second": 0.871,
      "step": 300
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0001,
      "loss": 0.5834,
      "step": 301
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0001,
      "loss": 0.5414,
      "step": 302
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0001,
      "loss": 0.5692,
      "step": 303
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0001,
      "loss": 0.5972,
      "step": 304
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0001,
      "loss": 0.5569,
      "step": 305
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0001,
      "loss": 0.5139,
      "step": 306
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0001,
      "loss": 0.506,
      "step": 307
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001,
      "loss": 0.5161,
      "step": 308
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001,
      "loss": 0.5095,
      "step": 309
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0001,
      "loss": 0.5668,
      "step": 310
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0001,
      "loss": 0.5141,
      "step": 311
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0001,
      "loss": 0.4863,
      "step": 312
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0001,
      "loss": 0.5355,
      "step": 313
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001,
      "loss": 0.5008,
      "step": 314
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001,
      "loss": 0.5084,
      "step": 315
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001,
      "loss": 0.5296,
      "step": 316
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001,
      "loss": 0.5431,
      "step": 317
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001,
      "loss": 0.5514,
      "step": 318
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0001,
      "loss": 0.5013,
      "step": 319
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0001,
      "loss": 0.5648,
      "step": 320
    },
    {
      "epoch": 1.42,
      "eval_loss": 0.5671387910842896,
      "eval_runtime": 24.1009,
      "eval_samples_per_second": 3.361,
      "eval_steps_per_second": 0.871,
      "step": 320
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0001,
      "loss": 0.4979,
      "step": 321
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0001,
      "loss": 0.5319,
      "step": 322
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001,
      "loss": 0.5366,
      "step": 323
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001,
      "loss": 0.4552,
      "step": 324
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001,
      "loss": 0.4957,
      "step": 325
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0001,
      "loss": 0.5296,
      "step": 326
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0001,
      "loss": 0.5525,
      "step": 327
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001,
      "loss": 0.5684,
      "step": 328
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001,
      "loss": 0.51,
      "step": 329
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001,
      "loss": 0.548,
      "step": 330
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001,
      "loss": 0.5697,
      "step": 331
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001,
      "loss": 0.5436,
      "step": 332
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001,
      "loss": 0.5071,
      "step": 333
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001,
      "loss": 0.4868,
      "step": 334
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0001,
      "loss": 0.5495,
      "step": 335
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0001,
      "loss": 0.5394,
      "step": 336
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0001,
      "loss": 0.5377,
      "step": 337
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0001,
      "loss": 0.4897,
      "step": 338
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0001,
      "loss": 0.532,
      "step": 339
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0001,
      "loss": 0.496,
      "step": 340
    },
    {
      "epoch": 1.51,
      "eval_loss": 0.5622614622116089,
      "eval_runtime": 24.1201,
      "eval_samples_per_second": 3.358,
      "eval_steps_per_second": 0.871,
      "step": 340
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0001,
      "loss": 0.5655,
      "step": 341
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0001,
      "loss": 0.4694,
      "step": 342
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0001,
      "loss": 0.504,
      "step": 343
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001,
      "loss": 0.5137,
      "step": 344
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001,
      "loss": 0.4883,
      "step": 345
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0001,
      "loss": 0.5091,
      "step": 346
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0001,
      "loss": 0.5201,
      "step": 347
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0001,
      "loss": 0.5358,
      "step": 348
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0001,
      "loss": 0.5073,
      "step": 349
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0001,
      "loss": 0.5519,
      "step": 350
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0001,
      "loss": 0.4829,
      "step": 351
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0001,
      "loss": 0.523,
      "step": 352
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0001,
      "loss": 0.4998,
      "step": 353
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0001,
      "loss": 0.5102,
      "step": 354
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0001,
      "loss": 0.5053,
      "step": 355
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0001,
      "loss": 0.4836,
      "step": 356
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0001,
      "loss": 0.5636,
      "step": 357
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0001,
      "loss": 0.5772,
      "step": 358
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0001,
      "loss": 0.4938,
      "step": 359
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0001,
      "loss": 0.5311,
      "step": 360
    },
    {
      "epoch": 1.6,
      "eval_loss": 0.5604732632637024,
      "eval_runtime": 24.1036,
      "eval_samples_per_second": 3.36,
      "eval_steps_per_second": 0.871,
      "step": 360
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0001,
      "loss": 0.5216,
      "step": 361
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0001,
      "loss": 0.5932,
      "step": 362
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0001,
      "loss": 0.534,
      "step": 363
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0001,
      "loss": 0.5036,
      "step": 364
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0001,
      "loss": 0.5509,
      "step": 365
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0001,
      "loss": 0.5137,
      "step": 366
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0001,
      "loss": 0.5431,
      "step": 367
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0001,
      "loss": 0.4895,
      "step": 368
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0001,
      "loss": 0.5196,
      "step": 369
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0001,
      "loss": 0.5139,
      "step": 370
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0001,
      "loss": 0.4992,
      "step": 371
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0001,
      "loss": 0.4901,
      "step": 372
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0001,
      "loss": 0.4701,
      "step": 373
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0001,
      "loss": 0.474,
      "step": 374
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0001,
      "loss": 0.5518,
      "step": 375
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0001,
      "loss": 0.6052,
      "step": 376
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0001,
      "loss": 0.5274,
      "step": 377
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0001,
