{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 20,
  "global_step": 699,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1.3333333333333334e-06,
      "loss": 3.6582,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 2.666666666666667e-06,
      "loss": 2.2305,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.5487,
      "step": 3
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.333333333333334e-06,
      "loss": 3.6515,
      "step": 4
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.666666666666667e-06,
      "loss": 3.7426,
      "step": 5
    },
    {
      "epoch": 0.01,
      "learning_rate": 8.000000000000001e-06,
      "loss": 3.4014,
      "step": 6
    },
    {
      "epoch": 0.01,
      "learning_rate": 9.333333333333334e-06,
      "loss": 3.2163,
      "step": 7
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.0666666666666667e-05,
      "loss": 3.823,
      "step": 8
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.2e-05,
      "loss": 4.3112,
      "step": 9
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 2.5267,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.4666666666666668e-05,
      "loss": 3.2481,
      "step": 11
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 2.4184,
      "step": 12
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.7333333333333336e-05,
      "loss": 2.7559,
      "step": 13
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.866666666666667e-05,
      "loss": 1.4468,
      "step": 14
    },
    {
      "epoch": 0.02,
      "learning_rate": 2e-05,
      "loss": 1.4686,
      "step": 15
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.1333333333333335e-05,
      "loss": 1.1665,
      "step": 16
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.2666666666666668e-05,
      "loss": 0.9044,
      "step": 17
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.4e-05,
      "loss": 0.9294,
      "step": 18
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.5333333333333337e-05,
      "loss": 0.8645,
      "step": 19
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 0.8973,
      "step": 20
    },
    {
      "epoch": 0.03,
      "eval_loss": 0.7028617262840271,
      "eval_runtime": 76.0257,
      "eval_samples_per_second": 21.427,
      "eval_steps_per_second": 10.72,
      "step": 20
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 0.8187,
      "step": 21
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.9333333333333336e-05,
      "loss": 0.9868,
      "step": 22
    },
    {
      "epoch": 0.03,
      "learning_rate": 3.066666666666667e-05,
      "loss": 0.8922,
      "step": 23
    },
    {
      "epoch": 0.03,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 0.8536,
      "step": 24
    },
    {
      "epoch": 0.04,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.9366,
      "step": 25
    },
    {
      "epoch": 0.04,
      "learning_rate": 3.466666666666667e-05,
      "loss": 1.0178,
      "step": 26
    },
    {
      "epoch": 0.04,
      "learning_rate": 3.6e-05,
      "loss": 1.1587,
      "step": 27
    },
    {
      "epoch": 0.04,
      "learning_rate": 3.733333333333334e-05,
      "loss": 0.7203,
      "step": 28
    },
    {
      "epoch": 0.04,
      "learning_rate": 3.866666666666667e-05,
      "loss": 0.6773,
      "step": 29
    },
    {
      "epoch": 0.04,
      "learning_rate": 4e-05,
      "loss": 1.0454,
      "step": 30
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.133333333333333e-05,
      "loss": 0.8861,
      "step": 31
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.266666666666667e-05,
      "loss": 0.9398,
      "step": 32
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.9018,
      "step": 33
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.5333333333333335e-05,
      "loss": 0.9051,
      "step": 34
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.666666666666667e-05,
      "loss": 0.7858,
      "step": 35
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.8e-05,
      "loss": 0.6311,
      "step": 36
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.933333333333334e-05,
      "loss": 0.7227,
      "step": 37
    },
    {
      "epoch": 0.05,
      "learning_rate": 5.0666666666666674e-05,
      "loss": 0.8281,
      "step": 38
    },
    {
      "epoch": 0.06,
      "learning_rate": 5.2000000000000004e-05,
      "loss": 0.7919,
      "step": 39
    },
    {
      "epoch": 0.06,
      "learning_rate": 5.333333333333333e-05,
      "loss": 0.6828,
      "step": 40
    },
    {
      "epoch": 0.06,
      "eval_loss": 0.6520687341690063,
      "eval_runtime": 75.9058,
      "eval_samples_per_second": 21.461,
      "eval_steps_per_second": 10.737,
      "step": 40
    },
    {
      "epoch": 0.06,
      "learning_rate": 5.466666666666666e-05,
      "loss": 0.9088,
      "step": 41
    },
    {
      "epoch": 0.06,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 0.6432,
      "step": 42
    },
    {
      "epoch": 0.06,
      "learning_rate": 5.7333333333333336e-05,
      "loss": 0.749,
      "step": 43
    },
    {
      "epoch": 0.06,
      "learning_rate": 5.866666666666667e-05,
      "loss": 0.7316,
      "step": 44
    },
    {
      "epoch": 0.06,
      "learning_rate": 6e-05,
      "loss": 0.6837,
      "step": 45
    },
    {
      "epoch": 0.07,
      "learning_rate": 6.133333333333334e-05,
      "loss": 0.717,
      "step": 46
    },
    {
      "epoch": 0.07,
      "learning_rate": 6.266666666666667e-05,
      "loss": 0.728,
      "step": 47
    },
    {
      "epoch": 0.07,
      "learning_rate": 6.400000000000001e-05,
      "loss": 0.8732,
      "step": 48
    },
    {
      "epoch": 0.07,
      "learning_rate": 6.533333333333334e-05,
      "loss": 0.6397,
      "step": 49
    },
    {
      "epoch": 0.07,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.654,
      "step": 50
    },
    {
      "epoch": 0.07,
      "learning_rate": 6.800000000000001e-05,
      "loss": 0.7031,
      "step": 51
    },
    {
      "epoch": 0.07,
      "learning_rate": 6.933333333333334e-05,
      "loss": 0.7092,
      "step": 52
    },
    {
      "epoch": 0.08,
      "learning_rate": 7.066666666666667e-05,
      "loss": 0.9424,
      "step": 53
    },
    {
      "epoch": 0.08,
      "learning_rate": 7.2e-05,
      "loss": 0.7815,
      "step": 54
    },
    {
      "epoch": 0.08,
      "learning_rate": 7.333333333333333e-05,
      "loss": 0.6843,
      "step": 55
    },
    {
      "epoch": 0.08,
      "learning_rate": 7.466666666666667e-05,
      "loss": 0.6269,
      "step": 56
    },
    {
      "epoch": 0.08,
      "learning_rate": 7.6e-05,
      "loss": 0.6285,
      "step": 57
    },
    {
      "epoch": 0.08,
      "learning_rate": 7.733333333333333e-05,
      "loss": 0.7555,
      "step": 58
    },
    {
      "epoch": 0.08,
      "learning_rate": 7.866666666666666e-05,
      "loss": 0.6724,
      "step": 59
    },
    {
      "epoch": 0.09,
      "learning_rate": 8e-05,
      "loss": 0.6521,
      "step": 60
    },
    {
      "epoch": 0.09,
      "eval_loss": 0.6198899149894714,
      "eval_runtime": 75.9771,
      "eval_samples_per_second": 21.441,
      "eval_steps_per_second": 10.727,
      "step": 60
    },
    {
      "epoch": 0.09,
      "learning_rate": 8.133333333333334e-05,
      "loss": 0.7455,
      "step": 61
    },
    {
      "epoch": 0.09,
      "learning_rate": 8.266666666666667e-05,
      "loss": 0.7496,
      "step": 62
    },
    {
      "epoch": 0.09,
      "learning_rate": 8.4e-05,
      "loss": 0.702,
      "step": 63
    },
    {
      "epoch": 0.09,
      "learning_rate": 8.533333333333334e-05,
      "loss": 0.8785,
      "step": 64
    },
    {
      "epoch": 0.09,
      "learning_rate": 8.666666666666667e-05,
      "loss": 0.7184,
      "step": 65
    },
    {
      "epoch": 0.09,
      "learning_rate": 8.800000000000001e-05,
      "loss": 0.7102,
      "step": 66
    },
    {
      "epoch": 0.1,
      "learning_rate": 8.933333333333334e-05,
      "loss": 0.795,
      "step": 67
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.066666666666667e-05,
      "loss": 0.5903,
      "step": 68
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.200000000000001e-05,
      "loss": 0.5926,
      "step": 69
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.333333333333334e-05,
      "loss": 0.6543,
      "step": 70
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.466666666666667e-05,
      "loss": 0.5384,
      "step": 71
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.6e-05,
      "loss": 0.7371,
      "step": 72
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.733333333333335e-05,
      "loss": 0.6365,
      "step": 73
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.866666666666668e-05,
      "loss": 0.7859,
      "step": 74
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0001,
      "loss": 0.5836,
      "step": 75
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00010133333333333335,
      "loss": 0.5666,
      "step": 76
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00010266666666666666,
      "loss": 0.4644,
      "step": 77
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00010400000000000001,
      "loss": 0.9123,
      "step": 78
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00010533333333333332,
      "loss": 0.7794,
      "step": 79
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00010666666666666667,
      "loss": 0.7857,
      "step": 80
    },
    {
      "epoch": 0.11,
      "eval_loss": 0.6065900325775146,
      "eval_runtime": 75.8059,
      "eval_samples_per_second": 21.489,
      "eval_steps_per_second": 10.751,
      "step": 80
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00010800000000000001,
      "loss": 0.7882,
      "step": 81
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00010933333333333333,
      "loss": 0.722,
      "step": 82
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00011066666666666667,
      "loss": 0.6808,
      "step": 83
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00011200000000000001,
      "loss": 0.7176,
      "step": 84
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00011333333333333334,
      "loss": 0.9055,
      "step": 85
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00011466666666666667,
      "loss": 0.6622,
      "step": 86
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.000116,
      "loss": 0.6591,
      "step": 87
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00011733333333333334,
      "loss": 0.6363,
      "step": 88
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00011866666666666669,
      "loss": 0.747,
      "step": 89
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00012,
      "loss": 0.609,
