{
  "best_metric": 3.2291667461395264,
  "best_model_checkpoint": "/leonardo_scratch/large/userexternal/fcocchi0/itserr/latin_llm/latin_gpt2_from_scratch/scratch_2-nodes_tokenizer_latbert-original_packing_fcocchi/checkpoint-6300",
  "epoch": 3.548647447921809,
  "eval_steps": 100,
  "global_step": 6300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 3e-07,
      "loss": 10.5505,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 6e-07,
      "loss": 10.55,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 9e-07,
      "loss": 10.5327,
      "step": 3
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.2e-06,
      "loss": 10.4992,
      "step": 4
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.4999999999999998e-06,
      "loss": 10.4523,
      "step": 5
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.8e-06,
      "loss": 10.3869,
      "step": 6
    },
    {
      "epoch": 0.0,
      "learning_rate": 2.1e-06,
      "loss": 10.3118,
      "step": 7
    },
    {
      "epoch": 0.0,
      "learning_rate": 2.4e-06,
      "loss": 10.2255,
      "step": 8
    },
    {
      "epoch": 0.0,
      "learning_rate": 2.6999999999999996e-06,
      "loss": 10.1289,
      "step": 9
    },
    {
      "epoch": 0.0,
      "learning_rate": 2.9999999999999997e-06,
      "loss": 10.0346,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 3.2999999999999993e-06,
      "loss": 9.9478,
      "step": 11
    },
    {
      "epoch": 0.01,
      "learning_rate": 3.6e-06,
      "loss": 9.8691,
      "step": 12
    },
    {
      "epoch": 0.01,
      "learning_rate": 3.899999999999999e-06,
      "loss": 9.7756,
      "step": 13
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.2e-06,
      "loss": 9.7086,
      "step": 14
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.499999999999999e-06,
      "loss": 9.6538,
      "step": 15
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.8e-06,
      "loss": 9.6026,
      "step": 16
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.1e-06,
      "loss": 9.5342,
      "step": 17
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.399999999999999e-06,
      "loss": 9.4872,
      "step": 18
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.7e-06,
      "loss": 9.4419,
      "step": 19
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.999999999999999e-06,
      "loss": 9.3978,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.3e-06,
      "loss": 9.3527,
      "step": 21
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.599999999999999e-06,
      "loss": 9.3241,
      "step": 22
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.899999999999999e-06,
      "loss": 9.2958,
      "step": 23
    },
    {
      "epoch": 0.01,
      "learning_rate": 7.2e-06,
      "loss": 9.2723,
      "step": 24
    },
    {
      "epoch": 0.01,
      "learning_rate": 7.499999999999999e-06,
      "loss": 9.2308,
      "step": 25
    },
    {
      "epoch": 0.01,
      "learning_rate": 7.799999999999998e-06,
      "loss": 9.2126,
      "step": 26
    },
    {
      "epoch": 0.01,
      "learning_rate": 8.099999999999999e-06,
      "loss": 9.1869,
      "step": 27
    },
    {
      "epoch": 0.01,
      "learning_rate": 8.4e-06,
      "loss": 9.1636,
      "step": 28
    },
    {
      "epoch": 0.01,
      "learning_rate": 8.7e-06,
      "loss": 9.1351,
      "step": 29
    },
    {
      "epoch": 0.01,
      "learning_rate": 8.999999999999999e-06,
      "loss": 9.1292,
      "step": 30
    },
    {
      "epoch": 0.02,
      "learning_rate": 9.299999999999999e-06,
      "loss": 9.1143,
      "step": 31
    },
    {
      "epoch": 0.02,
      "learning_rate": 9.6e-06,
      "loss": 9.1,
      "step": 32
    },
    {
      "epoch": 0.02,
      "learning_rate": 9.9e-06,
      "loss": 9.0882,
      "step": 33
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.02e-05,
      "loss": 9.0575,
      "step": 34
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.05e-05,
      "loss": 9.0302,
      "step": 35
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.0799999999999998e-05,
      "loss": 8.9985,
      "step": 36
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.1099999999999999e-05,
      "loss": 8.9902,
      "step": 37
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.14e-05,
      "loss": 8.956,
      "step": 38
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.17e-05,
      "loss": 8.9296,
      "step": 39
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.1999999999999999e-05,
      "loss": 8.9036,
      "step": 40
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.2299999999999999e-05,
      "loss": 8.9249,
      "step": 41
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.26e-05,
      "loss": 8.8919,
      "step": 42
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.2899999999999998e-05,
      "loss": 8.8674,
      "step": 43
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.3199999999999997e-05,
      "loss": 8.8374,
      "step": 44
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.3499999999999998e-05,
      "loss": 8.8298,
      "step": 45
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.3799999999999998e-05,
      "loss": 8.8043,
      "step": 46
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.4099999999999999e-05,
      "loss": 8.7985,
      "step": 47
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.44e-05,
      "loss": 8.7971,
      "step": 48
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.47e-05,
      "loss": 8.7547,
      "step": 49
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.4999999999999999e-05,
      "loss": 8.7363,
      "step": 50
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.53e-05,
      "loss": 8.7255,
      "step": 51
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.5599999999999996e-05,
      "loss": 8.7006,
      "step": 52
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.5899999999999997e-05,
      "loss": 8.6881,
      "step": 53
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6199999999999997e-05,
      "loss": 8.6703,
      "step": 54
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6499999999999998e-05,
      "loss": 8.651,
      "step": 55
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.68e-05,
      "loss": 8.6256,
      "step": 56
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.71e-05,
      "loss": 8.5972,
      "step": 57
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.74e-05,
      "loss": 8.5971,
      "step": 58
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.7699999999999997e-05,
      "loss": 8.5871,
      "step": 59
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.7999999999999997e-05,
      "loss": 8.5653,
      "step": 60
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.8299999999999998e-05,
      "loss": 8.5326,
      "step": 61
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.8599999999999998e-05,
      "loss": 8.5213,
      "step": 62
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.89e-05,
      "loss": 8.5018,
      "step": 63
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.92e-05,
      "loss": 8.4921,
      "step": 64
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.95e-05,
      "loss": 8.4692,
      "step": 65
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.98e-05,
      "loss": 8.4592,
      "step": 66
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.01e-05,
      "loss": 8.4458,
      "step": 67
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.04e-05,
      "loss": 8.4309,
      "step": 68
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.07e-05,
      "loss": 8.4086,
      "step": 69
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.1e-05,
      "loss": 8.3835,
      "step": 70
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.1299999999999996e-05,
      "loss": 8.3681,
      "step": 71
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.1599999999999996e-05,
      "loss": 8.3678,
      "step": 72
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.1899999999999997e-05,
      "loss": 8.3407,
      "step": 73
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.2199999999999998e-05,
      "loss": 8.3117,
      "step": 74
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.2499999999999998e-05,
      "loss": 8.3196,
      "step": 75
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.28e-05,
      "loss": 8.3008,
      "step": 76
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.31e-05,
      "loss": 8.2673,
      "step": 77
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.34e-05,
      "loss": 8.2407,
      "step": 78
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.3699999999999997e-05,
      "loss": 8.2235,
      "step": 79
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.3999999999999997e-05,
      "loss": 8.2034,
      "step": 80
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.4299999999999998e-05,
      "loss": 8.1748,
      "step": 81
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.4599999999999998e-05,
      "loss": 8.1644,
      "step": 82
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.49e-05,
      "loss": 8.1515,
      "step": 83
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.52e-05,
      "loss": 8.1324,
      "step": 84
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.55e-05,
      "loss": 8.1209,
      "step": 85
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.5799999999999997e-05,
      "loss": 8.0796,
      "step": 86
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.6099999999999997e-05,
      "loss": 8.0506,
      "step": 87
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.6399999999999995e-05,
      "loss": 8.0295,
      "step": 88
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.6699999999999995e-05,
      "loss": 8.0513,
      "step": 89
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.6999999999999996e-05,
      "loss": 8.0231,
      "step": 90
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.7299999999999996e-05,
      "loss": 7.9821,
      "step": 91
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.7599999999999997e-05,
      "loss": 7.9843,
      "step": 92
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.7899999999999997e-05,
      "loss": 7.9655,
      "step": 93
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.8199999999999998e-05,
      "loss": 7.9321,
      "step": 94
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.8499999999999998e-05,
      "loss": 7.8901,
      "step": 95
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.88e-05,
      "loss": 7.8696,
      "step": 96
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.91e-05,
      "loss": 7.8938,
      "step": 97
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.94e-05,
      "loss": 7.8579,
      "step": 98
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.97e-05,
      "loss": 7.8416,
      "step": 99
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 7.8367,
      "step": 100
    },
    {
      "epoch": 0.05,
      "eval_loss": 7.395714282989502,
      "eval_perplexity": 1641.7833251953125,
      "eval_runtime": 104.059,
      "eval_samples_per_second": 960.994,
      "eval_steps_per_second": 30.031,
      "step": 100
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.0299999999999998e-05,
      "loss": 7.7875,
      "step": 101
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.06e-05,
      "loss": 7.7814,
      "step": 102
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.09e-05,
      "loss": 7.7541,
      "step": 103
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.119999999999999e-05,
      "loss": 7.753,