      "loss": 0.5234,
      "step": 378
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0001,
      "loss": 0.5269,
      "step": 379
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0001,
      "loss": 0.5834,
      "step": 380
    },
    {
      "epoch": 1.69,
      "eval_loss": 0.5608329176902771,
      "eval_runtime": 24.1068,
      "eval_samples_per_second": 3.36,
      "eval_steps_per_second": 0.871,
      "step": 380
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0001,
      "loss": 0.5721,
      "step": 381
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0001,
      "loss": 0.5099,
      "step": 382
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0001,
      "loss": 0.5707,
      "step": 383
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0001,
      "loss": 0.5459,
      "step": 384
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0001,
      "loss": 0.4994,
      "step": 385
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0001,
      "loss": 0.513,
      "step": 386
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0001,
      "loss": 0.5423,
      "step": 387
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0001,
      "loss": 0.5675,
      "step": 388
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0001,
      "loss": 0.5498,
      "step": 389
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0001,
      "loss": 0.5062,
      "step": 390
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0001,
      "loss": 0.5369,
      "step": 391
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0001,
      "loss": 0.5517,
      "step": 392
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0001,
      "loss": 0.5299,
      "step": 393
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0001,
      "loss": 0.536,
      "step": 394
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0001,
      "loss": 0.5051,
      "step": 395
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0001,
      "loss": 0.5908,
      "step": 396
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0001,
      "loss": 0.5486,
      "step": 397
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0001,
      "loss": 0.58,
      "step": 398
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0001,
      "loss": 0.5417,
      "step": 399
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0001,
      "loss": 0.4939,
      "step": 400
    },
    {
      "epoch": 1.78,
      "eval_loss": 0.5567287802696228,
      "eval_runtime": 24.0718,
      "eval_samples_per_second": 3.365,
      "eval_steps_per_second": 0.872,
      "step": 400
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0001,
      "loss": 0.5521,
      "step": 401
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0001,
      "loss": 0.505,
      "step": 402
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0001,
      "loss": 0.5303,
      "step": 403
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0001,
      "loss": 0.4961,
      "step": 404
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0001,
      "loss": 0.5463,
      "step": 405
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0001,
      "loss": 0.5453,
      "step": 406
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0001,
      "loss": 0.5163,
      "step": 407
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0001,
      "loss": 0.5589,
      "step": 408
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0001,
      "loss": 0.5152,
      "step": 409
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0001,
      "loss": 0.5602,
      "step": 410
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0001,
      "loss": 0.5261,
      "step": 411
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0001,
      "loss": 0.5312,
      "step": 412
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0001,
      "loss": 0.5673,
      "step": 413
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0001,
      "loss": 0.515,
      "step": 414
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0001,
      "loss": 0.4986,
      "step": 415
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0001,
      "loss": 0.5095,
      "step": 416
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0001,
      "loss": 0.5371,
      "step": 417
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0001,
      "loss": 0.5161,
      "step": 418
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0001,
      "loss": 0.5198,
      "step": 419
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0001,
      "loss": 0.4753,
      "step": 420
    },
    {
      "epoch": 1.87,
      "eval_loss": 0.5573872923851013,
      "eval_runtime": 24.0896,
      "eval_samples_per_second": 3.362,
      "eval_steps_per_second": 0.872,
      "step": 420
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0001,
      "loss": 0.6157,
      "step": 421
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0001,
      "loss": 0.5122,
      "step": 422
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0001,
      "loss": 0.5375,
      "step": 423
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0001,
      "loss": 0.511,
      "step": 424
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0001,
      "loss": 0.4829,
      "step": 425
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0001,
      "loss": 0.5713,
      "step": 426
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0001,
      "loss": 0.5718,
      "step": 427
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0001,
      "loss": 0.5628,
      "step": 428
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0001,
      "loss": 0.5441,
      "step": 429
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0001,
      "loss": 0.5442,
      "step": 430
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0001,
      "loss": 0.5441,
      "step": 431
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0001,
      "loss": 0.5701,
      "step": 432
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0001,
      "loss": 0.4806,
      "step": 433
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0001,
      "loss": 0.5127,
      "step": 434
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0001,
      "loss": 0.4866,
      "step": 435
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0001,
      "loss": 0.5048,
      "step": 436
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0001,
      "loss": 0.5271,
      "step": 437
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0001,
      "loss": 0.4898,
      "step": 438
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0001,
      "loss": 0.5233,
      "step": 439
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0001,
      "loss": 0.5115,
      "step": 440
    },
    {
      "epoch": 1.96,
      "eval_loss": 0.5556459426879883,
      "eval_runtime": 24.0913,
      "eval_samples_per_second": 3.362,
      "eval_steps_per_second": 0.872,
      "step": 440
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0001,
      "loss": 0.4692,
      "step": 441
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0001,
      "loss": 0.5076,
      "step": 442
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0001,
      "loss": 0.5331,
      "step": 443
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0001,
      "loss": 0.4602,
      "step": 444
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.0001,
      "loss": 0.5247,
      "step": 445
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.0001,
      "loss": 0.4663,
      "step": 446
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.0001,
      "loss": 0.5308,
      "step": 447
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.0001,
      "loss": 0.5013,
      "step": 448
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0001,
      "loss": 0.5739,
      "step": 449
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0001,
      "loss": 0.4782,
      "step": 450
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0001,
      "loss": 0.5388,
      "step": 451
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0001,
      "loss": 0.497,
      "step": 452
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0001,
      "loss": 0.5305,
      "step": 453
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0001,