      "step": 90
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00012133333333333335,
      "loss": 0.8201,
      "step": 91
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00012266666666666668,
      "loss": 0.7042,
      "step": 92
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.000124,
      "loss": 0.5947,
      "step": 93
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00012533333333333334,
      "loss": 0.8148,
      "step": 94
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00012666666666666666,
      "loss": 0.5936,
      "step": 95
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00012800000000000002,
      "loss": 0.7983,
      "step": 96
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00012933333333333332,
      "loss": 0.907,
      "step": 97
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00013066666666666668,
      "loss": 0.8807,
      "step": 98
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.000132,
      "loss": 0.626,
      "step": 99
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.6208,
      "step": 100
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.6062822937965393,
      "eval_runtime": 76.0845,
      "eval_samples_per_second": 21.41,
      "eval_steps_per_second": 10.712,
      "step": 100
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00013466666666666667,
      "loss": 0.5916,
      "step": 101
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00013600000000000003,
      "loss": 0.6769,
      "step": 102
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00013733333333333333,
      "loss": 0.6558,
      "step": 103
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00013866666666666669,
      "loss": 0.6152,
      "step": 104
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00014,
      "loss": 0.6301,
      "step": 105
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00014133333333333334,
      "loss": 0.6504,
      "step": 106
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00014266666666666667,
      "loss": 0.6567,
      "step": 107
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.000144,
      "loss": 0.8578,
      "step": 108
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00014533333333333333,
      "loss": 0.7125,
      "step": 109
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00014666666666666666,
      "loss": 0.6192,
      "step": 110
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.000148,
      "loss": 0.6424,
      "step": 111
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00014933333333333335,
      "loss": 0.5399,
      "step": 112
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00015066666666666668,
      "loss": 0.6844,
      "step": 113
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.000152,
      "loss": 0.6255,
      "step": 114
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00015333333333333334,
      "loss": 0.7753,
      "step": 115
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00015466666666666667,
      "loss": 0.7796,
      "step": 116
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00015600000000000002,
      "loss": 0.6181,
      "step": 117
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00015733333333333333,
      "loss": 0.7465,
      "step": 118
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00015866666666666668,
      "loss": 0.7195,
      "step": 119
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00016,
      "loss": 0.6805,
      "step": 120
    },
    {
      "epoch": 0.17,
      "eval_loss": 0.5969421863555908,
      "eval_runtime": 75.9286,
      "eval_samples_per_second": 21.454,
      "eval_steps_per_second": 10.734,
      "step": 120
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00016133333333333334,
      "loss": 0.6499,
      "step": 121
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00016266666666666667,
      "loss": 0.5963,
      "step": 122
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.000164,
      "loss": 0.6295,
      "step": 123
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00016533333333333333,
      "loss": 0.7473,
      "step": 124
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001666666666666667,
      "loss": 0.65,
      "step": 125
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.000168,
      "loss": 0.5558,
      "step": 126
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00016933333333333335,
      "loss": 0.5604,
      "step": 127
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00017066666666666668,
      "loss": 0.5784,
      "step": 128
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.000172,
      "loss": 0.5938,
      "step": 129
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00017333333333333334,
      "loss": 0.8228,
      "step": 130
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00017466666666666667,
      "loss": 0.7301,
      "step": 131
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00017600000000000002,
      "loss": 0.863,
      "step": 132
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00017733333333333335,
      "loss": 0.5386,
      "step": 133
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00017866666666666668,
      "loss": 0.4405,
      "step": 134
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00018,
      "loss": 0.5679,
      "step": 135
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00018133333333333334,
      "loss": 0.718,
      "step": 136
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00018266666666666667,
      "loss": 0.6935,
      "step": 137
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00018400000000000003,
      "loss": 0.629,
      "step": 138
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00018533333333333333,
      "loss": 0.7571,
      "step": 139
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001866666666666667,
      "loss": 0.5928,
      "step": 140
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.5988821983337402,
      "eval_runtime": 76.0106,
      "eval_samples_per_second": 21.431,
      "eval_steps_per_second": 10.722,
      "step": 140
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.000188,
      "loss": 0.4976,
      "step": 141
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00018933333333333335,
      "loss": 0.5056,
      "step": 142
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019066666666666668,
      "loss": 0.688,
      "step": 143
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.000192,
      "loss": 0.618,
      "step": 144
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00019333333333333333,
      "loss": 0.5207,
      "step": 145
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001946666666666667,
      "loss": 0.45,
      "step": 146
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.000196,
      "loss": 0.6393,
      "step": 147
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00019733333333333335,
      "loss": 0.6003,
      "step": 148
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00019866666666666668,
      "loss": 0.8397,
      "step": 149
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002,
      "loss": 0.6742,
      "step": 150
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00020133333333333334,
      "loss": 0.5273,
      "step": 151
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002026666666666667,
      "loss": 0.5878,
      "step": 152
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00020400000000000003,
      "loss": 0.6262,
      "step": 153
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00020533333333333333,
      "loss": 0.9558,
      "step": 154
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00020666666666666668,
      "loss": 0.563,
      "step": 155
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00020800000000000001,
      "loss": 0.4969,
      "step": 156
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00020933333333333334,
      "loss": 0.6478,
      "step": 157
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00021066666666666665,
      "loss": 0.67,
      "step": 158
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00021200000000000003,
      "loss": 0.6035,
      "step": 159
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00021333333333333333,
      "loss": 0.715,
      "step": 160
    },
    {
      "epoch": 0.23,
      "eval_loss": 0.5843603014945984,
      "eval_runtime": 76.0289,
      "eval_samples_per_second": 21.426,
      "eval_steps_per_second": 10.72,
      "step": 160
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00021466666666666666,
      "loss": 0.5748,
      "step": 161
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00021600000000000002,
      "loss": 0.5162,
      "step": 162
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00021733333333333335,
      "loss": 0.4678,
      "step": 163
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00021866666666666665,
      "loss": 0.5409,
      "step": 164
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00022000000000000003,
      "loss": 0.6638,
      "step": 165
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00022133333333333334,
      "loss": 0.6777,
      "step": 166
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00022266666666666667,
      "loss": 0.5909,
      "step": 167
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00022400000000000002,
      "loss": 0.767,
      "step": 168
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00022533333333333335,
      "loss": 0.7222,
      "step": 169
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00022666666666666668,
      "loss": 0.4992,
      "step": 170
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00022799999999999999,
      "loss": 0.5737,
      "step": 171
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00022933333333333334,
      "loss": 0.4818,
      "step": 172
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00023066666666666667,
      "loss": 0.5393,
      "step": 173
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.000232,
      "loss": 0.4846,
      "step": 174
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00023333333333333336,
      "loss": 0.5463,
      "step": 175
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002346666666666667,
      "loss": 0.7732,
      "step": 176
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.000236,
      "loss": 0.723,
      "step": 177
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00023733333333333337,
      "loss": 0.5236,
      "step": 178
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00023866666666666668,
      "loss": 0.5844,
      "step": 179
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00024,
      "loss": 0.5647,