      "step": 104
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.149999999999999e-05,
      "loss": 7.726,
      "step": 105
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.1799999999999994e-05,
      "loss": 7.7173,
      "step": 106
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.2099999999999994e-05,
      "loss": 7.7112,
      "step": 107
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.2399999999999995e-05,
      "loss": 7.6813,
      "step": 108
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.2699999999999995e-05,
      "loss": 7.6615,
      "step": 109
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.2999999999999996e-05,
      "loss": 7.6425,
      "step": 110
    },
    {
      "epoch": 0.05,
      "learning_rate": 3.3299999999999996e-05,
      "loss": 7.6085,
      "step": 111
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.36e-05,
      "loss": 7.6122,
      "step": 112
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.39e-05,
      "loss": 7.5893,
      "step": 113
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.42e-05,
      "loss": 7.5572,
      "step": 114
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.45e-05,
      "loss": 7.5692,
      "step": 115
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.48e-05,
      "loss": 7.5718,
      "step": 116
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.51e-05,
      "loss": 7.5346,
      "step": 117
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.539999999999999e-05,
      "loss": 7.5333,
      "step": 118
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.5699999999999994e-05,
      "loss": 7.5087,
      "step": 119
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.5999999999999994e-05,
      "loss": 7.4706,
      "step": 120
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.6299999999999995e-05,
      "loss": 7.4646,
      "step": 121
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.6599999999999995e-05,
      "loss": 7.4424,
      "step": 122
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.6899999999999996e-05,
      "loss": 7.417,
      "step": 123
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.7199999999999996e-05,
      "loss": 7.4438,
      "step": 124
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.75e-05,
      "loss": 7.3714,
      "step": 125
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.78e-05,
      "loss": 7.3652,
      "step": 126
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.81e-05,
      "loss": 7.3716,
      "step": 127
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.84e-05,
      "loss": 7.3251,
      "step": 128
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.87e-05,
      "loss": 7.3463,
      "step": 129
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.9e-05,
      "loss": 7.325,
      "step": 130
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.93e-05,
      "loss": 7.3076,
      "step": 131
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.96e-05,
      "loss": 7.296,
      "step": 132
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.99e-05,
      "loss": 7.218,
      "step": 133
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.02e-05,
      "loss": 7.2249,
      "step": 134
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.05e-05,
      "loss": 7.2453,
      "step": 135
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.08e-05,
      "loss": 7.2464,
      "step": 136
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.11e-05,
      "loss": 7.2304,
      "step": 137
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.14e-05,
      "loss": 7.2442,
      "step": 138
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.17e-05,
      "loss": 7.2581,
      "step": 139
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.2e-05,
      "loss": 7.1581,
      "step": 140
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.229999999999999e-05,
      "loss": 7.1703,
      "step": 141
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.259999999999999e-05,
      "loss": 7.1596,
      "step": 142
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.289999999999999e-05,
      "loss": 7.1292,
      "step": 143
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.319999999999999e-05,
      "loss": 7.1428,
      "step": 144
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.3499999999999993e-05,
      "loss": 7.1585,
      "step": 145
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.3799999999999994e-05,
      "loss": 7.1357,
      "step": 146
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.4099999999999995e-05,
      "loss": 7.0859,
      "step": 147
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.4399999999999995e-05,
      "loss": 7.0889,
      "step": 148
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.4699999999999996e-05,
      "loss": 7.0271,
      "step": 149
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.4999999999999996e-05,
      "loss": 7.0211,
      "step": 150
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.5299999999999997e-05,
      "loss": 7.0439,
      "step": 151
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.56e-05,
      "loss": 7.0246,
      "step": 152
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.59e-05,
      "loss": 7.0444,
      "step": 153
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.62e-05,
      "loss": 7.0673,
      "step": 154
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.65e-05,
      "loss": 7.0133,
      "step": 155
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.68e-05,
      "loss": 6.9997,
      "step": 156
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.709999999999999e-05,
      "loss": 7.0195,
      "step": 157
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.7399999999999993e-05,
      "loss": 7.0069,
      "step": 158
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.7699999999999994e-05,
      "loss": 6.9496,
      "step": 159
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.7999999999999994e-05,
      "loss": 7.0044,
      "step": 160
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.8299999999999995e-05,
      "loss": 6.9773,
      "step": 161
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.8599999999999995e-05,
      "loss": 6.9693,
      "step": 162
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.8899999999999996e-05,
      "loss": 6.9713,
      "step": 163
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.9199999999999997e-05,
      "loss": 6.9645,
      "step": 164
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.95e-05,
      "loss": 6.9289,
      "step": 165
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.98e-05,
      "loss": 6.9019,
      "step": 166
    },
    {
      "epoch": 0.08,
      "learning_rate": 5.01e-05,
      "loss": 6.8871,
      "step": 167
    },
    {
      "epoch": 0.08,
      "learning_rate": 5.04e-05,
      "loss": 6.9074,
      "step": 168
    },
    {
      "epoch": 0.08,
      "learning_rate": 5.07e-05,
      "loss": 6.9036,
      "step": 169
    },
    {
      "epoch": 0.08,
      "learning_rate": 5.1e-05,
      "loss": 6.9018,
      "step": 170
    },
    {
      "epoch": 0.08,
      "learning_rate": 5.13e-05,
      "loss": 6.8619,
      "step": 171
    },
    {
      "epoch": 0.08,
      "learning_rate": 5.1599999999999994e-05,
      "loss": 6.8403,
      "step": 172
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.1899999999999994e-05,
      "loss": 6.8643,
      "step": 173
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.2199999999999995e-05,
      "loss": 6.8651,
      "step": 174
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.2499999999999995e-05,
      "loss": 6.8386,
      "step": 175
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.279999999999999e-05,
      "loss": 6.7607,
      "step": 176
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.309999999999999e-05,
      "loss": 6.8308,
      "step": 177
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.339999999999999e-05,
      "loss": 6.8756,
      "step": 178
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.369999999999999e-05,
      "loss": 6.8236,
      "step": 179
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.399999999999999e-05,
      "loss": 6.824,
      "step": 180
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.429999999999999e-05,
      "loss": 6.7949,
      "step": 181
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.459999999999999e-05,
      "loss": 6.7748,
      "step": 182
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.489999999999999e-05,
      "loss": 6.8412,
      "step": 183
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.519999999999999e-05,
      "loss": 6.7812,
      "step": 184
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.5499999999999994e-05,
      "loss": 6.7697,
      "step": 185
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.5799999999999994e-05,
      "loss": 6.7746,
      "step": 186
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.6099999999999995e-05,
      "loss": 6.7341,
      "step": 187
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.6399999999999995e-05,
      "loss": 6.7575,
      "step": 188
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.6699999999999996e-05,
      "loss": 6.7391,
      "step": 189
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.6999999999999996e-05,
      "loss": 6.7521,
      "step": 190
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.73e-05,
      "loss": 6.7203,
      "step": 191
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.76e-05,
      "loss": 6.6694,
      "step": 192
    },
    {
      "epoch": 0.1,
      "learning_rate": 5.79e-05,
      "loss": 6.688,
      "step": 193
    },
    {
      "epoch": 0.1,
      "learning_rate": 5.82e-05,
      "loss": 6.6694,
      "step": 194
    },
    {
      "epoch": 0.1,
      "learning_rate": 5.85e-05,
      "loss": 6.6944,
      "step": 195
    },
    {
      "epoch": 0.1,
      "learning_rate": 5.88e-05,
      "loss": 6.6918,
      "step": 196
    },
    {
      "epoch": 0.1,
      "learning_rate": 5.91e-05,
      "loss": 6.7016,
      "step": 197
    },
    {
      "epoch": 0.1,
      "learning_rate": 5.94e-05,
      "loss": 6.6725,
      "step": 198
    },
    {
      "epoch": 0.1,
      "learning_rate": 5.97e-05,
      "loss": 6.6608,
      "step": 199
    },
    {
      "epoch": 0.1,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 6.6462,
      "step": 200
    },
    {
      "epoch": 0.1,
      "eval_loss": 6.360668182373047,
      "eval_perplexity": 588.505615234375,
      "eval_runtime": 47.3128,
      "eval_samples_per_second": 2113.595,
      "eval_steps_per_second": 66.05,
      "step": 200
    },
    {
      "epoch": 0.1,
      "learning_rate": 6.0299999999999995e-05,
      "loss": 6.6237,
      "step": 201
    },
    {
      "epoch": 0.1,
      "learning_rate": 6.0599999999999996e-05,
      "loss": 6.6443,
      "step": 202
    },
    {
      "epoch": 0.1,
      "learning_rate": 6.0899999999999996e-05,
      "loss": 6.6778,
      "step": 203
    },
    {
      "epoch": 0.1,
      "learning_rate": 6.12e-05,
      "loss": 6.6503,
      "step": 204
    },
    {
      "epoch": 0.1,
      "learning_rate": 6.149999999999999e-05,
      "loss": 6.5883,
      "step": 205
    },
    {
      "epoch": 0.1,
      "learning_rate": 6.18e-05,
      "loss": 6.5883,
      "step": 206
    },
    {
      "epoch": 0.1,
      "learning_rate": 6.209999999999999e-05,
      "loss": 6.6014,
      "step": 207
    },
    {
      "epoch": 0.1,
      "learning_rate": 6.239999999999999e-05,
      "loss": 6.6172,