      "loss": 0.4813,
      "step": 454
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0001,
      "loss": 0.5366,
      "step": 455
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0001,
      "loss": 0.5395,
      "step": 456
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0001,
      "loss": 0.4534,
      "step": 457
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0001,
      "loss": 0.4877,
      "step": 458
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0001,
      "loss": 0.4893,
      "step": 459
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0001,
      "loss": 0.5742,
      "step": 460
    },
    {
      "epoch": 2.04,
      "eval_loss": 0.5570523738861084,
      "eval_runtime": 24.0958,
      "eval_samples_per_second": 3.362,
      "eval_steps_per_second": 0.872,
      "step": 460
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0001,
      "loss": 0.4744,
      "step": 461
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0001,
      "loss": 0.5214,
      "step": 462
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0001,
      "loss": 0.5228,
      "step": 463
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0001,
      "loss": 0.4773,
      "step": 464
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0001,
      "loss": 0.4839,
      "step": 465
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0001,
      "loss": 0.4434,
      "step": 466
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0001,
      "loss": 0.4146,
      "step": 467
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0001,
      "loss": 0.4636,
      "step": 468
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0001,
      "loss": 0.4773,
      "step": 469
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0001,
      "loss": 0.4388,
      "step": 470
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0001,
      "loss": 0.5304,
      "step": 471
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0001,
      "loss": 0.4787,
      "step": 472
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0001,
      "loss": 0.4495,
      "step": 473
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0001,
      "loss": 0.4295,
      "step": 474
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0001,
      "loss": 0.5033,
      "step": 475
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0001,
      "loss": 0.4499,
      "step": 476
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0001,
      "loss": 0.4977,
      "step": 477
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0001,
      "loss": 0.5306,
      "step": 478
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0001,
      "loss": 0.4181,
      "step": 479
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0001,
      "loss": 0.4056,
      "step": 480
    },
    {
      "epoch": 2.13,
      "eval_loss": 0.5588575601577759,
      "eval_runtime": 24.1124,
      "eval_samples_per_second": 3.359,
      "eval_steps_per_second": 0.871,
      "step": 480
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0001,
      "loss": 0.4385,
      "step": 481
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0001,
      "loss": 0.5353,
      "step": 482
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0001,
      "loss": 0.442,
      "step": 483
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0001,
      "loss": 0.464,
      "step": 484
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0001,
      "loss": 0.4987,
      "step": 485
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0001,
      "loss": 0.4449,
      "step": 486
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0001,
      "loss": 0.4393,
      "step": 487
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0001,
      "loss": 0.4425,
      "step": 488
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0001,
      "loss": 0.4644,
      "step": 489
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0001,
      "loss": 0.415,
      "step": 490
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0001,
      "loss": 0.4426,
      "step": 491
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0001,
      "loss": 0.4531,
      "step": 492
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0001,
      "loss": 0.4703,
      "step": 493
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0001,
      "loss": 0.3959,
      "step": 494
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0001,
      "loss": 0.4202,
      "step": 495
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0001,
      "loss": 0.3943,
      "step": 496
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0001,
      "loss": 0.4682,
      "step": 497
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0001,
      "loss": 0.4149,
      "step": 498
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0001,
      "loss": 0.3943,
      "step": 499
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0001,
      "loss": 0.4107,
      "step": 500
    },
    {
      "epoch": 2.22,
      "eval_loss": 0.5702491998672485,
      "eval_runtime": 24.116,
      "eval_samples_per_second": 3.359,
      "eval_steps_per_second": 0.871,
      "step": 500
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0001,
      "loss": 0.4747,
      "step": 501
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0001,
      "loss": 0.5236,
      "step": 502
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0001,
      "loss": 0.5227,
      "step": 503
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0001,
      "loss": 0.4735,
      "step": 504
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0001,
      "loss": 0.5027,
      "step": 505
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0001,
      "loss": 0.5624,
      "step": 506
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0001,
      "loss": 0.5559,
      "step": 507
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0001,
      "loss": 0.4389,
      "step": 508
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0001,
      "loss": 0.5332,
      "step": 509
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0001,
      "loss": 0.5011,
      "step": 510
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0001,
      "loss": 0.4945,
      "step": 511
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0001,
      "loss": 0.5385,
      "step": 512
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0001,
      "loss": 0.455,
      "step": 513
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0001,
      "loss": 0.4694,
      "step": 514
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0001,
      "loss": 0.5,
      "step": 515
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0001,
      "loss": 0.4812,
      "step": 516
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0001,
      "loss": 0.481,
      "step": 517
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0001,
      "loss": 0.4765,
      "step": 518
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0001,
      "loss": 0.4208,
      "step": 519
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0001,
      "loss": 0.4539,
      "step": 520
    },
    {
      "epoch": 2.31,
      "eval_loss": 0.5589257478713989,
      "eval_runtime": 24.1012,
      "eval_samples_per_second": 3.361,
      "eval_steps_per_second": 0.871,
      "step": 520
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0001,
      "loss": 0.4978,
      "step": 521
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0001,
      "loss": 0.4789,
      "step": 522
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0001,
      "loss": 0.5233,
      "step": 523
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0001,
      "loss": 0.5108,
      "step": 524
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0001,
      "loss": 0.4863,
      "step": 525
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0001,
      "loss": 0.4018,
      "step": 526
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0001,
      "loss": 0.4524,
      "step": 527
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0001,
      "loss": 0.4495,
      "step": 528
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0001,
      "loss": 0.4484,
      "step": 529