      "step": 180
    },
    {
      "epoch": 0.26,
      "eval_loss": 0.5979331731796265,
      "eval_runtime": 75.9556,
      "eval_samples_per_second": 21.447,
      "eval_steps_per_second": 10.73,
      "step": 180
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00024133333333333336,
      "loss": 0.5554,
      "step": 181
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002426666666666667,
      "loss": 0.6322,
      "step": 182
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.000244,
      "loss": 0.6285,
      "step": 183
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00024533333333333335,
      "loss": 0.6452,
      "step": 184
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002466666666666667,
      "loss": 0.663,
      "step": 185
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.000248,
      "loss": 0.4447,
      "step": 186
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00024933333333333334,
      "loss": 0.5616,
      "step": 187
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025066666666666667,
      "loss": 0.6607,
      "step": 188
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.000252,
      "loss": 0.5731,
      "step": 189
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025333333333333333,
      "loss": 0.8153,
      "step": 190
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002546666666666667,
      "loss": 0.7403,
      "step": 191
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00025600000000000004,
      "loss": 0.5603,
      "step": 192
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002573333333333333,
      "loss": 0.5311,
      "step": 193
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00025866666666666665,
      "loss": 0.8732,
      "step": 194
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00026000000000000003,
      "loss": 0.627,
      "step": 195
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00026133333333333336,
      "loss": 0.9056,
      "step": 196
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00026266666666666664,
      "loss": 0.4675,
      "step": 197
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.000264,
      "loss": 0.5102,
      "step": 198
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00026533333333333335,
      "loss": 0.6037,
      "step": 199
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002666666666666667,
      "loss": 0.6778,
      "step": 200
    },
    {
      "epoch": 0.29,
      "eval_loss": 0.5888569355010986,
      "eval_runtime": 75.9422,
      "eval_samples_per_second": 21.451,
      "eval_steps_per_second": 10.732,
      "step": 200
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.000268,
      "loss": 0.6108,
      "step": 201
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00026933333333333334,
      "loss": 0.5723,
      "step": 202
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00027066666666666667,
      "loss": 0.6239,
      "step": 203
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00027200000000000005,
      "loss": 0.6304,
      "step": 204
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00027333333333333333,
      "loss": 0.7902,
      "step": 205
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00027466666666666666,
      "loss": 0.5582,
      "step": 206
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.000276,
      "loss": 0.5657,
      "step": 207
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00027733333333333337,
      "loss": 0.6505,
      "step": 208
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002786666666666667,
      "loss": 0.5841,
      "step": 209
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00028,
      "loss": 0.5383,
      "step": 210
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00028133333333333336,
      "loss": 0.4945,
      "step": 211
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002826666666666667,
      "loss": 0.5968,
      "step": 212
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.000284,
      "loss": 0.6705,
      "step": 213
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00028533333333333335,
      "loss": 0.571,
      "step": 214
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002866666666666667,
      "loss": 0.5133,
      "step": 215
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.000288,
      "loss": 0.6208,
      "step": 216
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002893333333333334,
      "loss": 0.6128,
      "step": 217
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00029066666666666667,
      "loss": 0.7007,
      "step": 218
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.000292,
      "loss": 0.5285,
      "step": 219
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002933333333333333,
      "loss": 0.5907,
      "step": 220
    },
    {
      "epoch": 0.31,
      "eval_loss": 0.5771830081939697,
      "eval_runtime": 75.9871,
      "eval_samples_per_second": 21.438,
      "eval_steps_per_second": 10.725,
      "step": 220
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002946666666666667,
      "loss": 0.5352,
      "step": 221
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.000296,
      "loss": 0.5487,
      "step": 222
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002973333333333333,
      "loss": 0.7642,
      "step": 223
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002986666666666667,
      "loss": 0.4925,
      "step": 224
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00030000000000000003,
      "loss": 0.6328,
      "step": 225
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00030133333333333336,
      "loss": 0.8737,
      "step": 226
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0003026666666666667,
      "loss": 0.7009,
      "step": 227
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.000304,
      "loss": 0.6476,
      "step": 228
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00030533333333333335,
      "loss": 0.5962,
      "step": 229
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0003066666666666667,
      "loss": 0.4297,
      "step": 230
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.000308,
      "loss": 0.4098,
      "step": 231
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00030933333333333334,
      "loss": 0.466,
      "step": 232
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00031066666666666666,
      "loss": 0.557,
      "step": 233
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00031200000000000005,
      "loss": 0.5769,
      "step": 234
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0003133333333333333,
      "loss": 0.6945,
      "step": 235
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00031466666666666665,
      "loss": 0.4355,
      "step": 236
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00031600000000000004,
      "loss": 0.6799,
      "step": 237
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00031733333333333337,
      "loss": 0.5999,
      "step": 238
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00031866666666666664,
      "loss": 0.4519,
      "step": 239
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00032,
      "loss": 0.5536,
      "step": 240
    },
    {
      "epoch": 0.34,
      "eval_loss": 0.5916756987571716,
      "eval_runtime": 75.9583,
      "eval_samples_per_second": 21.446,
      "eval_steps_per_second": 10.73,
      "step": 240
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00032133333333333336,
      "loss": 0.5568,
      "step": 241
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0003226666666666667,
      "loss": 0.6273,
      "step": 242
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.000324,
      "loss": 0.4417,
      "step": 243
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00032533333333333334,
      "loss": 0.5871,
      "step": 244
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0003266666666666667,
      "loss": 0.5515,
      "step": 245
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.000328,
      "loss": 0.5438,
      "step": 246
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00032933333333333333,
      "loss": 0.6791,
      "step": 247
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00033066666666666666,
      "loss": 0.5649,
      "step": 248
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.000332,
      "loss": 0.8795,
      "step": 249
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0003333333333333334,
      "loss": 0.5306,
      "step": 250
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0003346666666666667,
      "loss": 0.5975,
      "step": 251
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.000336,
      "loss": 0.6095,
      "step": 252
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00033733333333333337,
      "loss": 0.5929,
      "step": 253
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0003386666666666667,
      "loss": 0.581,
      "step": 254
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00034,
      "loss": 0.5226,
      "step": 255
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00034133333333333335,
      "loss": 0.6894,
      "step": 256
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0003426666666666667,
      "loss": 0.5807,
      "step": 257
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.000344,
      "loss": 0.5059,
      "step": 258
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00034533333333333334,
      "loss": 0.67,
      "step": 259
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00034666666666666667,
      "loss": 0.7422,
      "step": 260
    },
    {
      "epoch": 0.37,
      "eval_loss": 0.6780916452407837,
      "eval_runtime": 75.9236,
      "eval_samples_per_second": 21.456,
      "eval_steps_per_second": 10.734,
      "step": 260
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.000348,
      "loss": 0.5771,
      "step": 261
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00034933333333333333,
      "loss": 0.525,
      "step": 262
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0003506666666666667,
      "loss": 0.5917,
      "step": 263
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00035200000000000005,
      "loss": 0.5993,
      "step": 264
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0003533333333333333,
      "loss": 0.5338,
      "step": 265
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0003546666666666667,
      "loss": 0.5121,
      "step": 266
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00035600000000000003,
      "loss": 0.6388,
      "step": 267
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00035733333333333336,
      "loss": 0.764,
      "step": 268