      "step": 208
    },
    {
      "epoch": 0.1,
      "learning_rate": 6.269999999999999e-05,
      "loss": 6.5974,
      "step": 209
    },
    {
      "epoch": 0.1,
      "learning_rate": 6.299999999999999e-05,
      "loss": 6.5607,
      "step": 210
    },
    {
      "epoch": 0.1,
      "learning_rate": 6.33e-05,
      "loss": 6.5335,
      "step": 211
    },
    {
      "epoch": 0.1,
      "learning_rate": 6.359999999999999e-05,
      "loss": 6.5254,
      "step": 212
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.39e-05,
      "loss": 6.5476,
      "step": 213
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.419999999999999e-05,
      "loss": 6.6115,
      "step": 214
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.45e-05,
      "loss": 6.5606,
      "step": 215
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.479999999999999e-05,
      "loss": 6.5545,
      "step": 216
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.51e-05,
      "loss": 6.542,
      "step": 217
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.539999999999999e-05,
      "loss": 6.5752,
      "step": 218
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.57e-05,
      "loss": 6.5697,
      "step": 219
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.599999999999999e-05,
      "loss": 6.5713,
      "step": 220
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.63e-05,
      "loss": 6.5193,
      "step": 221
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.659999999999999e-05,
      "loss": 6.4797,
      "step": 222
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.69e-05,
      "loss": 6.5563,
      "step": 223
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.72e-05,
      "loss": 6.5155,
      "step": 224
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.75e-05,
      "loss": 6.519,
      "step": 225
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.78e-05,
      "loss": 6.4943,
      "step": 226
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.81e-05,
      "loss": 6.4333,
      "step": 227
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.84e-05,
      "loss": 6.4428,
      "step": 228
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.87e-05,
      "loss": 6.4403,
      "step": 229
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.9e-05,
      "loss": 6.4462,
      "step": 230
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.93e-05,
      "loss": 6.4837,
      "step": 231
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.96e-05,
      "loss": 6.4349,
      "step": 232
    },
    {
      "epoch": 0.12,
      "learning_rate": 6.989999999999999e-05,
      "loss": 6.4084,
      "step": 233
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.02e-05,
      "loss": 6.3892,
      "step": 234
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.049999999999999e-05,
      "loss": 6.4615,
      "step": 235
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.079999999999999e-05,
      "loss": 6.4714,
      "step": 236
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.11e-05,
      "loss": 6.4407,
      "step": 237
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.139999999999999e-05,
      "loss": 6.4199,
      "step": 238
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.17e-05,
      "loss": 6.4517,
      "step": 239
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.199999999999999e-05,
      "loss": 6.3998,
      "step": 240
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.23e-05,
      "loss": 6.3704,
      "step": 241
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.259999999999999e-05,
      "loss": 6.3732,
      "step": 242
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.29e-05,
      "loss": 6.3412,
      "step": 243
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.319999999999999e-05,
      "loss": 6.3835,
      "step": 244
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.35e-05,
      "loss": 6.4028,
      "step": 245
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.379999999999999e-05,
      "loss": 6.3879,
      "step": 246
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.41e-05,
      "loss": 6.4299,
      "step": 247
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.439999999999999e-05,
      "loss": 6.4034,
      "step": 248
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.47e-05,
      "loss": 6.3951,
      "step": 249
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.5e-05,
      "loss": 6.4044,
      "step": 250
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.529999999999999e-05,
      "loss": 6.3788,
      "step": 251
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.56e-05,
      "loss": 6.3588,
      "step": 252
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.589999999999999e-05,
      "loss": 6.361,
      "step": 253
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.62e-05,
      "loss": 6.3462,
      "step": 254
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.649999999999999e-05,
      "loss": 6.3355,
      "step": 255
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.68e-05,
      "loss": 6.2993,
      "step": 256
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.709999999999999e-05,
      "loss": 6.2947,
      "step": 257
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.74e-05,
      "loss": 6.3233,
      "step": 258
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.769999999999999e-05,
      "loss": 6.3308,
      "step": 259
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.8e-05,
      "loss": 6.3022,
      "step": 260
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.829999999999999e-05,
      "loss": 6.2699,
      "step": 261
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.86e-05,
      "loss": 6.3163,
      "step": 262
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.89e-05,
      "loss": 6.2407,
      "step": 263
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.92e-05,
      "loss": 6.3039,
      "step": 264
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.95e-05,
      "loss": 6.3269,
      "step": 265
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.98e-05,
      "loss": 6.3141,
      "step": 266
    },
    {
      "epoch": 0.13,
      "learning_rate": 8.01e-05,
      "loss": 6.2301,
      "step": 267
    },
    {
      "epoch": 0.13,
      "learning_rate": 8.04e-05,
      "loss": 6.2314,
      "step": 268
    },
    {
      "epoch": 0.13,
      "learning_rate": 8.07e-05,
      "loss": 6.2585,
      "step": 269
    },
    {
      "epoch": 0.13,
      "learning_rate": 8.1e-05,
      "loss": 6.3191,
      "step": 270
    },
    {
      "epoch": 0.13,
      "learning_rate": 8.13e-05,
      "loss": 6.2622,
      "step": 271
    },
    {
      "epoch": 0.13,
      "learning_rate": 8.16e-05,
      "loss": 6.2737,
      "step": 272
    },
    {
      "epoch": 0.13,
      "learning_rate": 8.19e-05,
      "loss": 6.2433,
      "step": 273
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.22e-05,
      "loss": 6.1509,
      "step": 274
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.25e-05,
      "loss": 6.2474,
      "step": 275
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.28e-05,
      "loss": 6.2867,
      "step": 276
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.31e-05,
      "loss": 6.2867,
      "step": 277
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.34e-05,
      "loss": 6.2708,
      "step": 278
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.37e-05,
      "loss": 6.2431,
      "step": 279
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.4e-05,
      "loss": 6.2033,
      "step": 280
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.43e-05,
      "loss": 6.2028,
      "step": 281
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.459999999999998e-05,
      "loss": 6.1418,
      "step": 282
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.489999999999999e-05,
      "loss": 6.241,
      "step": 283
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.519999999999998e-05,
      "loss": 6.2212,
      "step": 284
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.549999999999999e-05,
      "loss": 6.1629,
      "step": 285
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.579999999999998e-05,
      "loss": 6.2423,
      "step": 286
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.609999999999999e-05,
      "loss": 6.1569,
      "step": 287
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.639999999999999e-05,
      "loss": 6.165,
      "step": 288
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.669999999999998e-05,
      "loss": 6.2393,
      "step": 289
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.699999999999999e-05,
      "loss": 6.183,
      "step": 290
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.729999999999998e-05,
      "loss": 6.1804,
      "step": 291
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.759999999999999e-05,
      "loss": 6.1409,
      "step": 292
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.789999999999998e-05,
      "loss": 6.122,
      "step": 293
    },
    {
      "epoch": 0.15,
      "learning_rate": 8.819999999999999e-05,
      "loss": 6.2016,
      "step": 294
    },
    {
      "epoch": 0.15,
      "learning_rate": 8.849999999999998e-05,
      "loss": 6.2088,
      "step": 295
    },
    {
      "epoch": 0.15,
      "learning_rate": 8.879999999999999e-05,
      "loss": 6.2174,
      "step": 296
    },
    {
      "epoch": 0.15,
      "learning_rate": 8.909999999999998e-05,
      "loss": 6.1718,
      "step": 297
    },
    {
      "epoch": 0.15,
      "learning_rate": 8.939999999999999e-05,
      "loss": 6.1371,
      "step": 298
    },
    {
      "epoch": 0.15,
      "learning_rate": 8.969999999999998e-05,
      "loss": 6.1563,
      "step": 299
    },
    {
      "epoch": 0.15,
      "learning_rate": 8.999999999999999e-05,
      "loss": 6.141,
      "step": 300
    },
    {
      "epoch": 0.15,
      "eval_loss": 5.846449375152588,
      "eval_perplexity": 355.3304138183594,
      "eval_runtime": 47.348,
      "eval_samples_per_second": 2112.02,
      "eval_steps_per_second": 66.001,
      "step": 300
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.029999999999999e-05,
      "loss": 6.1273,
      "step": 301
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.059999999999999e-05,
      "loss": 6.1483,
      "step": 302
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.089999999999999e-05,
      "loss": 6.1726,
      "step": 303
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.12e-05,
      "loss": 6.1578,
      "step": 304
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.149999999999999e-05,
      "loss": 6.1498,
      "step": 305
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.18e-05,
      "loss": 6.1934,
      "step": 306
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.209999999999999e-05,
      "loss": 6.1273,
      "step": 307
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.24e-05,
      "loss": 6.14,
      "step": 308
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.269999999999999e-05,
      "loss": 6.1552,
      "step": 309
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.3e-05,
      "loss": 6.089,
      "step": 310
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.329999999999999e-05,
      "loss": 6.0914,
      "step": 311
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.36e-05,
      "loss": 6.0915,
      "step": 312
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.389999999999999e-05,
      "loss": 6.067,
      "step": 313
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.419999999999999e-05,