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0001,
      "loss": 0.4857,
      "step": 530
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0001,
      "loss": 0.4388,
      "step": 531
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0001,
      "loss": 0.5182,
      "step": 532
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0001,
      "loss": 0.4622,
      "step": 533
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0001,
      "loss": 0.4832,
      "step": 534
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0001,
      "loss": 0.4628,
      "step": 535
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0001,
      "loss": 0.4723,
      "step": 536
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0001,
      "loss": 0.4359,
      "step": 537
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0001,
      "loss": 0.4265,
      "step": 538
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0001,
      "loss": 0.4319,
      "step": 539
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0001,
      "loss": 0.5404,
      "step": 540
    },
    {
      "epoch": 2.4,
      "eval_loss": 0.5609443783760071,
      "eval_runtime": 24.1082,
      "eval_samples_per_second": 3.36,
      "eval_steps_per_second": 0.871,
      "step": 540
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0001,
      "loss": 0.4683,
      "step": 541
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0001,
      "loss": 0.4834,
      "step": 542
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0001,
      "loss": 0.4519,
      "step": 543
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0001,
      "loss": 0.4503,
      "step": 544
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0001,
      "loss": 0.4313,
      "step": 545
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0001,
      "loss": 0.478,
      "step": 546
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0001,
      "loss": 0.4184,
      "step": 547
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0001,
      "loss": 0.415,
      "step": 548
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0001,
      "loss": 0.4029,
      "step": 549
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0001,
      "loss": 0.4797,
      "step": 550
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0001,
      "loss": 0.5585,
      "step": 551
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0001,
      "loss": 0.56,
      "step": 552
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0001,
      "loss": 0.5369,
      "step": 553
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0001,
      "loss": 0.466,
      "step": 554
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0001,
      "loss": 0.5259,
      "step": 555
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0001,
      "loss": 0.5448,
      "step": 556
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0001,
      "loss": 0.4472,
      "step": 557
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0001,
      "loss": 0.5056,
      "step": 558
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0001,
      "loss": 0.4782,
      "step": 559
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0001,
      "loss": 0.5448,
      "step": 560
    },
    {
      "epoch": 2.49,
      "eval_loss": 0.5567753314971924,
      "eval_runtime": 24.1173,
      "eval_samples_per_second": 3.359,
      "eval_steps_per_second": 0.871,
      "step": 560
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0001,
      "loss": 0.4553,
      "step": 561
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0001,
      "loss": 0.5287,
      "step": 562
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0001,
      "loss": 0.4369,
      "step": 563
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0001,
      "loss": 0.5289,
      "step": 564
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0001,
      "loss": 0.4811,
      "step": 565
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0001,
      "loss": 0.4602,
      "step": 566
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0001,
      "loss": 0.4847,
      "step": 567
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0001,
      "loss": 0.4409,
      "step": 568
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0001,
      "loss": 0.4455,
      "step": 569
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0001,
      "loss": 0.46,
      "step": 570
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0001,
      "loss": 0.4576,
      "step": 571
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0001,
      "loss": 0.5368,
      "step": 572
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0001,
      "loss": 0.482,
      "step": 573
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0001,
      "loss": 0.4726,
      "step": 574
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0001,
      "loss": 0.4965,
      "step": 575
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0001,
      "loss": 0.5308,
      "step": 576
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0001,
      "loss": 0.4908,
      "step": 577
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0001,
      "loss": 0.4788,
      "step": 578
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0001,
      "loss": 0.4657,
      "step": 579
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0001,
      "loss": 0.4344,
      "step": 580
    },
    {
      "epoch": 2.58,
      "eval_loss": 0.5566776394844055,
      "eval_runtime": 24.1316,
      "eval_samples_per_second": 3.357,
      "eval_steps_per_second": 0.87,
      "step": 580
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0001,
      "loss": 0.4564,
      "step": 581
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0001,
      "loss": 0.4481,
      "step": 582
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0001,
      "loss": 0.4549,
      "step": 583
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0001,
      "loss": 0.4598,
      "step": 584
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0001,
      "loss": 0.534,
      "step": 585
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0001,
      "loss": 0.4528,
      "step": 586
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0001,
      "loss": 0.443,
      "step": 587
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0001,
      "loss": 0.4663,
      "step": 588
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0001,
      "loss": 0.4555,
      "step": 589
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0001,
      "loss": 0.454,
      "step": 590
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0001,
      "loss": 0.4626,
      "step": 591
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0001,
      "loss": 0.4164,
      "step": 592
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0001,
      "loss": 0.4086,
      "step": 593
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0001,
      "loss": 0.4684,
      "step": 594
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0001,
      "loss": 0.5382,
      "step": 595
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0001,
      "loss": 0.4771,
      "step": 596
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0001,
      "loss": 0.4459,
      "step": 597
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0001,
      "loss": 0.4068,
      "step": 598
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0001,
      "loss": 0.421,
      "step": 599
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0001,
      "loss": 0.3814,
      "step": 600
    },
    {
      "epoch": 2.67,
      "eval_loss": 0.5660998225212097,
      "eval_runtime": 24.1229,
      "eval_samples_per_second": 3.358,
      "eval_steps_per_second": 0.871,
      "step": 600
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0001,
      "loss": 0.5285,
      "step": 601
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0001,
      "loss": 0.4892,
      "step": 602
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0001,
      "loss": 0.4713,
      "step": 603
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0001,
      "loss": 0.4877,
      "step": 604
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0001,
      "loss": 0.5598,
      "step": 605