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00035866666666666664,
      "loss": 0.5963,
      "step": 269
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00036,
      "loss": 0.565,
      "step": 270
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00036133333333333335,
      "loss": 0.6174,
      "step": 271
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0003626666666666667,
      "loss": 0.629,
      "step": 272
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.000364,
      "loss": 0.5634,
      "step": 273
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00036533333333333334,
      "loss": 0.535,
      "step": 274
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00036666666666666667,
      "loss": 0.654,
      "step": 275
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00036800000000000005,
      "loss": 0.6187,
      "step": 276
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00036933333333333333,
      "loss": 0.4855,
      "step": 277
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00037066666666666666,
      "loss": 0.4726,
      "step": 278
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00037200000000000004,
      "loss": 0.5277,
      "step": 279
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0003733333333333334,
      "loss": 0.6328,
      "step": 280
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.5785369277000427,
      "eval_runtime": 76.0263,
      "eval_samples_per_second": 21.427,
      "eval_steps_per_second": 10.72,
      "step": 280
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0003746666666666667,
      "loss": 0.6309,
      "step": 281
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.000376,
      "loss": 0.7785,
      "step": 282
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00037733333333333336,
      "loss": 0.4615,
      "step": 283
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0003786666666666667,
      "loss": 0.7048,
      "step": 284
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00038,
      "loss": 0.8294,
      "step": 285
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00038133333333333335,
      "loss": 0.6389,
      "step": 286
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0003826666666666667,
      "loss": 0.5951,
      "step": 287
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.000384,
      "loss": 0.4733,
      "step": 288
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0003853333333333334,
      "loss": 0.5251,
      "step": 289
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00038666666666666667,
      "loss": 0.5581,
      "step": 290
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.000388,
      "loss": 0.5923,
      "step": 291
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0003893333333333334,
      "loss": 0.5028,
      "step": 292
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0003906666666666667,
      "loss": 0.4401,
      "step": 293
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.000392,
      "loss": 0.8113,
      "step": 294
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0003933333333333333,
      "loss": 0.621,
      "step": 295
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0003946666666666667,
      "loss": 0.6678,
      "step": 296
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00039600000000000003,
      "loss": 0.5044,
      "step": 297
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00039733333333333336,
      "loss": 0.5511,
      "step": 298
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0003986666666666667,
      "loss": 0.5147,
      "step": 299
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004,
      "loss": 0.5705,
      "step": 300
    },
    {
      "epoch": 0.43,
      "eval_loss": 0.5719897747039795,
      "eval_runtime": 76.0506,
      "eval_samples_per_second": 21.42,
      "eval_steps_per_second": 10.717,
      "step": 300
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0003999938005707171,
      "loss": 0.9111,
      "step": 301
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0003999752026671977,
      "loss": 0.5359,
      "step": 302
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0003999442074424055,
      "loss": 0.7822,
      "step": 303
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00039990081681786775,
      "loss": 0.5033,
      "step": 304
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00039984503348355536,
      "loss": 0.4551,
      "step": 305
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0003997768608977168,
      "loss": 0.6405,
      "step": 306
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00039969630328666326,
      "loss": 0.4667,
      "step": 307
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0003996033656445069,
      "loss": 0.5991,
      "step": 308
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00039949805373285115,
      "loss": 0.5846,
      "step": 309
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00039938037408043353,
      "loss": 0.6403,
      "step": 310
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00039925033398272074,
      "loss": 0.5268,
      "step": 311
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00039910794150145677,
      "loss": 0.7475,
      "step": 312
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0003989532054641627,
      "loss": 0.5654,
      "step": 313
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00039878613546358993,
      "loss": 0.601,
      "step": 314
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00039860674185712486,
      "loss": 0.5911,
      "step": 315
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0003984150357661472,
      "loss": 0.5018,
      "step": 316
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0003982110290753407,
      "loss": 0.6736,
      "step": 317
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0003979947344319557,
      "loss": 0.5516,
      "step": 318
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0003977661652450257,
      "loss": 0.5712,
      "step": 319
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0003975253356845359,
      "loss": 0.6124,
      "step": 320
    },
    {
      "epoch": 0.46,
      "eval_loss": 0.5753471255302429,
      "eval_runtime": 75.9905,
      "eval_samples_per_second": 21.437,
      "eval_steps_per_second": 10.725,
      "step": 320
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0003972722606805445,
      "loss": 0.6516,
      "step": 321
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0003970069559222574,
      "loss": 0.8156,
      "step": 322
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00039672943785705553,
      "loss": 0.6279,
      "step": 323
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0003964397236894751,
      "loss": 0.7026,
      "step": 324
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00039613783138014097,
      "loss": 0.5975,
      "step": 325
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0003958237796446534,
      "loss": 0.5961,
      "step": 326
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0003954975879524277,
      "loss": 0.5417,
      "step": 327
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00039515927652548715,
      "loss": 0.7113,
      "step": 328
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00039480886633720935,
      "loss": 0.5137,
      "step": 329
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0003944463791110261,
      "loss": 0.5216,
      "step": 330
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0003940718373190768,
      "loss": 0.4807,
      "step": 331
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0003936852641808148,
      "loss": 0.5484,
      "step": 332
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00039328668366156856,
      "loss": 0.523,
      "step": 333
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0003928761204710554,
      "loss": 0.6463,
      "step": 334
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00039245360006185005,
      "loss": 0.6086,
      "step": 335
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00039201914862780653,
      "loss": 0.6831,
      "step": 336
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0003915727931024342,
      "loss": 0.446,
      "step": 337
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0003911145611572282,
      "loss": 0.73,
      "step": 338
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00039064448119995385,
      "loss": 0.5218,
      "step": 339
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00039016258237288586,
      "loss": 0.4735,
      "step": 340
    },
    {
      "epoch": 0.49,
      "eval_loss": 0.6202576160430908,
      "eval_runtime": 75.948,
      "eval_samples_per_second": 21.449,
      "eval_steps_per_second": 10.731,
      "step": 340
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0003896688945510011,
      "loss": 0.6596,
      "step": 341
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00038916344834012695,
      "loss": 0.4636,
      "step": 342
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0003886462750750439,
      "loss": 0.6812,
      "step": 343
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00038811740681754276,
      "loss": 0.435,
      "step": 344
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0003875768763544371,
      "loss": 0.6217,
      "step": 345
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0003870247171955308,
      "loss": 0.6063,
      "step": 346
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00038646096357154036,
      "loss": 0.8118,
      "step": 347
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0003858856504319732,
      "loss": 0.6175,
      "step": 348
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00038529881344296035,
      "loss": 0.6884,
      "step": 349
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0003847004889850461,
      "loss": 0.6609,
      "step": 350
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00038409071415093204,
      "loss": 0.6236,
      "step": 351
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00038346952674317783,
      "loss": 0.3354,
      "step": 352
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00038283696527185746,
      "loss": 0.6189,
      "step": 353
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0003821930689521721,
      "loss": 0.4758,
      "step": 354
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00038153787770201875,
      "loss": 0.6045,
      "step": 355
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00038087143213951554,
      "loss": 0.7085,
      "step": 356
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0003801937735804838,