      "loss": 6.0831,
      "step": 314
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.449999999999999e-05,
      "loss": 6.0904,
      "step": 315
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.479999999999999e-05,
      "loss": 6.1102,
      "step": 316
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.51e-05,
      "loss": 6.0906,
      "step": 317
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.539999999999999e-05,
      "loss": 6.0969,
      "step": 318
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.57e-05,
      "loss": 6.0847,
      "step": 319
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.599999999999999e-05,
      "loss": 6.0684,
      "step": 320
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.63e-05,
      "loss": 6.0476,
      "step": 321
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.659999999999999e-05,
      "loss": 6.0594,
      "step": 322
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.69e-05,
      "loss": 6.0101,
      "step": 323
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.719999999999999e-05,
      "loss": 6.0098,
      "step": 324
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.75e-05,
      "loss": 6.048,
      "step": 325
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.779999999999999e-05,
      "loss": 6.058,
      "step": 326
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.81e-05,
      "loss": 6.074,
      "step": 327
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.839999999999999e-05,
      "loss": 6.063,
      "step": 328
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.87e-05,
      "loss": 6.0217,
      "step": 329
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.9e-05,
      "loss": 6.0167,
      "step": 330
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.93e-05,
      "loss": 6.0324,
      "step": 331
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.96e-05,
      "loss": 5.983,
      "step": 332
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.99e-05,
      "loss": 6.0015,
      "step": 333
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001002,
      "loss": 6.0464,
      "step": 334
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001005,
      "loss": 5.994,
      "step": 335
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001008,
      "loss": 5.9946,
      "step": 336
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001011,
      "loss": 6.0028,
      "step": 337
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001014,
      "loss": 5.9715,
      "step": 338
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00010169999999999999,
      "loss": 5.9778,
      "step": 339
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.000102,
      "loss": 6.0087,
      "step": 340
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00010229999999999999,
      "loss": 6.0072,
      "step": 341
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001026,
      "loss": 5.985,
      "step": 342
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001029,
      "loss": 5.9916,
      "step": 343
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00010319999999999999,
      "loss": 5.9897,
      "step": 344
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00010349999999999998,
      "loss": 6.022,
      "step": 345
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00010379999999999999,
      "loss": 5.9359,
      "step": 346
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00010409999999999998,
      "loss": 5.9185,
      "step": 347
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00010439999999999999,
      "loss": 5.9329,
      "step": 348
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00010469999999999998,
      "loss": 6.0124,
      "step": 349
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00010499999999999999,
      "loss": 5.957,
      "step": 350
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00010529999999999998,
      "loss": 5.9441,
      "step": 351
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00010559999999999998,
      "loss": 5.9706,
      "step": 352
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00010589999999999999,
      "loss": 5.9568,
      "step": 353
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00010619999999999998,
      "loss": 5.9044,
      "step": 354
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00010649999999999999,
      "loss": 5.9164,
      "step": 355
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00010679999999999998,
      "loss": 5.9563,
      "step": 356
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00010709999999999999,
      "loss": 5.9935,
      "step": 357
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00010739999999999998,
      "loss": 5.9687,
      "step": 358
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00010769999999999999,
      "loss": 5.9596,
      "step": 359
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00010799999999999998,
      "loss": 5.884,
      "step": 360
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00010829999999999999,
      "loss": 5.813,
      "step": 361
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00010859999999999998,
      "loss": 5.9526,
      "step": 362
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00010889999999999999,
      "loss": 5.9453,
      "step": 363
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00010919999999999998,
      "loss": 5.9494,
      "step": 364
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00010949999999999999,
      "loss": 5.8922,
      "step": 365
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00010979999999999999,
      "loss": 5.956,
      "step": 366
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00011009999999999999,
      "loss": 5.9456,
      "step": 367
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00011039999999999999,
      "loss": 5.8863,
      "step": 368
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001107,
      "loss": 5.8773,
      "step": 369
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00011099999999999999,
      "loss": 5.8503,
      "step": 370
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001113,
      "loss": 5.9039,
      "step": 371
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00011159999999999999,
      "loss": 5.934,
      "step": 372
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001119,
      "loss": 5.8915,
      "step": 373
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00011219999999999999,
      "loss": 5.8451,
      "step": 374
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001125,
      "loss": 5.8894,
      "step": 375
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00011279999999999999,
      "loss": 5.9067,
      "step": 376
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00011309999999999998,
      "loss": 5.8137,
      "step": 377
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00011339999999999999,
      "loss": 5.8673,
      "step": 378
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00011369999999999999,
      "loss": 5.8261,
      "step": 379
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00011399999999999999,
      "loss": 5.8207,
      "step": 380
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00011429999999999999,
      "loss": 5.8385,
      "step": 381
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001146,
      "loss": 5.8338,
      "step": 382
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00011489999999999999,
      "loss": 5.8626,
      "step": 383
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001152,
      "loss": 5.8835,
      "step": 384
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00011549999999999999,
      "loss": 5.8526,
      "step": 385
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001158,
      "loss": 5.8532,
      "step": 386
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00011609999999999999,
      "loss": 5.8494,
      "step": 387
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001164,
      "loss": 5.8673,
      "step": 388
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00011669999999999999,
      "loss": 5.9206,
      "step": 389
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.000117,
      "loss": 5.8781,
      "step": 390
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00011729999999999999,
      "loss": 5.8708,
      "step": 391
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001176,
      "loss": 5.9089,
      "step": 392
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00011789999999999999,
      "loss": 5.8747,
      "step": 393
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001182,
      "loss": 5.8297,
      "step": 394
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001185,
      "loss": 5.8424,
      "step": 395
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001188,
      "loss": 5.8373,
      "step": 396
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001191,
      "loss": 5.8204,
      "step": 397
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001194,
      "loss": 5.7772,
      "step": 398
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001197,
      "loss": 5.7894,
      "step": 399
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00011999999999999999,
      "loss": 5.7917,
      "step": 400
    },
    {
      "epoch": 0.2,
      "eval_loss": 5.498251914978027,
      "eval_perplexity": 252.27073669433594,
      "eval_runtime": 46.8895,
      "eval_samples_per_second": 2132.672,
      "eval_steps_per_second": 66.646,
      "step": 400
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001203,
      "loss": 5.7939,
      "step": 401
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00012059999999999999,
      "loss": 5.7812,
      "step": 402
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001209,
      "loss": 5.8048,
      "step": 403
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00012119999999999999,
      "loss": 5.8283,
      "step": 404
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001215,
      "loss": 5.801,
      "step": 405
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00012179999999999999,
      "loss": 5.7634,
      "step": 406
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00012209999999999999,
      "loss": 5.7784,
      "step": 407
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001224,
      "loss": 5.797,
      "step": 408
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00012269999999999997,
      "loss": 5.7394,
      "step": 409
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00012299999999999998,
      "loss": 5.79,
      "step": 410
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001233,
      "loss": 5.7906,
      "step": 411
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001236,
      "loss": 5.7817,
      "step": 412
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00012389999999999998,
      "loss": 5.806,
      "step": 413
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00012419999999999998,
      "loss": 5.7556,
      "step": 414
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001245,
      "loss": 5.7396,
      "step": 415
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00012479999999999997,
      "loss": 5.7418,
      "step": 416
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00012509999999999998,
      "loss": 5.7139,
      "step": 417
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00012539999999999999,
      "loss": 5.713,
      "step": 418
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001257,
      "loss": 5.733,