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0001,
      "loss": 0.4981,
      "step": 606
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0001,
      "loss": 0.523,
      "step": 607
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0001,
      "loss": 0.6273,
      "step": 608
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0001,
      "loss": 0.4796,
      "step": 609
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0001,
      "loss": 0.4946,
      "step": 610
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0001,
      "loss": 0.4764,
      "step": 611
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0001,
      "loss": 0.4761,
      "step": 612
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0001,
      "loss": 0.4968,
      "step": 613
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0001,
      "loss": 0.4851,
      "step": 614
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0001,
      "loss": 0.577,
      "step": 615
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0001,
      "loss": 0.4857,
      "step": 616
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0001,
      "loss": 0.4485,
      "step": 617
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0001,
      "loss": 0.4615,
      "step": 618
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0001,
      "loss": 0.4864,
      "step": 619
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0001,
      "loss": 0.5176,
      "step": 620
    },
    {
      "epoch": 2.76,
      "eval_loss": 0.5543871521949768,
      "eval_runtime": 24.1187,
      "eval_samples_per_second": 3.358,
      "eval_steps_per_second": 0.871,
      "step": 620
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0001,
      "loss": 0.4426,
      "step": 621
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0001,
      "loss": 0.5002,
      "step": 622
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0001,
      "loss": 0.4899,
      "step": 623
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0001,
      "loss": 0.5254,
      "step": 624
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0001,
      "loss": 0.4458,
      "step": 625
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0001,
      "loss": 0.4558,
      "step": 626
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.0001,
      "loss": 0.4568,
      "step": 627
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.0001,
      "loss": 0.4449,
      "step": 628
    },
    {
      "epoch": 2.8,
      "learning_rate": 0.0001,
      "loss": 0.4423,
      "step": 629
    },
    {
      "epoch": 2.8,
      "learning_rate": 0.0001,
      "loss": 0.4581,
      "step": 630
    },
    {
      "epoch": 2.8,
      "learning_rate": 0.0001,
      "loss": 0.4878,
      "step": 631
    },
    {
      "epoch": 2.81,
      "learning_rate": 0.0001,
      "loss": 0.4397,
      "step": 632
    },
    {
      "epoch": 2.81,
      "learning_rate": 0.0001,
      "loss": 0.5023,
      "step": 633
    },
    {
      "epoch": 2.82,
      "learning_rate": 0.0001,
      "loss": 0.5071,
      "step": 634
    },
    {
      "epoch": 2.82,
      "learning_rate": 0.0001,
      "loss": 0.4481,
      "step": 635
    },
    {
      "epoch": 2.83,
      "learning_rate": 0.0001,
      "loss": 0.4919,
      "step": 636
    },
    {
      "epoch": 2.83,
      "learning_rate": 0.0001,
      "loss": 0.4873,
      "step": 637
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.0001,
      "loss": 0.4878,
      "step": 638
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.0001,
      "loss": 0.4719,
      "step": 639
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.0001,
      "loss": 0.4563,
      "step": 640
    },
    {
      "epoch": 2.84,
      "eval_loss": 0.554156482219696,
      "eval_runtime": 24.0905,
      "eval_samples_per_second": 3.362,
      "eval_steps_per_second": 0.872,
      "step": 640
    },
    {
      "epoch": 2.85,
      "learning_rate": 0.0001,
      "loss": 0.4496,
      "step": 641
    },
    {
      "epoch": 2.85,
      "learning_rate": 0.0001,
      "loss": 0.4601,
      "step": 642
    },
    {
      "epoch": 2.86,
      "learning_rate": 0.0001,
      "loss": 0.4357,
      "step": 643
    },
    {
      "epoch": 2.86,
      "learning_rate": 0.0001,
      "loss": 0.4693,
      "step": 644
    },
    {
      "epoch": 2.87,
      "learning_rate": 0.0001,
      "loss": 0.503,
      "step": 645
    },
    {
      "epoch": 2.87,
      "learning_rate": 0.0001,
      "loss": 0.4489,
      "step": 646
    },
    {
      "epoch": 2.88,
      "learning_rate": 0.0001,
      "loss": 0.4086,
      "step": 647
    },
    {
      "epoch": 2.88,
      "learning_rate": 0.0001,
      "loss": 0.4727,
      "step": 648
    },
    {
      "epoch": 2.88,
      "learning_rate": 0.0001,
      "loss": 0.4289,
      "step": 649
    },
    {
      "epoch": 2.89,
      "learning_rate": 0.0001,
      "loss": 0.4302,
      "step": 650
    },
    {
      "epoch": 2.89,
      "learning_rate": 0.0001,
      "loss": 0.5574,
      "step": 651
    },
    {
      "epoch": 2.9,
      "learning_rate": 0.0001,
      "loss": 0.4812,
      "step": 652
    },
    {
      "epoch": 2.9,
      "learning_rate": 0.0001,
      "loss": 0.4935,
      "step": 653
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.0001,
      "loss": 0.519,
      "step": 654
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.0001,
      "loss": 0.5092,
      "step": 655
    },
    {
      "epoch": 2.92,
      "learning_rate": 0.0001,
      "loss": 0.4837,
      "step": 656
    },
    {
      "epoch": 2.92,
      "learning_rate": 0.0001,
      "loss": 0.4728,
      "step": 657
    },
    {
      "epoch": 2.92,
      "learning_rate": 0.0001,
      "loss": 0.4994,
      "step": 658
    },
    {
      "epoch": 2.93,
      "learning_rate": 0.0001,
      "loss": 0.4842,
      "step": 659
    },
    {
      "epoch": 2.93,
      "learning_rate": 0.0001,
      "loss": 0.4697,
      "step": 660
    },
    {
      "epoch": 2.93,
      "eval_loss": 0.5526395440101624,
      "eval_runtime": 24.11,
      "eval_samples_per_second": 3.36,
      "eval_steps_per_second": 0.871,
      "step": 660
    },
    {
      "epoch": 2.94,
      "learning_rate": 0.0001,
      "loss": 0.4133,
      "step": 661
    },
    {
      "epoch": 2.94,
      "learning_rate": 0.0001,
      "loss": 0.4212,
      "step": 662
    },
    {
      "epoch": 2.95,
      "learning_rate": 0.0001,
      "loss": 0.4936,
      "step": 663
    },
    {
      "epoch": 2.95,
      "learning_rate": 0.0001,
      "loss": 0.5138,
      "step": 664
    },
    {
      "epoch": 2.96,
      "learning_rate": 0.0001,
      "loss": 0.5091,
      "step": 665
    },
    {
      "epoch": 2.96,
      "learning_rate": 0.0001,
      "loss": 0.454,
      "step": 666
    },
    {
      "epoch": 2.96,
      "learning_rate": 0.0001,
      "loss": 0.4532,
      "step": 667
    },
    {
      "epoch": 2.97,
      "learning_rate": 0.0001,
      "loss": 0.4718,
      "step": 668
    },
    {
      "epoch": 2.97,
      "learning_rate": 0.0001,
      "loss": 0.4531,
      "step": 669
    },
    {
      "epoch": 2.98,
      "learning_rate": 0.0001,
      "loss": 0.43,
      "step": 670
    },
    {
      "epoch": 2.98,
      "learning_rate": 0.0001,
      "loss": 0.4563,
      "step": 671
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.0001,
      "loss": 0.4249,
      "step": 672
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.0001,
      "loss": 0.4351,
      "step": 673
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0001,
      "loss": 0.3934,
      "step": 674
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0001,
      "loss": 0.4883,
      "step": 675
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0001,
      "loss": 0.5506,
      "step": 676
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0001,
      "loss": 0.4671,
      "step": 677
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0001,
      "loss": 0.4837,
      "step": 678
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0001,
      "loss": 0.4255,
      "step": 679
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0001,
      "loss": 0.4636,
      "step": 680
    },
    {
      "epoch": 3.02,
      "eval_loss": 0.5592077970504761,