      "loss": 0.8512,
      "step": 357
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0003795049440358868,
      "loss": 0.6723,
      "step": 358
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00037880498620922507,
      "loss": 0.5035,
      "step": 359
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00037809394349388895,
      "loss": 0.4602,
      "step": 360
    },
    {
      "epoch": 0.52,
      "eval_loss": 0.5771754384040833,
      "eval_runtime": 75.9251,
      "eval_samples_per_second": 21.455,
      "eval_steps_per_second": 10.734,
      "step": 360
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00037737185997046874,
      "loss": 0.5802,
      "step": 361
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00037663878040402194,
      "loss": 0.5308,
      "step": 362
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0003758947502412978,
      "loss": 0.527,
      "step": 363
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00037513981560792015,
      "loss": 0.6222,
      "step": 364
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00037437402330552766,
      "loss": 0.544,
      "step": 365
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00037359742080887265,
      "loss": 0.6717,
      "step": 366
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0003728100562628776,
      "loss": 0.5324,
      "step": 367
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00037201197847965076,
      "loss": 0.4345,
      "step": 368
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00037120323693546004,
      "loss": 0.7623,
      "step": 369
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00037038388176766546,
      "loss": 0.6668,
      "step": 370
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0003695539637716112,
      "loss": 0.6174,
      "step": 371
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0003687135343974766,
      "loss": 0.388,
      "step": 372
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0003678626457470863,
      "loss": 0.7039,
      "step": 373
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003670013505706805,
      "loss": 0.5392,
      "step": 374
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003661297022636446,
      "loss": 0.5504,
      "step": 375
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.000365247754863199,
      "loss": 0.7533,
      "step": 376
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.000364355563045049,
      "loss": 0.4881,
      "step": 377
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00036345318211999564,
      "loss": 1.0078,
      "step": 378
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00036254066803050603,
      "loss": 0.4862,
      "step": 379
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003616180773472459,
      "loss": 0.8571,
      "step": 380
    },
    {
      "epoch": 0.54,
      "eval_loss": 0.5749652981758118,
      "eval_runtime": 75.9786,
      "eval_samples_per_second": 21.44,
      "eval_steps_per_second": 10.727,
      "step": 380
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003606854672655723,
      "loss": 0.4636,
      "step": 381
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00035974289560198756,
      "loss": 0.6757,
      "step": 382
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00035879042079055566,
      "loss": 0.3554,
      "step": 383
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00035782810187927875,
      "loss": 0.6187,
      "step": 384
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003568559985264373,
      "loss": 0.4611,
      "step": 385
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00035587417099689124,
      "loss": 0.4761,
      "step": 386
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003548826801583439,
      "loss": 0.6782,
      "step": 387
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00035388158747756866,
      "loss": 0.6693,
      "step": 388
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00035287095501659844,
      "loss": 0.5747,
      "step": 389
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00035185084542887774,
      "loss": 0.6823,
      "step": 390
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00035082132195537926,
      "loss": 0.6062,
      "step": 391
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0003497824484206826,
      "loss": 0.6462,
      "step": 392
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00034873428922901797,
      "loss": 0.6891,
      "step": 393
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0003476769093602731,
      "loss": 0.5871,
      "step": 394
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0003466103743659653,
      "loss": 0.857,
      "step": 395
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034553475036517726,
      "loss": 0.6893,
      "step": 396
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034445010404045827,
      "loss": 0.6563,
      "step": 397
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034335650263369027,
      "loss": 0.545,
      "step": 398
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0003422540139419191,
      "loss": 0.7041,
      "step": 399
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034114270631315137,
      "loss": 0.5504,
      "step": 400
    },
    {
      "epoch": 0.57,
      "eval_loss": 0.6039557456970215,
      "eval_runtime": 75.7007,
      "eval_samples_per_second": 21.519,
      "eval_steps_per_second": 10.766,
      "step": 400
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034002264864211793,
      "loss": 0.6144,
      "step": 401
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0003388939103660017,
      "loss": 0.5391,
      "step": 402
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0003377565614601343,
      "loss": 0.5719,
      "step": 403
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00033661067243365644,
      "loss": 0.4687,
      "step": 404
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00033545631432514825,
      "loss": 0.5341,
      "step": 405
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00033429355869822423,
      "loss": 0.4259,
      "step": 406
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0003331224776370972,
      "loss": 0.9081,
      "step": 407
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0003319431437421095,
      "loss": 0.4992,
      "step": 408
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00033075563012523174,
      "loss": 0.6903,
      "step": 409
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00032956001040553105,
      "loss": 0.6846,
      "step": 410
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00032835635870460635,
      "loss": 0.8808,
      "step": 411
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0003271447496419937,
      "loss": 0.7433,
      "step": 412
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0003259252583305399,
      "loss": 0.8327,
      "step": 413
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00032469796037174674,
      "loss": 0.5437,
      "step": 414
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00032346293185108306,
      "loss": 0.7766,
      "step": 415
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00032222024933326873,
      "loss": 0.5782,
      "step": 416
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00032096998985752754,
      "loss": 0.5614,
      "step": 417
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00031971223093281166,
      "loss": 0.4574,
      "step": 418
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00031844705053299605,
      "loss": 0.5042,
      "step": 419
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00031717452709204496,
      "loss": 0.6307,
      "step": 420
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.579613447189331,
      "eval_runtime": 75.9225,
      "eval_samples_per_second": 21.456,
      "eval_steps_per_second": 10.735,
      "step": 420
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0003158947394991493,
      "loss": 0.5218,
      "step": 421
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0003146077670938357,
      "loss": 0.687,
      "step": 422
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003133136896610485,
      "loss": 0.5384,
      "step": 423
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003120125874262029,
      "loss": 0.4349,
      "step": 424
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003107045410502118,
      "loss": 0.5447,
      "step": 425
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00030938963162448544,
      "loss": 0.4783,
      "step": 426
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003080679406659036,
      "loss": 0.6773,
      "step": 427
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003067395501117627,
      "loss": 0.705,
      "step": 428
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003054045423146958,
      "loss": 0.4308,
      "step": 429
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0003040630000375671,
      "loss": 0.5592,
      "step": 430
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0003027150064483414,
      "loss": 0.7458,
      "step": 431
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00030136064511492803,
      "loss": 0.5513,
      "step": 432
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00030000000000000003,
      "loss": 0.6097,
      "step": 433
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00029863315545578916,
      "loss": 0.5079,
      "step": 434
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0002972601962188563,
      "loss": 0.6411,
      "step": 435
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00029588120740483843,
      "loss": 0.4658,
      "step": 436
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002944962745031719,
      "loss": 0.4725,
      "step": 437
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00029310548337179267,
      "loss": 0.5727,
      "step": 438
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002917089202318132,
      "loss": 0.4842,
      "step": 439
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002903066716621779,
      "loss": 0.4782,
      "step": 440
    },
    {
      "epoch": 0.63,
      "eval_loss": 0.563869297504425,
      "eval_runtime": 75.946,
      "eval_samples_per_second": 21.449,
      "eval_steps_per_second": 10.731,
      "step": 440
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002888988245942952,
      "loss": 0.5742,
      "step": 441
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002874854663066483,
      "loss": 0.5106,
      "step": 442
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.000286066684419385,