      "step": 419
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00012599999999999997,
      "loss": 5.7591,
      "step": 420
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00012629999999999998,
      "loss": 5.779,
      "step": 421
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001266,
      "loss": 5.7978,
      "step": 422
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001269,
      "loss": 5.7836,
      "step": 423
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00012719999999999997,
      "loss": 5.7381,
      "step": 424
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00012749999999999998,
      "loss": 5.7036,
      "step": 425
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001278,
      "loss": 5.7095,
      "step": 426
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001281,
      "loss": 5.7488,
      "step": 427
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00012839999999999998,
      "loss": 5.7717,
      "step": 428
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00012869999999999998,
      "loss": 5.6837,
      "step": 429
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.000129,
      "loss": 5.7164,
      "step": 430
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001293,
      "loss": 5.7416,
      "step": 431
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00012959999999999998,
      "loss": 5.7687,
      "step": 432
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00012989999999999999,
      "loss": 5.7038,
      "step": 433
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001302,
      "loss": 5.7239,
      "step": 434
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001305,
      "loss": 5.7219,
      "step": 435
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00013079999999999998,
      "loss": 5.707,
      "step": 436
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001311,
      "loss": 5.6438,
      "step": 437
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001314,
      "loss": 5.69,
      "step": 438
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00013169999999999998,
      "loss": 5.6808,
      "step": 439
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00013199999999999998,
      "loss": 5.6899,
      "step": 440
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001323,
      "loss": 5.6269,
      "step": 441
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001326,
      "loss": 5.692,
      "step": 442
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00013289999999999998,
      "loss": 5.6941,
      "step": 443
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00013319999999999999,
      "loss": 5.6445,
      "step": 444
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001335,
      "loss": 5.7082,
      "step": 445
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001338,
      "loss": 5.6975,
      "step": 446
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00013409999999999998,
      "loss": 5.6711,
      "step": 447
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001344,
      "loss": 5.6334,
      "step": 448
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001347,
      "loss": 5.7378,
      "step": 449
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.000135,
      "loss": 5.6745,
      "step": 450
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00013529999999999998,
      "loss": 5.643,
      "step": 451
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001356,
      "loss": 5.6971,
      "step": 452
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001359,
      "loss": 5.6211,
      "step": 453
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001362,
      "loss": 5.6385,
      "step": 454
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00013649999999999998,
      "loss": 5.6933,
      "step": 455
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001368,
      "loss": 5.7149,
      "step": 456
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001371,
      "loss": 5.6703,
      "step": 457
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001374,
      "loss": 5.6551,
      "step": 458
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00013769999999999999,
      "loss": 5.6503,
      "step": 459
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.000138,
      "loss": 5.6532,
      "step": 460
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001383,
      "loss": 5.6759,
      "step": 461
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001386,
      "loss": 5.6403,
      "step": 462
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001389,
      "loss": 5.6361,
      "step": 463
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001392,
      "loss": 5.6027,
      "step": 464
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001395,
      "loss": 5.6436,
      "step": 465
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00013979999999999998,
      "loss": 5.6753,
      "step": 466
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001401,
      "loss": 5.6609,
      "step": 467
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001404,
      "loss": 5.6202,
      "step": 468
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00014069999999999998,
      "loss": 5.661,
      "step": 469
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00014099999999999998,
      "loss": 5.6272,
      "step": 470
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001413,
      "loss": 5.5941,
      "step": 471
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00014159999999999997,
      "loss": 5.5625,
      "step": 472
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00014189999999999998,
      "loss": 5.6037,
      "step": 473
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001422,
      "loss": 5.5885,
      "step": 474
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001425,
      "loss": 5.5818,
      "step": 475
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00014279999999999997,
      "loss": 5.665,
      "step": 476
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00014309999999999998,
      "loss": 5.6206,
      "step": 477
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001434,
      "loss": 5.5953,
      "step": 478
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00014369999999999997,
      "loss": 5.576,
      "step": 479
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00014399999999999998,
      "loss": 5.5506,
      "step": 480
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00014429999999999998,
      "loss": 5.6017,
      "step": 481
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001446,
      "loss": 5.6169,
      "step": 482
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00014489999999999997,
      "loss": 5.6171,
      "step": 483
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00014519999999999998,
      "loss": 5.5472,
      "step": 484
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00014549999999999999,
      "loss": 5.6092,
      "step": 485
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001458,
      "loss": 5.5975,
      "step": 486
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00014609999999999997,
      "loss": 5.5889,
      "step": 487
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00014639999999999998,
      "loss": 5.5462,
      "step": 488
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001467,
      "loss": 5.5345,
      "step": 489
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.000147,
      "loss": 5.5809,
      "step": 490
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00014729999999999998,
      "loss": 5.5681,
      "step": 491
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00014759999999999998,
      "loss": 5.639,
      "step": 492
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001479,
      "loss": 5.5547,
      "step": 493
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001482,
      "loss": 5.5284,
      "step": 494
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00014849999999999998,
      "loss": 5.557,
      "step": 495
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00014879999999999998,
      "loss": 5.5465,
      "step": 496
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001491,
      "loss": 5.5514,
      "step": 497
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001494,
      "loss": 5.5389,
      "step": 498
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00014969999999999998,
      "loss": 5.5197,
      "step": 499
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00015,
      "loss": 5.4984,
      "step": 500
    },
    {
      "epoch": 0.25,
      "eval_loss": 5.209429740905762,
      "eval_perplexity": 189.92770385742188,
      "eval_runtime": 46.9536,
      "eval_samples_per_second": 2129.763,
      "eval_steps_per_second": 66.555,
      "step": 500
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001503,
      "loss": 5.5614,
      "step": 501
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00015059999999999997,
      "loss": 5.5647,
      "step": 502
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00015089999999999998,
      "loss": 5.5128,
      "step": 503
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001512,
      "loss": 5.5092,
      "step": 504
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001515,
      "loss": 5.5328,
      "step": 505
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00015179999999999998,
      "loss": 5.5468,
      "step": 506
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00015209999999999998,
      "loss": 5.6035,
      "step": 507
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001524,
      "loss": 5.5067,
      "step": 508
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001527,
      "loss": 5.5219,
      "step": 509
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00015299999999999998,
      "loss": 5.4892,
      "step": 510
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00015329999999999999,
      "loss": 5.5009,
      "step": 511
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001536,
      "loss": 5.569,
      "step": 512
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001539,
      "loss": 5.5417,
      "step": 513
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00015419999999999998,
      "loss": 5.5095,
      "step": 514
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001545,
      "loss": 5.5455,
      "step": 515
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001548,
      "loss": 5.5567,
      "step": 516
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001551,
      "loss": 5.522,
      "step": 517
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00015539999999999998,
      "loss": 5.4761,
      "step": 518
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001557,
      "loss": 5.4973,
      "step": 519
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.000156,
      "loss": 5.58,
      "step": 520
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001563,
      "loss": 5.5154,
      "step": 521
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00015659999999999998,
      "loss": 5.4738,
      "step": 522
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001569,
      "loss": 5.431,
      "step": 523
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001572,
      "loss": 5.5447,
      "step": 524
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00015749999999999998,
      "loss": 5.5007,
      "step": 525
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001578,
      "loss": 5.4625,
      "step": 526