      "eval_runtime": 24.0715,
      "eval_samples_per_second": 3.365,
      "eval_steps_per_second": 0.872,
      "step": 680
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0001,
      "loss": 0.4747,
      "step": 681
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0001,
      "loss": 0.4245,
      "step": 682
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0001,
      "loss": 0.4396,
      "step": 683
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0001,
      "loss": 0.4285,
      "step": 684
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0001,
      "loss": 0.4468,
      "step": 685
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0001,
      "loss": 0.4381,
      "step": 686
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0001,
      "loss": 0.4515,
      "step": 687
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0001,
      "loss": 0.4223,
      "step": 688
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0001,
      "loss": 0.4254,
      "step": 689
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0001,
      "loss": 0.4104,
      "step": 690
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0001,
      "loss": 0.4124,
      "step": 691
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0001,
      "loss": 0.4246,
      "step": 692
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0001,
      "loss": 0.3641,
      "step": 693
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0001,
      "loss": 0.4427,
      "step": 694
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0001,
      "loss": 0.4322,
      "step": 695
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0001,
      "loss": 0.3643,
      "step": 696
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0001,
      "loss": 0.459,
      "step": 697
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0001,
      "loss": 0.4377,
      "step": 698
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0001,
      "loss": 0.3954,
      "step": 699
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0001,
      "loss": 0.4524,
      "step": 700
    },
    {
      "epoch": 3.11,
      "eval_loss": 0.5644525289535522,
      "eval_runtime": 24.1024,
      "eval_samples_per_second": 3.361,
      "eval_steps_per_second": 0.871,
      "step": 700
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0001,
      "loss": 0.4379,
      "step": 701
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0001,
      "loss": 0.3573,
      "step": 702
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0001,
      "loss": 0.4522,
      "step": 703
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0001,
      "loss": 0.3428,
      "step": 704
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0001,
      "loss": 0.3873,
      "step": 705
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0001,
      "loss": 0.4135,
      "step": 706
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0001,
      "loss": 0.4353,
      "step": 707
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0001,
      "loss": 0.4034,
      "step": 708
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0001,
      "loss": 0.4034,
      "step": 709
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0001,
      "loss": 0.3874,
      "step": 710
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0001,
      "loss": 0.4126,
      "step": 711
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0001,
      "loss": 0.367,
      "step": 712
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0001,
      "loss": 0.3941,
      "step": 713
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0001,
      "loss": 0.355,
      "step": 714
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0001,
      "loss": 0.3581,
      "step": 715
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0001,
      "loss": 0.3784,
      "step": 716
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0001,
      "loss": 0.3789,
      "step": 717
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0001,
      "loss": 0.3545,
      "step": 718
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0001,
      "loss": 0.3727,
      "step": 719
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0001,
      "loss": 0.3919,
      "step": 720
    },
    {
      "epoch": 3.2,
      "eval_loss": 0.5801900029182434,
      "eval_runtime": 24.1157,
      "eval_samples_per_second": 3.359,
      "eval_steps_per_second": 0.871,
      "step": 720
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0001,
      "loss": 0.3662,
      "step": 721
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0001,
      "loss": 0.398,
      "step": 722
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0001,
      "loss": 0.32,
      "step": 723
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0001,
      "loss": 0.3239,
      "step": 724
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0001,
      "loss": 0.3296,
      "step": 725
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0001,
      "loss": 0.5341,
      "step": 726
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0001,
      "loss": 0.4906,
      "step": 727
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0001,
      "loss": 0.4745,
      "step": 728
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0001,
      "loss": 0.4542,
      "step": 729
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0001,
      "loss": 0.4518,
      "step": 730
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0001,
      "loss": 0.449,
      "step": 731
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0001,
      "loss": 0.4719,
      "step": 732
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0001,
      "loss": 0.4844,
      "step": 733
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0001,
      "loss": 0.5027,
      "step": 734
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0001,
      "loss": 0.4252,
      "step": 735
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0001,
      "loss": 0.4273,
      "step": 736
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0001,
      "loss": 0.3969,
      "step": 737
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0001,
      "loss": 0.4431,
      "step": 738
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0001,
      "loss": 0.4409,
      "step": 739
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0001,
      "loss": 0.4234,
      "step": 740
    },
    {
      "epoch": 3.29,
      "eval_loss": 0.5601220726966858,
      "eval_runtime": 24.0933,
      "eval_samples_per_second": 3.362,
      "eval_steps_per_second": 0.872,
      "step": 740
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0001,
      "loss": 0.4385,
      "step": 741
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0001,
      "loss": 0.3971,
      "step": 742
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0001,
      "loss": 0.4323,
      "step": 743
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0001,
      "loss": 0.3851,
      "step": 744
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0001,
      "loss": 0.4015,
      "step": 745
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0001,
      "loss": 0.4419,
      "step": 746
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0001,
      "loss": 0.4382,
      "step": 747
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0001,
      "loss": 0.3991,
      "step": 748
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0001,
      "loss": 0.4382,
      "step": 749
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0001,
      "loss": 0.3881,
      "step": 750
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0001,
      "loss": 0.4292,
      "step": 751
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0001,
      "loss": 0.4199,
      "step": 752
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0001,
      "loss": 0.4382,
      "step": 753
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0001,
      "loss": 0.3792,
      "step": 754
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0001,
      "loss": 0.4032,
      "step": 755
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0001,
      "loss": 0.4279,
      "step": 756
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0001,