      "loss": 0.7239,
      "step": 443
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.000284642566888885,
      "loss": 0.4501,
      "step": 444
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002832132020023075,
      "loss": 0.6713,
      "step": 445
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00028177867837211775,
      "loss": 0.7934,
      "step": 446
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002803390849305939,
      "loss": 0.6831,
      "step": 447
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002788945109243133,
      "loss": 0.7223,
      "step": 448
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00027744504590861976,
      "loss": 0.8328,
      "step": 449
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.000275990779742072,
      "loss": 0.4744,
      "step": 450
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00027453180258087264,
      "loss": 0.5313,
      "step": 451
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00027306820487327905,
      "loss": 0.4472,
      "step": 452
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00027160007735399597,
      "loss": 0.623,
      "step": 453
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00027012751103855094,
      "loss": 0.5294,
      "step": 454
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00026865059721765114,
      "loss": 0.6812,
      "step": 455
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002671694274515246,
      "loss": 0.5081,
      "step": 456
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00026568409356424335,
      "loss": 0.6002,
      "step": 457
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00026419468763803157,
      "loss": 0.5973,
      "step": 458
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002627013020075563,
      "loss": 0.526,
      "step": 459
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00026120402925420357,
      "loss": 0.4159,
      "step": 460
    },
    {
      "epoch": 0.66,
      "eval_loss": 0.5689454078674316,
      "eval_runtime": 76.0037,
      "eval_samples_per_second": 21.433,
      "eval_steps_per_second": 10.723,
      "step": 460
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00025970296220033894,
      "loss": 0.6363,
      "step": 461
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00025819819390355295,
      "loss": 0.4499,
      "step": 462
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.000256689817650892,
      "loss": 0.4329,
      "step": 463
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00025517792695307505,
      "loss": 0.4992,
      "step": 464
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.000253662615538697,
      "loss": 0.4652,
      "step": 465
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002521439773484172,
      "loss": 0.5941,
      "step": 466
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002506221065291364,
      "loss": 0.4763,
      "step": 467
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002490970974281599,
      "loss": 0.5573,
      "step": 468
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002475690445873483,
      "loss": 0.6524,
      "step": 469
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00024603804273725695,
      "loss": 0.8046,
      "step": 470
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002445041867912629,
      "loss": 0.4634,
      "step": 471
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002429675718396808,
      "loss": 0.9086,
      "step": 472
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00024142829314386792,
      "loss": 0.5438,
      "step": 473
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00023988644613031853,
      "loss": 0.825,
      "step": 474
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00023834212638474772,
      "loss": 0.4561,
      "step": 475
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00023679542964616618,
      "loss": 0.5544,
      "step": 476
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00023524645180094438,
      "loss": 0.5037,
      "step": 477
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00023369528887686843,
      "loss": 0.5007,
      "step": 478
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00023214203703718692,
      "loss": 0.4977,
      "step": 479
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0002305867925746492,
      "loss": 0.6393,
      "step": 480
    },
    {
      "epoch": 0.69,
      "eval_loss": 0.5660532116889954,
      "eval_runtime": 75.8556,
      "eval_samples_per_second": 21.475,
      "eval_steps_per_second": 10.744,
      "step": 480
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00022902965190553593,
      "loss": 0.8045,
      "step": 481
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00022747071156368164,
      "loss": 0.5848,
      "step": 482
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0002259100681944905,
      "loss": 0.533,
      "step": 483
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00022434781854894454,
      "loss": 0.4729,
      "step": 484
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0002227840594776056,
      "loss": 0.5992,
      "step": 485
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00022121888792461163,
      "loss": 0.4929,
      "step": 486
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00021965240092166615,
      "loss": 0.4443,
      "step": 487
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00021808469558202317,
      "loss": 0.7139,
      "step": 488
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002165158690944665,
      "loss": 0.5218,
      "step": 489
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00021494601871728487,
      "loss": 0.5698,
      "step": 490
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00021337524177224228,
      "loss": 0.4386,
      "step": 491
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00021180363563854458,
      "loss": 0.6144,
      "step": 492
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002102312977468027,
      "loss": 0.6238,
      "step": 493
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00020865832557299222,
      "loss": 0.5725,
      "step": 494
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00020708481663241082,
      "loss": 0.5518,
      "step": 495
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00020551086847363242,
      "loss": 0.6726,
      "step": 496
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00020393657867246018,
      "loss": 0.7597,
      "step": 497
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00020236204482587696,
      "loss": 0.7842,
      "step": 498
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002007873645459951,
      "loss": 0.438,
      "step": 499
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.000199212635454005,
      "loss": 0.8243,
      "step": 500
    },
    {
      "epoch": 0.72,
      "eval_loss": 0.5698496103286743,
      "eval_runtime": 75.9995,
      "eval_samples_per_second": 21.434,
      "eval_steps_per_second": 10.724,
      "step": 500
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001976379551741231,
      "loss": 0.6012,
      "step": 501
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00019606342132753992,
      "loss": 0.5885,
      "step": 502
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00019448913152636762,
      "loss": 0.5198,
      "step": 503
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00019291518336758925,
      "loss": 0.555,
      "step": 504
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001913416744270078,
      "loss": 0.5222,
      "step": 505
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001897687022531974,
      "loss": 0.4807,
      "step": 506
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00018819636436145546,
      "loss": 0.538,
      "step": 507
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00018662475822775777,
      "loss": 0.3728,
      "step": 508
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00018505398128271515,
      "loss": 0.5586,
      "step": 509
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00018348413090553354,
      "loss": 0.821,
      "step": 510
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00018191530441797688,
      "loss": 0.7031,
      "step": 511
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00018034759907833392,
      "loss": 0.5291,
      "step": 512
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00017878111207538842,
      "loss": 0.5437,
      "step": 513
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00017721594052239444,
      "loss": 0.5258,
      "step": 514
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00017565218145105553,
      "loss": 0.9788,
      "step": 515
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00017408993180550956,
      "loss": 0.5201,
      "step": 516
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00017252928843631838,
      "loss": 0.6211,
      "step": 517
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00017097034809446414,
      "loss": 0.4348,
      "step": 518
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001694132074253508,
      "loss": 0.465,
      "step": 519
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00016785796296281316,
      "loss": 0.4744,
      "step": 520
    },
    {
      "epoch": 0.74,
      "eval_loss": 0.5536297559738159,
      "eval_runtime": 75.93,
      "eval_samples_per_second": 21.454,
      "eval_steps_per_second": 10.734,
      "step": 520
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00016630471112313158,
      "loss": 0.5797,
      "step": 521
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001647535481990557,
      "loss": 0.6363,
      "step": 522
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00016320457035383384,
      "loss": 0.4378,
      "step": 523
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00016165787361525238,
      "loss": 0.5703,
      "step": 524
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00016011355386968154,
      "loss": 0.3862,
      "step": 525
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00015857170685613215,
      "loss": 0.5217,
      "step": 526
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00015703242816031922,
      "loss": 0.4774,
      "step": 527
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00015549581320873715,
      "loss": 0.8355,
      "step": 528
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00015396195726274304,
      "loss": 0.4639,
      "step": 529
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001524309554126518,
      "loss": 0.4704,