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001581,
      "loss": 5.4816,
      "step": 527
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001584,
      "loss": 5.4859,
      "step": 528
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00015869999999999998,
      "loss": 5.4734,
      "step": 529
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.000159,
      "loss": 5.4469,
      "step": 530
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001593,
      "loss": 5.4558,
      "step": 531
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001596,
      "loss": 5.5202,
      "step": 532
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00015989999999999998,
      "loss": 5.4743,
      "step": 533
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001602,
      "loss": 5.4849,
      "step": 534
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001605,
      "loss": 5.4497,
      "step": 535
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001608,
      "loss": 5.4564,
      "step": 536
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00016109999999999999,
      "loss": 5.4788,
      "step": 537
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001614,
      "loss": 5.4638,
      "step": 538
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001617,
      "loss": 5.4938,
      "step": 539
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.000162,
      "loss": 5.47,
      "step": 540
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001623,
      "loss": 5.4581,
      "step": 541
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001626,
      "loss": 5.4642,
      "step": 542
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001629,
      "loss": 5.503,
      "step": 543
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001632,
      "loss": 5.4886,
      "step": 544
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001635,
      "loss": 5.4773,
      "step": 545
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001638,
      "loss": 5.4303,
      "step": 546
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001641,
      "loss": 5.385,
      "step": 547
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001644,
      "loss": 5.4361,
      "step": 548
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001647,
      "loss": 5.4771,
      "step": 549
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.000165,
      "loss": 5.4719,
      "step": 550
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001653,
      "loss": 5.4371,
      "step": 551
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001656,
      "loss": 5.4577,
      "step": 552
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001659,
      "loss": 5.4885,
      "step": 553
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001662,
      "loss": 5.4563,
      "step": 554
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001665,
      "loss": 5.4411,
      "step": 555
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001668,
      "loss": 5.4415,
      "step": 556
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001671,
      "loss": 5.4331,
      "step": 557
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001674,
      "loss": 5.4186,
      "step": 558
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001677,
      "loss": 5.3755,
      "step": 559
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.000168,
      "loss": 5.4182,
      "step": 560
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001683,
      "loss": 5.4295,
      "step": 561
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001686,
      "loss": 5.4293,
      "step": 562
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00016889999999999996,
      "loss": 5.4286,
      "step": 563
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00016919999999999997,
      "loss": 5.4138,
      "step": 564
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00016949999999999997,
      "loss": 5.4009,
      "step": 565
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00016979999999999998,
      "loss": 5.3919,
      "step": 566
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00017009999999999996,
      "loss": 5.4125,
      "step": 567
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00017039999999999997,
      "loss": 5.4017,
      "step": 568
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00017069999999999998,
      "loss": 5.4021,
      "step": 569
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00017099999999999998,
      "loss": 5.4107,
      "step": 570
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00017129999999999996,
      "loss": 5.4379,
      "step": 571
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00017159999999999997,
      "loss": 5.3738,
      "step": 572
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00017189999999999998,
      "loss": 5.3541,
      "step": 573
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00017219999999999998,
      "loss": 5.342,
      "step": 574
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00017249999999999996,
      "loss": 5.3987,
      "step": 575
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00017279999999999997,
      "loss": 5.3671,
      "step": 576
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00017309999999999998,
      "loss": 5.3479,
      "step": 577
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017339999999999996,
      "loss": 5.3216,
      "step": 578
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017369999999999997,
      "loss": 5.4174,
      "step": 579
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017399999999999997,
      "loss": 5.4539,
      "step": 580
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017429999999999998,
      "loss": 5.4378,
      "step": 581
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017459999999999996,
      "loss": 5.4082,
      "step": 582
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017489999999999997,
      "loss": 5.383,
      "step": 583
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017519999999999998,
      "loss": 5.3478,
      "step": 584
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017549999999999998,
      "loss": 5.3286,
      "step": 585
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017579999999999996,
      "loss": 5.3557,
      "step": 586
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017609999999999997,
      "loss": 5.3801,
      "step": 587
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017639999999999998,
      "loss": 5.3468,
      "step": 588
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017669999999999999,
      "loss": 5.343,
      "step": 589
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017699999999999997,
      "loss": 5.3649,
      "step": 590
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017729999999999997,
      "loss": 5.3479,
      "step": 591
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017759999999999998,
      "loss": 5.3754,
      "step": 592
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0001779,
      "loss": 5.3866,
      "step": 593
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017819999999999997,
      "loss": 5.385,
      "step": 594
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017849999999999997,
      "loss": 5.3148,
      "step": 595
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017879999999999998,
      "loss": 5.3634,
      "step": 596
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0001791,
      "loss": 5.3466,
      "step": 597
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00017939999999999997,
      "loss": 5.3581,
      "step": 598
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00017969999999999998,
      "loss": 5.3336,
      "step": 599
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00017999999999999998,
      "loss": 5.3143,
      "step": 600
    },
    {
      "epoch": 0.3,
      "eval_loss": 5.0090532302856445,
      "eval_perplexity": 155.6486358642578,
      "eval_runtime": 252.6677,
      "eval_samples_per_second": 395.777,
      "eval_steps_per_second": 12.368,
      "step": 600
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018029999999999996,
      "loss": 5.352,
      "step": 601
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018059999999999997,
      "loss": 5.3792,
      "step": 602
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018089999999999998,
      "loss": 5.3183,
      "step": 603
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018119999999999999,
      "loss": 5.3597,
      "step": 604
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018149999999999997,
      "loss": 5.3758,
      "step": 605
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018179999999999997,
      "loss": 5.4077,
      "step": 606
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018209999999999998,
      "loss": 5.3193,
      "step": 607
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001824,
      "loss": 5.3396,
      "step": 608
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018269999999999997,
      "loss": 5.327,
      "step": 609
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018299999999999998,
      "loss": 5.3163,
      "step": 610
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018329999999999998,
      "loss": 5.3238,
      "step": 611
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001836,
      "loss": 5.3296,
      "step": 612
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018389999999999997,
      "loss": 5.2978,
      "step": 613
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018419999999999998,
      "loss": 5.2662,
      "step": 614
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018449999999999999,
      "loss": 5.3181,
      "step": 615
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001848,
      "loss": 5.3191,
      "step": 616
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018509999999999997,
      "loss": 5.3315,
      "step": 617
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018539999999999998,
      "loss": 5.331,
      "step": 618
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001857,
      "loss": 5.3058,
      "step": 619
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.000186,
      "loss": 5.2922,
      "step": 620
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018629999999999997,
      "loss": 5.3042,
      "step": 621
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018659999999999998,
      "loss": 5.2904,
      "step": 622
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001869,
      "loss": 5.2971,
      "step": 623
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001872,
      "loss": 5.3,
      "step": 624
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018749999999999998,
      "loss": 5.284,
      "step": 625
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018779999999999998,
      "loss": 5.273,
      "step": 626
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001881,
      "loss": 5.2989,
      "step": 627
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018839999999999997,
      "loss": 5.2936,
      "step": 628
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018869999999999998,
      "loss": 5.306,
      "step": 629
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018899999999999999,
      "loss": 5.3336,
      "step": 630
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001893,
      "loss": 5.2254,
      "step": 631
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018959999999999997,
      "loss": 5.2462,
      "step": 632