      "loss": 0.3657,
      "step": 757
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0001,
      "loss": 0.4118,
      "step": 758
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0001,
      "loss": 0.4667,
      "step": 759
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0001,
      "loss": 0.4281,
      "step": 760
    },
    {
      "epoch": 3.38,
      "eval_loss": 0.5698632597923279,
      "eval_runtime": 24.099,
      "eval_samples_per_second": 3.361,
      "eval_steps_per_second": 0.871,
      "step": 760
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0001,
      "loss": 0.428,
      "step": 761
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0001,
      "loss": 0.3752,
      "step": 762
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0001,
      "loss": 0.3762,
      "step": 763
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0001,
      "loss": 0.3779,
      "step": 764
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0001,
      "loss": 0.3607,
      "step": 765
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0001,
      "loss": 0.3495,
      "step": 766
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0001,
      "loss": 0.4067,
      "step": 767
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0001,
      "loss": 0.3657,
      "step": 768
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0001,
      "loss": 0.3845,
      "step": 769
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0001,
      "loss": 0.3801,
      "step": 770
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0001,
      "loss": 0.3608,
      "step": 771
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0001,
      "loss": 0.3927,
      "step": 772
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0001,
      "loss": 0.2951,
      "step": 773
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0001,
      "loss": 0.3267,
      "step": 774
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0001,
      "loss": 0.3367,
      "step": 775
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0001,
      "loss": 0.4904,
      "step": 776
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0001,
      "loss": 0.4673,
      "step": 777
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0001,
      "loss": 0.4591,
      "step": 778
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0001,
      "loss": 0.4776,
      "step": 779
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0001,
      "loss": 0.5314,
      "step": 780
    },
    {
      "epoch": 3.47,
      "eval_loss": 0.5754256248474121,
      "eval_runtime": 24.1165,
      "eval_samples_per_second": 3.359,
      "eval_steps_per_second": 0.871,
      "step": 780
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0001,
      "loss": 0.4819,
      "step": 781
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0001,
      "loss": 0.4466,
      "step": 782
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0001,
      "loss": 0.5092,
      "step": 783
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0001,
      "loss": 0.4889,
      "step": 784
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0001,
      "loss": 0.4197,
      "step": 785
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0001,
      "loss": 0.5453,
      "step": 786
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0001,
      "loss": 0.4253,
      "step": 787
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0001,
      "loss": 0.4924,
      "step": 788
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0001,
      "loss": 0.474,
      "step": 789
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0001,
      "loss": 0.4691,
      "step": 790
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0001,
      "loss": 0.488,
      "step": 791
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0001,
      "loss": 0.4487,
      "step": 792
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0001,
      "loss": 0.4285,
      "step": 793
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0001,
      "loss": 0.411,
      "step": 794
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0001,
      "loss": 0.4045,
      "step": 795
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0001,
      "loss": 0.4255,
      "step": 796
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0001,
      "loss": 0.4022,
      "step": 797
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.0001,
      "loss": 0.417,
      "step": 798
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.0001,
      "loss": 0.4677,
      "step": 799
    },
    {
      "epoch": 3.56,
      "learning_rate": 0.0001,
      "loss": 0.4362,
      "step": 800
    },
    {
      "epoch": 3.56,
      "eval_loss": 0.5662307739257812,
      "eval_runtime": 24.0683,
      "eval_samples_per_second": 3.365,
      "eval_steps_per_second": 0.873,
      "step": 800
    },
    {
      "epoch": 3.56,
      "learning_rate": 0.0001,
      "loss": 0.3927,
      "step": 801
    },
    {
      "epoch": 3.56,
      "learning_rate": 0.0001,
      "loss": 0.3961,
      "step": 802
    },
    {
      "epoch": 3.57,
      "learning_rate": 0.0001,
      "loss": 0.4362,
      "step": 803
    },
    {
      "epoch": 3.57,
      "learning_rate": 0.0001,
      "loss": 0.3974,
      "step": 804
    },
    {
      "epoch": 3.58,
      "learning_rate": 0.0001,
      "loss": 0.4463,
      "step": 805
    },
    {
      "epoch": 3.58,
      "learning_rate": 0.0001,
      "loss": 0.3887,
      "step": 806
    },
    {
      "epoch": 3.59,
      "learning_rate": 0.0001,
      "loss": 0.4105,
      "step": 807
    },
    {
      "epoch": 3.59,
      "learning_rate": 0.0001,
      "loss": 0.3885,
      "step": 808
    },
    {
      "epoch": 3.6,
      "learning_rate": 0.0001,
      "loss": 0.442,
      "step": 809
    },
    {
      "epoch": 3.6,
      "learning_rate": 0.0001,
      "loss": 0.3711,
      "step": 810
    },
    {
      "epoch": 3.6,
      "learning_rate": 0.0001,
      "loss": 0.393,
      "step": 811
    },
    {
      "epoch": 3.61,
      "learning_rate": 0.0001,
      "loss": 0.416,
      "step": 812
    },
    {
      "epoch": 3.61,
      "learning_rate": 0.0001,
      "loss": 0.4141,
      "step": 813
    },
    {
      "epoch": 3.62,
      "learning_rate": 0.0001,
      "loss": 0.4338,
      "step": 814
    },
    {
      "epoch": 3.62,
      "learning_rate": 0.0001,
      "loss": 0.4081,
      "step": 815
    },
    {
      "epoch": 3.63,
      "learning_rate": 0.0001,
      "loss": 0.4054,
      "step": 816
    },
    {
      "epoch": 3.63,
      "learning_rate": 0.0001,
      "loss": 0.3894,
      "step": 817
    },
    {
      "epoch": 3.64,
      "learning_rate": 0.0001,
      "loss": 0.3737,
      "step": 818
    },
    {
      "epoch": 3.64,
      "learning_rate": 0.0001,
      "loss": 0.4137,
      "step": 819
    },
    {
      "epoch": 3.64,
      "learning_rate": 0.0001,
      "loss": 0.3944,
      "step": 820
    },
    {
      "epoch": 3.64,
      "eval_loss": 0.5764873027801514,
      "eval_runtime": 24.0959,
      "eval_samples_per_second": 3.362,
      "eval_steps_per_second": 0.872,
      "step": 820
    },
    {
      "epoch": 3.65,
      "learning_rate": 0.0001,
      "loss": 0.3385,
      "step": 821
    },
    {
      "epoch": 3.65,
      "learning_rate": 0.0001,
      "loss": 0.3736,
      "step": 822
    },
    {
      "epoch": 3.66,
      "learning_rate": 0.0001,
      "loss": 0.3769,
      "step": 823
    },
    {
      "epoch": 3.66,
      "learning_rate": 0.0001,
      "loss": 0.3751,
      "step": 824
    },
    {
      "epoch": 3.67,
      "learning_rate": 0.0001,
      "loss": 0.353,
      "step": 825
    },
    {
      "epoch": 3.67,
      "learning_rate": 0.0001,
      "loss": 0.5471,
      "step": 826
    },
    {
      "epoch": 3.68,
      "learning_rate": 0.0001,
      "loss": 0.4697,
      "step": 827
    },
    {
      "epoch": 3.68,
      "learning_rate": 0.0001,
      "loss": 0.4157,
      "step": 828
    },
    {
      "epoch": 3.68,
      "learning_rate": 0.0001,
      "loss": 0.4827,
      "step": 829
    },
    {
      "epoch": 3.69,
      "learning_rate": 0.0001,
      "loss": 0.4743,
      "step": 830
    },
    {
      "epoch": 3.69,