      "step": 530
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001509029025718402,
      "loss": 0.3754,
      "step": 531
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00014937789347086364,
      "loss": 0.9537,
      "step": 532
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00014785602265158281,
      "loss": 0.4096,
      "step": 533
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00014633738446130312,
      "loss": 0.7183,
      "step": 534
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.000144822073046925,
      "loss": 0.5815,
      "step": 535
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00014331018234910813,
      "loss": 0.4413,
      "step": 536
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001418018060964471,
      "loss": 0.573,
      "step": 537
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00014029703779966114,
      "loss": 0.6439,
      "step": 538
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001387959707457965,
      "loss": 0.4307,
      "step": 539
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00013729869799244382,
      "loss": 0.4395,
      "step": 540
    },
    {
      "epoch": 0.77,
      "eval_loss": 0.5535804629325867,
      "eval_runtime": 75.9708,
      "eval_samples_per_second": 21.442,
      "eval_steps_per_second": 10.728,
      "step": 540
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00013580531236196848,
      "loss": 0.3237,
      "step": 541
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00013431590643575673,
      "loss": 0.4944,
      "step": 542
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001328305725484755,
      "loss": 0.4503,
      "step": 543
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00013134940278234896,
      "loss": 0.4831,
      "step": 544
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00012987248896144914,
      "loss": 0.555,
      "step": 545
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00012839992264600407,
      "loss": 0.5443,
      "step": 546
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.000126931795126721,
      "loss": 0.6869,
      "step": 547
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001254681974191274,
      "loss": 0.5072,
      "step": 548
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00012400922025792804,
      "loss": 0.5121,
      "step": 549
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00012255495409138032,
      "loss": 0.6873,
      "step": 550
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00012110548907568677,
      "loss": 0.4515,
      "step": 551
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00011966091506940616,
      "loss": 0.5488,
      "step": 552
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00011822132162788226,
      "loss": 0.5171,
      "step": 553
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00011678679799769256,
      "loss": 0.5864,
      "step": 554
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00011535743311111502,
      "loss": 0.5402,
      "step": 555
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.000113933315580615,
      "loss": 0.4402,
      "step": 556
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00011251453369335172,
      "loss": 0.5805,
      "step": 557
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00011110117540570493,
      "loss": 0.8398,
      "step": 558
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00010969332833782218,
      "loss": 0.653,
      "step": 559
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00010829107976818687,
      "loss": 0.543,
      "step": 560
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.5492686033248901,
      "eval_runtime": 75.8627,
      "eval_samples_per_second": 21.473,
      "eval_steps_per_second": 10.743,
      "step": 560
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00010689451662820737,
      "loss": 0.582,
      "step": 561
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00010550372549682808,
      "loss": 0.4656,
      "step": 562
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00010411879259516158,
      "loss": 0.6783,
      "step": 563
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00010273980378114374,
      "loss": 0.4788,
      "step": 564
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00010136684454421091,
      "loss": 0.4883,
      "step": 565
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00010000000000000005,
      "loss": 0.685,
      "step": 566
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.863935488507201e-05,
      "loss": 0.5525,
      "step": 567
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.72849935516587e-05,
      "loss": 0.3223,
      "step": 568
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.593699996243295e-05,
      "loss": 0.5418,
      "step": 569
    },
    {
      "epoch": 0.82,
      "learning_rate": 9.459545768530425e-05,
      "loss": 0.5551,
      "step": 570
    },
    {
      "epoch": 0.82,
      "learning_rate": 9.326044988823733e-05,
      "loss": 0.5034,
      "step": 571
    },
    {
      "epoch": 0.82,
      "learning_rate": 9.193205933409646e-05,
      "loss": 0.3789,
      "step": 572
    },
    {
      "epoch": 0.82,
      "learning_rate": 9.061036837551466e-05,
      "loss": 0.5687,
      "step": 573
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.929545894978825e-05,
      "loss": 0.3573,
      "step": 574
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.798741257379714e-05,
      "loss": 0.4189,
      "step": 575
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.66863103389516e-05,
      "loss": 0.4463,
      "step": 576
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.53922329061643e-05,
      "loss": 0.7484,
      "step": 577
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.410526050085077e-05,
      "loss": 0.5024,
      "step": 578
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.282547290795509e-05,
      "loss": 0.5854,
      "step": 579
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.155294946700402e-05,
      "loss": 0.4451,
      "step": 580
    },
    {
      "epoch": 0.83,
      "eval_loss": 0.5421439409255981,
      "eval_runtime": 75.9322,
      "eval_samples_per_second": 21.453,
      "eval_steps_per_second": 10.733,
      "step": 580
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.02877690671884e-05,
      "loss": 0.2904,
      "step": 581
    },
    {
      "epoch": 0.83,
      "learning_rate": 7.903001014247255e-05,
      "loss": 0.5851,
      "step": 582
    },
    {
      "epoch": 0.83,
      "learning_rate": 7.777975066673133e-05,
      "loss": 0.7889,
      "step": 583
    },
    {
      "epoch": 0.84,
      "learning_rate": 7.653706814891697e-05,
      "loss": 0.4611,
      "step": 584
    },
    {
      "epoch": 0.84,
      "learning_rate": 7.530203962825331e-05,
      "loss": 0.5179,
      "step": 585
    },
    {
      "epoch": 0.84,
      "learning_rate": 7.407474166946011e-05,
      "loss": 0.5432,
      "step": 586
    },
    {
      "epoch": 0.84,
      "learning_rate": 7.285525035800644e-05,
      "loss": 0.4547,
      "step": 587
    },
    {
      "epoch": 0.84,
      "learning_rate": 7.164364129539373e-05,
      "loss": 0.8264,
      "step": 588
    },
    {
      "epoch": 0.84,
      "learning_rate": 7.043998959446896e-05,
      "loss": 0.768,
      "step": 589
    },
    {
      "epoch": 0.84,
      "learning_rate": 6.924436987476832e-05,
      "loss": 0.5146,
      "step": 590
    },
    {
      "epoch": 0.85,
      "learning_rate": 6.805685625789057e-05,
      "loss": 0.7737,
      "step": 591
    },
    {
      "epoch": 0.85,
      "learning_rate": 6.687752236290283e-05,
      "loss": 0.7187,
      "step": 592
    },
    {
      "epoch": 0.85,
      "learning_rate": 6.570644130177584e-05,
      "loss": 0.3896,
      "step": 593
    },
    {
      "epoch": 0.85,
      "learning_rate": 6.454368567485183e-05,
      "loss": 0.4378,
      "step": 594
    },
    {
      "epoch": 0.85,
      "learning_rate": 6.338932756634354e-05,
      "loss": 0.4103,
      "step": 595
    },
    {
      "epoch": 0.85,
      "learning_rate": 6.224343853986583e-05,
      "loss": 0.6346,
      "step": 596
    },
    {
      "epoch": 0.85,
      "learning_rate": 6.110608963399827e-05,
      "loss": 0.5081,
      "step": 597
    },
    {
      "epoch": 0.86,
      "learning_rate": 5.997735135788216e-05,
      "loss": 0.414,
      "step": 598
    },
    {
      "epoch": 0.86,
      "learning_rate": 5.8857293686848646e-05,
      "loss": 0.5079,
      "step": 599
    },
    {
      "epoch": 0.86,
      "learning_rate": 5.7745986058081016e-05,
      "loss": 0.5384,
      "step": 600
    },
    {
      "epoch": 0.86,
      "eval_loss": 0.5467327237129211,
      "eval_runtime": 75.8761,
      "eval_samples_per_second": 21.469,
      "eval_steps_per_second": 10.741,
      "step": 600
    },
    {
      "epoch": 0.86,
      "learning_rate": 5.6643497366309785e-05,
      "loss": 0.4515,
      "step": 601
    },
    {
      "epoch": 0.86,
      "learning_rate": 5.554989595954177e-05,
      "loss": 0.6003,
      "step": 602
    },
    {
      "epoch": 0.86,
      "learning_rate": 5.4465249634822756e-05,
      "loss": 0.7991,
      "step": 603
    },
    {
      "epoch": 0.86,
      "learning_rate": 5.338962563403478e-05,
      "loss": 0.5519,
      "step": 604
    },
    {
      "epoch": 0.87,
      "learning_rate": 5.232309063972691e-05,
      "loss": 0.5531,
      "step": 605
    },
    {
      "epoch": 0.87,
      "learning_rate": 5.126571077098208e-05,
      "loss": 0.6466,
      "step": 606
    },
    {
      "epoch": 0.87,
      "learning_rate": 5.02175515793174e-05,
      "loss": 0.4096,
      "step": 607
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.917867804462077e-05,
      "loss": 0.4503,
      "step": 608
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.81491545711223e-05,
      "loss": 0.4375,
      "step": 609
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.712904498340167e-05,
      "loss": 0.4413,
      "step": 610
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.611841252243134e-05,
      "loss": 0.5404,
      "step": 611
    },
    {
      "epoch": 0.88,
      "learning_rate": 4.511731984165619e-05,
      "loss": 0.532,
      "step": 612
    },
    {
      "epoch": 0.88,
      "learning_rate": 4.4125829003108824e-05,
      "loss": 0.5145,
      "step": 613
    },
    {
      "epoch": 0.88,
      "learning_rate": 4.3144001473562746e-05,
      "loss": 0.5592,
      "step": 614
    },
    {
      "epoch": 0.88,
      "learning_rate": 4.217189812072131e-05,
      "loss": 0.5453,
      "step": 615
    },
    {
      "epoch": 0.88,
      "learning_rate": 4.1209579209444416e-05,