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018989999999999998,
      "loss": 5.2636,
      "step": 633
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001902,
      "loss": 5.2814,
      "step": 634
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001905,
      "loss": 5.287,
      "step": 635
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019079999999999998,
      "loss": 5.2987,
      "step": 636
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019109999999999998,
      "loss": 5.2919,
      "step": 637
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001914,
      "loss": 5.2718,
      "step": 638
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001917,
      "loss": 5.1902,
      "step": 639
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019199999999999998,
      "loss": 5.225,
      "step": 640
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019229999999999999,
      "loss": 5.2423,
      "step": 641
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001926,
      "loss": 5.2477,
      "step": 642
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001929,
      "loss": 5.2903,
      "step": 643
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019319999999999998,
      "loss": 5.2348,
      "step": 644
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001935,
      "loss": 5.2247,
      "step": 645
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001938,
      "loss": 5.25,
      "step": 646
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001941,
      "loss": 5.2249,
      "step": 647
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019439999999999998,
      "loss": 5.2375,
      "step": 648
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001947,
      "loss": 5.2607,
      "step": 649
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.000195,
      "loss": 5.2954,
      "step": 650
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019529999999999998,
      "loss": 5.293,
      "step": 651
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019559999999999998,
      "loss": 5.268,
      "step": 652
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001959,
      "loss": 5.2307,
      "step": 653
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001962,
      "loss": 5.1975,
      "step": 654
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019649999999999998,
      "loss": 5.2649,
      "step": 655
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019679999999999999,
      "loss": 5.2608,
      "step": 656
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001971,
      "loss": 5.2383,
      "step": 657
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001974,
      "loss": 5.2051,
      "step": 658
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019769999999999998,
      "loss": 5.2028,
      "step": 659
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.000198,
      "loss": 5.2124,
      "step": 660
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001983,
      "loss": 5.2123,
      "step": 661
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001986,
      "loss": 5.1987,
      "step": 662
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019889999999999998,
      "loss": 5.1943,
      "step": 663
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001992,
      "loss": 5.2265,
      "step": 664
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001995,
      "loss": 5.1989,
      "step": 665
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001998,
      "loss": 5.1964,
      "step": 666
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00020009999999999998,
      "loss": 5.2227,
      "step": 667
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002004,
      "loss": 5.2015,
      "step": 668
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002007,
      "loss": 5.1409,
      "step": 669
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.000201,
      "loss": 5.1971,
      "step": 670
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002013,
      "loss": 5.2304,
      "step": 671
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002016,
      "loss": 5.2254,
      "step": 672
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002019,
      "loss": 5.1783,
      "step": 673
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002022,
      "loss": 5.2126,
      "step": 674
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002025,
      "loss": 5.1856,
      "step": 675
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002028,
      "loss": 5.2046,
      "step": 676
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002031,
      "loss": 5.1773,
      "step": 677
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00020339999999999998,
      "loss": 5.1333,
      "step": 678
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002037,
      "loss": 5.1828,
      "step": 679
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.000204,
      "loss": 5.2085,
      "step": 680
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002043,
      "loss": 5.1954,
      "step": 681
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00020459999999999999,
      "loss": 5.1488,
      "step": 682
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002049,
      "loss": 5.1773,
      "step": 683
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002052,
      "loss": 5.2171,
      "step": 684
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002055,
      "loss": 5.2159,
      "step": 685
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002058,
      "loss": 5.1852,
      "step": 686
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002061,
      "loss": 5.1892,
      "step": 687
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00020639999999999998,
      "loss": 5.1829,
      "step": 688
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00020669999999999996,
      "loss": 5.192,
      "step": 689
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00020699999999999996,
      "loss": 5.1885,
      "step": 690
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00020729999999999997,
      "loss": 5.1745,
      "step": 691
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00020759999999999998,
      "loss": 5.1509,
      "step": 692
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00020789999999999996,
      "loss": 5.1369,
      "step": 693
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00020819999999999996,
      "loss": 5.246,
      "step": 694
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00020849999999999997,
      "loss": 5.2076,
      "step": 695
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00020879999999999998,
      "loss": 5.154,
      "step": 696
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00020909999999999996,
      "loss": 5.1081,
      "step": 697
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00020939999999999997,
      "loss": 5.1648,
      "step": 698
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00020969999999999997,
      "loss": 5.1876,
      "step": 699
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00020999999999999998,
      "loss": 5.1901,
      "step": 700
    },
    {
      "epoch": 0.35,
      "eval_loss": 4.838101863861084,
      "eval_perplexity": 131.38815307617188,
      "eval_runtime": 47.2408,
      "eval_samples_per_second": 2116.814,
      "eval_steps_per_second": 66.15,
      "step": 700
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021029999999999996,
      "loss": 5.162,
      "step": 701
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021059999999999997,
      "loss": 5.1363,
      "step": 702
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021089999999999998,
      "loss": 5.1535,
      "step": 703
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021119999999999996,
      "loss": 5.1477,
      "step": 704
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021149999999999996,
      "loss": 5.107,
      "step": 705
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021179999999999997,
      "loss": 5.1919,
      "step": 706
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021209999999999998,
      "loss": 5.1927,
      "step": 707
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021239999999999996,
      "loss": 5.1696,
      "step": 708
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021269999999999997,
      "loss": 5.1792,
      "step": 709
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021299999999999997,
      "loss": 5.1488,
      "step": 710
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021329999999999998,
      "loss": 5.1706,
      "step": 711
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021359999999999996,
      "loss": 5.1645,
      "step": 712
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021389999999999997,
      "loss": 5.1883,
      "step": 713
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021419999999999998,
      "loss": 5.2007,
      "step": 714
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021449999999999998,
      "loss": 5.1649,
      "step": 715
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021479999999999996,
      "loss": 5.1611,
      "step": 716
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021509999999999997,
      "loss": 5.1929,
      "step": 717
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021539999999999998,
      "loss": 5.1354,
      "step": 718
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00021569999999999998,
      "loss": 5.0996,
      "step": 719
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00021599999999999996,
      "loss": 5.1231,
      "step": 720
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00021629999999999997,
      "loss": 5.1662,
      "step": 721
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00021659999999999998,
      "loss": 5.1025,
      "step": 722
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002169,
      "loss": 5.1083,
      "step": 723
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00021719999999999997,
      "loss": 5.1139,
      "step": 724
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00021749999999999997,
      "loss": 5.0901,
      "step": 725
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00021779999999999998,
      "loss": 5.0921,
      "step": 726
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00021809999999999996,
      "loss": 5.0778,
      "step": 727
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00021839999999999997,
      "loss": 5.1057,
      "step": 728
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00021869999999999998,
      "loss": 5.1658,
      "step": 729
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00021899999999999998,
      "loss": 5.0783,
      "step": 730
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00021929999999999996,
      "loss": 5.0922,
      "step": 731
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00021959999999999997,
      "loss": 5.0884,
      "step": 732
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00021989999999999998,
      "loss": 5.0537,
      "step": 733
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022019999999999999,
      "loss": 5.0421,
      "step": 734
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022049999999999997,
      "loss": 5.0471,
      "step": 735
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022079999999999997,
      "loss": 5.0957,
      "step": 736
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022109999999999998,
      "loss": 5.0782,
      "step": 737
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002214,
      "loss": 5.0794,
      "step": 738