      "learning_rate": 0.0001,
      "loss": 0.4747,
      "step": 831
    },
    {
      "epoch": 3.7,
      "learning_rate": 0.0001,
      "loss": 0.4999,
      "step": 832
    },
    {
      "epoch": 3.7,
      "learning_rate": 0.0001,
      "loss": 0.4967,
      "step": 833
    },
    {
      "epoch": 3.71,
      "learning_rate": 0.0001,
      "loss": 0.4716,
      "step": 834
    },
    {
      "epoch": 3.71,
      "learning_rate": 0.0001,
      "loss": 0.4255,
      "step": 835
    },
    {
      "epoch": 3.72,
      "learning_rate": 0.0001,
      "loss": 0.4114,
      "step": 836
    },
    {
      "epoch": 3.72,
      "learning_rate": 0.0001,
      "loss": 0.4663,
      "step": 837
    },
    {
      "epoch": 3.72,
      "learning_rate": 0.0001,
      "loss": 0.4501,
      "step": 838
    },
    {
      "epoch": 3.73,
      "learning_rate": 0.0001,
      "loss": 0.421,
      "step": 839
    },
    {
      "epoch": 3.73,
      "learning_rate": 0.0001,
      "loss": 0.483,
      "step": 840
    },
    {
      "epoch": 3.73,
      "eval_loss": 0.5593394041061401,
      "eval_runtime": 24.079,
      "eval_samples_per_second": 3.364,
      "eval_steps_per_second": 0.872,
      "step": 840
    },
    {
      "epoch": 3.74,
      "learning_rate": 0.0001,
      "loss": 0.3762,
      "step": 841
    },
    {
      "epoch": 3.74,
      "learning_rate": 0.0001,
      "loss": 0.4226,
      "step": 842
    },
    {
      "epoch": 3.75,
      "learning_rate": 0.0001,
      "loss": 0.3946,
      "step": 843
    },
    {
      "epoch": 3.75,
      "learning_rate": 0.0001,
      "loss": 0.3725,
      "step": 844
    },
    {
      "epoch": 3.76,
      "learning_rate": 0.0001,
      "loss": 0.3999,
      "step": 845
    },
    {
      "epoch": 3.76,
      "learning_rate": 0.0001,
      "loss": 0.4291,
      "step": 846
    },
    {
      "epoch": 3.76,
      "learning_rate": 0.0001,
      "loss": 0.4719,
      "step": 847
    },
    {
      "epoch": 3.77,
      "learning_rate": 0.0001,
      "loss": 0.4469,
      "step": 848
    },
    {
      "epoch": 3.77,
      "learning_rate": 0.0001,
      "loss": 0.4611,
      "step": 849
    },
    {
      "epoch": 3.78,
      "learning_rate": 0.0001,
      "loss": 0.3809,
      "step": 850
    },
    {
      "epoch": 3.78,
      "learning_rate": 0.0001,
      "loss": 0.3868,
      "step": 851
    },
    {
      "epoch": 3.79,
      "learning_rate": 0.0001,
      "loss": 0.3798,
      "step": 852
    },
    {
      "epoch": 3.79,
      "learning_rate": 0.0001,
      "loss": 0.4012,
      "step": 853
    },
    {
      "epoch": 3.8,
      "learning_rate": 0.0001,
      "loss": 0.488,
      "step": 854
    },
    {
      "epoch": 3.8,
      "learning_rate": 0.0001,
      "loss": 0.4302,
      "step": 855
    },
    {
      "epoch": 3.8,
      "learning_rate": 0.0001,
      "loss": 0.4116,
      "step": 856
    },
    {
      "epoch": 3.81,
      "learning_rate": 0.0001,
      "loss": 0.383,
      "step": 857
    },
    {
      "epoch": 3.81,
      "learning_rate": 0.0001,
      "loss": 0.4067,
      "step": 858
    },
    {
      "epoch": 3.82,
      "learning_rate": 0.0001,
      "loss": 0.3928,
      "step": 859
    },
    {
      "epoch": 3.82,
      "learning_rate": 0.0001,
      "loss": 0.4219,
      "step": 860
    },
    {
      "epoch": 3.82,
      "eval_loss": 0.5652073621749878,
      "eval_runtime": 24.0769,
      "eval_samples_per_second": 3.364,
      "eval_steps_per_second": 0.872,
      "step": 860
    },
    {
      "epoch": 3.83,
      "learning_rate": 0.0001,
      "loss": 0.4008,
      "step": 861
    },
    {
      "epoch": 3.83,
      "learning_rate": 0.0001,
      "loss": 0.4098,
      "step": 862
    },
    {
      "epoch": 3.84,
      "learning_rate": 0.0001,
      "loss": 0.4219,
      "step": 863
    },
    {
      "epoch": 3.84,
      "learning_rate": 0.0001,
      "loss": 0.4334,
      "step": 864
    },
    {
      "epoch": 3.84,
      "learning_rate": 0.0001,
      "loss": 0.381,
      "step": 865
    },
    {
      "epoch": 3.85,
      "learning_rate": 0.0001,
      "loss": 0.383,
      "step": 866
    },
    {
      "epoch": 3.85,
      "learning_rate": 0.0001,
      "loss": 0.3759,
      "step": 867
    },
    {
      "epoch": 3.86,
      "learning_rate": 0.0001,
      "loss": 0.3683,
      "step": 868
    },
    {
      "epoch": 3.86,
      "learning_rate": 0.0001,
      "loss": 0.3264,
      "step": 869
    },
    {
      "epoch": 3.87,
      "learning_rate": 0.0001,
      "loss": 0.3933,
      "step": 870
    },
    {
      "epoch": 3.87,
      "learning_rate": 0.0001,
      "loss": 0.3305,
      "step": 871
    },
    {
      "epoch": 3.88,
      "learning_rate": 0.0001,
      "loss": 0.3587,
      "step": 872
    },
    {
      "epoch": 3.88,
      "learning_rate": 0.0001,
      "loss": 0.3626,
      "step": 873
    },
    {
      "epoch": 3.88,
      "learning_rate": 0.0001,
      "loss": 0.3271,
      "step": 874
    },
    {
      "epoch": 3.89,
      "learning_rate": 0.0001,
      "loss": 0.3471,
      "step": 875
    },
    {
      "epoch": 3.89,
      "learning_rate": 0.0001,
      "loss": 0.4683,
      "step": 876
    },
    {
      "epoch": 3.9,
      "learning_rate": 0.0001,
      "loss": 0.4862,
      "step": 877
    },
    {
      "epoch": 3.9,
      "learning_rate": 0.0001,
      "loss": 0.4679,
      "step": 878
    },
    {
      "epoch": 3.91,
      "learning_rate": 0.0001,
      "loss": 0.5138,
      "step": 879
    },
    {
      "epoch": 3.91,
      "learning_rate": 0.0001,
      "loss": 0.5079,
      "step": 880
    },
    {
      "epoch": 3.91,
      "eval_loss": 0.5725839138031006,
      "eval_runtime": 24.0962,
      "eval_samples_per_second": 3.362,
      "eval_steps_per_second": 0.872,
      "step": 880
    },
    {
      "epoch": 3.92,
      "learning_rate": 0.0001,
      "loss": 0.4478,
      "step": 881
    },
    {
      "epoch": 3.92,
      "learning_rate": 0.0001,
      "loss": 0.4254,
      "step": 882
    },
    {
      "epoch": 3.92,
      "learning_rate": 0.0001,
      "loss": 0.4443,
      "step": 883
    },
    {
      "epoch": 3.93,
      "learning_rate": 0.0001,
      "loss": 0.4429,
      "step": 884
    },
    {
      "epoch": 3.93,
      "learning_rate": 0.0001,
      "loss": 0.4355,
      "step": 885
    },
    {
      "epoch": 3.94,
      "learning_rate": 0.0001,
      "loss": 0.4033,
      "step": 886
    },
    {
      "epoch": 3.94,
      "learning_rate": 0.0001,
      "loss": 0.4727,
      "step": 887
    },
    {
      "epoch": 3.95,
      "learning_rate": 0.0001,
      "loss": 0.3945,
      "step": 888
    },
    {
      "epoch": 3.95,
      "learning_rate": 0.0001,
      "loss": 0.4685,
      "step": 889
    },
    {
      "epoch": 3.96,
      "learning_rate": 0.0001,
      "loss": 0.3842,
      "step": 890
    },
    {
      "epoch": 3.96,
      "learning_rate": 0.0001,
      "loss": 0.4557,
      "step": 891
    },
    {
      "epoch": 3.96,
      "learning_rate": 0.0001,
      "loss": 0.3775,
      "step": 892
    },
    {
      "epoch": 3.97,
      "learning_rate": 0.0001,
      "loss": 0.3677,
      "step": 893
    },
    {
      "epoch": 3.97,
      "learning_rate": 0.0001,
      "loss": 0.3936,
      "step": 894
    },
    {
      "epoch": 3.98,
      "learning_rate": 0.0001,
      "loss": 0.42,
      "step": 895
    },
    {
      "epoch": 3.98,
      "learning_rate": 0.0001,
      "loss": 0.3614,
      "step": 896
    },
    {
      "epoch": 3.99,
      "learning_rate": 0.0001,
      "loss": 0.399,
      "step": 897
    },
    {
      "epoch": 3.99,
      "learning_rate": 0.0001,
      "loss": 0.3546,
      "step": 898
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.0001,
      "loss": 0.3126,
      "step": 899
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.0001,
      "loss": 0.3412,
      "step": 900
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.5830541253089905,
      "eval_runtime": 24.0906,
      "eval_samples_per_second": 3.362,
      "eval_steps_per_second": 0.872,
      "step": 900
    }
  ],
  "logging_steps": 1,
  "max_steps": 2250,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 20,
  "total_flos": 2.8312765231398912e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}