      "loss": 0.5213,
      "step": 616
    },
    {
      "epoch": 0.88,
      "learning_rate": 4.02571043980124e-05,
      "loss": 0.5131,
      "step": 617
    },
    {
      "epoch": 0.88,
      "learning_rate": 3.93145327344278e-05,
      "loss": 0.5372,
      "step": 618
    },
    {
      "epoch": 0.89,
      "learning_rate": 3.838192265275411e-05,
      "loss": 0.4571,
      "step": 619
    },
    {
      "epoch": 0.89,
      "learning_rate": 3.7459331969494e-05,
      "loss": 0.4438,
      "step": 620
    },
    {
      "epoch": 0.89,
      "eval_loss": 0.5379120111465454,
      "eval_runtime": 75.9975,
      "eval_samples_per_second": 21.435,
      "eval_steps_per_second": 10.724,
      "step": 620
    },
    {
      "epoch": 0.89,
      "learning_rate": 3.65468178800044e-05,
      "loss": 0.5414,
      "step": 621
    },
    {
      "epoch": 0.89,
      "learning_rate": 3.564443695495099e-05,
      "loss": 0.5231,
      "step": 622
    },
    {
      "epoch": 0.89,
      "learning_rate": 3.4752245136801065e-05,
      "loss": 0.5973,
      "step": 623
    },
    {
      "epoch": 0.89,
      "learning_rate": 3.387029773635544e-05,
      "loss": 0.2479,
      "step": 624
    },
    {
      "epoch": 0.89,
      "learning_rate": 3.299864942931952e-05,
      "loss": 0.5817,
      "step": 625
    },
    {
      "epoch": 0.9,
      "learning_rate": 3.21373542529138e-05,
      "loss": 0.5529,
      "step": 626
    },
    {
      "epoch": 0.9,
      "learning_rate": 3.128646560252346e-05,
      "loss": 0.6231,
      "step": 627
    },
    {
      "epoch": 0.9,
      "learning_rate": 3.0446036228388842e-05,
      "loss": 0.7211,
      "step": 628
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.961611823233461e-05,
      "loss": 0.4726,
      "step": 629
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.879676306454e-05,
      "loss": 0.5618,
      "step": 630
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.7988021520349207e-05,
      "loss": 0.4254,
      "step": 631
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.718994373712247e-05,
      "loss": 0.5931,
      "step": 632
    },
    {
      "epoch": 0.91,
      "learning_rate": 2.6402579191127387e-05,
      "loss": 0.9991,
      "step": 633
    },
    {
      "epoch": 0.91,
      "learning_rate": 2.562597669447233e-05,
      "loss": 0.5177,
      "step": 634
    },
    {
      "epoch": 0.91,
      "learning_rate": 2.4860184392079865e-05,
      "loss": 0.4434,
      "step": 635
    },
    {
      "epoch": 0.91,
      "learning_rate": 2.410524975870221e-05,
      "loss": 0.4668,
      "step": 636
    },
    {
      "epoch": 0.91,
      "learning_rate": 2.336121959597808e-05,
      "loss": 0.4063,
      "step": 637
    },
    {
      "epoch": 0.91,
      "learning_rate": 2.2628140029531286e-05,
      "loss": 0.6887,
      "step": 638
    },
    {
      "epoch": 0.91,
      "learning_rate": 2.1906056506111083e-05,
      "loss": 0.3671,
      "step": 639
    },
    {
      "epoch": 0.92,
      "learning_rate": 2.1195013790775e-05,
      "loss": 0.4168,
      "step": 640
    },
    {
      "epoch": 0.92,
      "eval_loss": 0.5397519469261169,
      "eval_runtime": 75.9178,
      "eval_samples_per_second": 21.457,
      "eval_steps_per_second": 10.735,
      "step": 640
    },
    {
      "epoch": 0.92,
      "learning_rate": 2.0495055964113165e-05,
      "loss": 0.5466,
      "step": 641
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.9806226419516192e-05,
      "loss": 0.5021,
      "step": 642
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.9128567860484513e-05,
      "loss": 0.668,
      "step": 643
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.84621222979813e-05,
      "loss": 0.6027,
      "step": 644
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.7806931047827914e-05,
      "loss": 0.8838,
      "step": 645
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.7163034728142537e-05,
      "loss": 0.5397,
      "step": 646
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.6530473256822175e-05,
      "loss": 0.388,
      "step": 647
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.590928584906799e-05,
      "loss": 0.39,
      "step": 648
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.5299511014953904e-05,
      "loss": 0.4025,
      "step": 649
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.4701186557039648e-05,
      "loss": 0.5292,
      "step": 650
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.4114349568026863e-05,
      "loss": 0.4569,
      "step": 651
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.353903642845964e-05,
      "loss": 0.3907,
      "step": 652
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.2975282804469225e-05,
      "loss": 0.3466,
      "step": 653
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.2423123645562951e-05,
      "loss": 0.5364,
      "step": 654
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.188259318245728e-05,
      "loss": 0.5549,
      "step": 655
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.1353724924956121e-05,
      "loss": 0.5845,
      "step": 656
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.0836551659873074e-05,
      "loss": 0.5839,
      "step": 657
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.0331105448998957e-05,
      "loss": 0.4616,
      "step": 658
    },
    {
      "epoch": 0.94,
      "learning_rate": 9.837417627114187e-06,
      "loss": 0.5385,
      "step": 659
    },
    {
      "epoch": 0.94,
      "learning_rate": 9.355518800046171e-06,
      "loss": 0.469,
      "step": 660
    },
    {
      "epoch": 0.94,
      "eval_loss": 0.5402470827102661,
      "eval_runtime": 75.9472,
      "eval_samples_per_second": 21.449,
      "eval_steps_per_second": 10.731,
      "step": 660
    },
    {
      "epoch": 0.95,
      "learning_rate": 8.885438842771843e-06,
      "loss": 0.5539,
      "step": 661
    },
    {
      "epoch": 0.95,
      "learning_rate": 8.427206897565865e-06,
      "loss": 0.313,
      "step": 662
    },
    {
      "epoch": 0.95,
      "learning_rate": 7.980851372193488e-06,
      "loss": 0.4776,
      "step": 663
    },
    {
      "epoch": 0.95,
      "learning_rate": 7.546399938149917e-06,
      "loss": 0.5348,
      "step": 664
    },
    {
      "epoch": 0.95,
      "learning_rate": 7.123879528944599e-06,
      "loss": 0.4273,
      "step": 665
    },
    {
      "epoch": 0.95,
      "learning_rate": 6.713316338431486e-06,
      "loss": 0.5493,
      "step": 666
    },
    {
      "epoch": 0.95,
      "learning_rate": 6.314735819185202e-06,
      "loss": 0.5291,
      "step": 667
    },
    {
      "epoch": 0.96,
      "learning_rate": 5.928162680923266e-06,
      "loss": 0.5378,
      "step": 668
    },
    {
      "epoch": 0.96,
      "learning_rate": 5.553620888973887e-06,
      "loss": 0.7479,
      "step": 669
    },
    {
      "epoch": 0.96,
      "learning_rate": 5.191133662790692e-06,
      "loss": 0.6171,
      "step": 670
    },
    {
      "epoch": 0.96,
      "learning_rate": 4.840723474512876e-06,
      "loss": 0.5556,
      "step": 671
    },
    {
      "epoch": 0.96,
      "learning_rate": 4.502412047572291e-06,
      "loss": 0.4105,
      "step": 672
    },
    {
      "epoch": 0.96,
      "learning_rate": 4.176220355346594e-06,
      "loss": 0.5382,
      "step": 673
    },
    {
      "epoch": 0.96,
      "learning_rate": 3.862168619859041e-06,
      "loss": 0.4685,
      "step": 674
    },
    {
      "epoch": 0.97,
      "learning_rate": 3.5602763105249125e-06,
      "loss": 0.4685,
      "step": 675
    },
    {
      "epoch": 0.97,
      "learning_rate": 3.270562142944478e-06,
      "loss": 0.5683,
      "step": 676
    },
    {
      "epoch": 0.97,
      "learning_rate": 2.9930440777425905e-06,
      "loss": 0.376,
      "step": 677
    },
    {
      "epoch": 0.97,
      "learning_rate": 2.7277393194555358e-06,
      "loss": 0.4913,
      "step": 678
    },
    {
      "epoch": 0.97,
      "learning_rate": 2.474664315464126e-06,
      "loss": 0.4781,
      "step": 679
    },
    {
      "epoch": 0.97,
      "learning_rate": 2.2338347549742956e-06,
      "loss": 0.6766,
      "step": 680
    },
    {
      "epoch": 0.97,
      "eval_loss": 0.5407062768936157,
      "eval_runtime": 76.0027,
      "eval_samples_per_second": 21.433,
      "eval_steps_per_second": 10.723,
      "step": 680
    },
    {
      "epoch": 0.97,
      "learning_rate": 2.0052655680443457e-06,
      "loss": 0.6168,
      "step": 681
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.78897092465935e-06,
      "loss": 0.3628,
      "step": 682
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.5849642338527926e-06,
      "loss": 0.6984,
      "step": 683
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3932581428751867e-06,
      "loss": 0.64,
      "step": 684
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.2138645364101031e-06,
      "loss": 0.5111,
      "step": 685
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.0467945358372921e-06,
      "loss": 0.6078,
      "step": 686
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.920584985433023e-07,
      "loss": 0.368,
      "step": 687
    },
    {
      "epoch": 0.98,
      "learning_rate": 7.496660172793046e-07,
      "loss": 0.714,
      "step": 688
    },
    {
      "epoch": 0.99,
      "learning_rate": 6.196259195665244e-07,
      "loss": 0.3712,
      "step": 689
    },
    {
      "epoch": 0.99,
      "learning_rate": 5.019462671488562e-07,
      "loss": 0.5028,
      "step": 690
    },
    {
      "epoch": 0.99,
      "learning_rate": 3.9663435549308625e-07,
      "loss": 0.6019,
      "step": 691
    },
    {
      "epoch": 0.99,
      "learning_rate": 3.036967133367652e-07,
      "loss": 0.5255,
      "step": 692
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.2313910228324297e-07,
      "loss": 0.5835,
      "step": 693
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.5496651644468786e-07,
      "loss": 0.391,
      "step": 694
    },
    {
      "epoch": 0.99,
      "learning_rate": 9.918318213231193e-08,
      "loss": 0.486,
      "step": 695
    },
    {
      "epoch": 1.0,
      "learning_rate": 5.579255759451396e-08,
      "loss": 0.6019,
      "step": 696
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.479733280236207e-08,
      "loss": 0.4126,
      "step": 697
    },
    {
      "epoch": 1.0,
      "learning_rate": 6.199429282904845e-09,
      "loss": 0.4922,
      "step": 698
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0,
      "loss": 0.4515,
      "step": 699
    }
  ],
  "logging_steps": 1,
  "max_steps": 699,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 1.8006457547803853e+18,
  "trial_name": null,
  "trial_params": null
}