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00022169999999999997,
      "loss": 5.0584,
      "step": 739
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00022199999999999998,
      "loss": 5.0228,
      "step": 740
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00022229999999999998,
      "loss": 5.0229,
      "step": 741
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002226,
      "loss": 5.0852,
      "step": 742
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00022289999999999997,
      "loss": 5.1055,
      "step": 743
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00022319999999999998,
      "loss": 5.1162,
      "step": 744
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00022349999999999998,
      "loss": 5.0746,
      "step": 745
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002238,
      "loss": 5.0535,
      "step": 746
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00022409999999999997,
      "loss": 5.0686,
      "step": 747
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00022439999999999998,
      "loss": 5.077,
      "step": 748
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002247,
      "loss": 5.0957,
      "step": 749
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.000225,
      "loss": 5.0532,
      "step": 750
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00022529999999999997,
      "loss": 5.0774,
      "step": 751
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00022559999999999998,
      "loss": 5.1037,
      "step": 752
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002259,
      "loss": 5.0532,
      "step": 753
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00022619999999999997,
      "loss": 5.054,
      "step": 754
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00022649999999999998,
      "loss": 5.0547,
      "step": 755
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00022679999999999998,
      "loss": 5.0335,
      "step": 756
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002271,
      "loss": 5.0383,
      "step": 757
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00022739999999999997,
      "loss": 5.1104,
      "step": 758
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00022769999999999998,
      "loss": 5.118,
      "step": 759
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00022799999999999999,
      "loss": 5.0724,
      "step": 760
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002283,
      "loss": 5.0317,
      "step": 761
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00022859999999999997,
      "loss": 5.0413,
      "step": 762
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00022889999999999998,
      "loss": 5.0506,
      "step": 763
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002292,
      "loss": 5.0505,
      "step": 764
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002295,
      "loss": 5.0566,
      "step": 765
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00022979999999999997,
      "loss": 5.0312,
      "step": 766
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00023009999999999998,
      "loss": 5.073,
      "step": 767
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002304,
      "loss": 5.0955,
      "step": 768
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002307,
      "loss": 5.0716,
      "step": 769
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00023099999999999998,
      "loss": 5.0646,
      "step": 770
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00023129999999999998,
      "loss": 5.021,
      "step": 771
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002316,
      "loss": 4.9615,
      "step": 772
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002319,
      "loss": 5.0506,
      "step": 773
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00023219999999999998,
      "loss": 5.0452,
      "step": 774
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00023249999999999999,
      "loss": 5.0465,
      "step": 775
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002328,
      "loss": 5.0296,
      "step": 776
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00023309999999999997,
      "loss": 5.0215,
      "step": 777
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00023339999999999998,
      "loss": 4.9976,
      "step": 778
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002337,
      "loss": 4.9959,
      "step": 779
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.000234,
      "loss": 5.0255,
      "step": 780
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00023429999999999998,
      "loss": 5.0329,
      "step": 781
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00023459999999999998,
      "loss": 5.0002,
      "step": 782
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002349,
      "loss": 4.9756,
      "step": 783
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002352,
      "loss": 4.9813,
      "step": 784
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00023549999999999998,
      "loss": 5.059,
      "step": 785
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00023579999999999999,
      "loss": 4.9748,
      "step": 786
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002361,
      "loss": 4.9578,
      "step": 787
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002364,
      "loss": 4.9555,
      "step": 788
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00023669999999999998,
      "loss": 4.9601,
      "step": 789
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.000237,
      "loss": 4.9726,
      "step": 790
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002373,
      "loss": 4.9592,
      "step": 791
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002376,
      "loss": 4.9603,
      "step": 792
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00023789999999999998,
      "loss": 4.9651,
      "step": 793
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002382,
      "loss": 5.0196,
      "step": 794
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002385,
      "loss": 5.0344,
      "step": 795
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002388,
      "loss": 5.002,
      "step": 796
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00023909999999999998,
      "loss": 4.9393,
      "step": 797
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002394,
      "loss": 4.9777,
      "step": 798
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002397,
      "loss": 4.9731,
      "step": 799
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00023999999999999998,
      "loss": 4.9959,
      "step": 800
    },
    {
      "epoch": 0.39,
      "eval_loss": 4.695063591003418,
      "eval_perplexity": 113.9029312133789,
      "eval_runtime": 49.1317,
      "eval_samples_per_second": 2035.347,
      "eval_steps_per_second": 63.605,
      "step": 800
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00024029999999999999,
      "loss": 4.9919,
      "step": 801
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002406,
      "loss": 4.9769,
      "step": 802
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002409,
      "loss": 4.9649,
      "step": 803
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00024119999999999998,
      "loss": 5.0252,
      "step": 804
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002415,
      "loss": 4.9033,
      "step": 805
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002418,
      "loss": 4.8893,
      "step": 806
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002421,
      "loss": 4.9099,
      "step": 807
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00024239999999999998,
      "loss": 4.9895,
      "step": 808
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002427,
      "loss": 5.0208,
      "step": 809
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.000243,
      "loss": 4.9822,
      "step": 810
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002433,
      "loss": 4.9528,
      "step": 811
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00024359999999999999,
      "loss": 4.9376,
      "step": 812
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00024389999999999997,
      "loss": 4.9317,
      "step": 813
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00024419999999999997,
      "loss": 4.9308,
      "step": 814
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002445,
      "loss": 4.9345,
      "step": 815
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002448,
      "loss": 5.0027,
      "step": 816
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00024509999999999994,
      "loss": 4.9322,
      "step": 817
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00024539999999999995,
      "loss": 4.9443,
      "step": 818
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00024569999999999995,
      "loss": 4.9532,
      "step": 819
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00024599999999999996,
      "loss": 4.9724,
      "step": 820
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00024629999999999997,
      "loss": 4.9363,
      "step": 821
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002466,
      "loss": 4.9565,
      "step": 822
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002469,
      "loss": 4.9905,
      "step": 823
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002472,
      "loss": 4.9406,
      "step": 824
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00024749999999999994,
      "loss": 4.9344,
      "step": 825
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00024779999999999995,
      "loss": 4.9086,
      "step": 826
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00024809999999999996,
      "loss": 4.8726,
      "step": 827
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00024839999999999997,
      "loss": 4.9949,
      "step": 828
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002487,
      "loss": 4.9945,
      "step": 829
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.000249,
      "loss": 4.9861,
      "step": 830
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002493,
      "loss": 4.9962,
      "step": 831
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00024959999999999994,
      "loss": 4.9248,
      "step": 832
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00024989999999999995,
      "loss": 4.9576,
      "step": 833
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00025019999999999996,
      "loss": 4.9649,
      "step": 834
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00025049999999999996,
      "loss": 4.9779,
      "step": 835
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00025079999999999997,
      "loss": 4.9462,
      "step": 836
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002511,
      "loss": 4.923,
      "step": 837
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002514,
      "loss": 4.9478,
      "step": 838
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002517,
      "loss": 4.9513,
      "step": 839
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00025199999999999995,
      "loss": 4.9501,
      "step": 840
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00025229999999999995,
      "loss": 4.9232,
      "step": 841
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00025259999999999996,
      "loss": 4.8662,
      "step": 842
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00025289999999999997,
      "loss": 4.902,
      "step": 843
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002532,
      "loss": 4.9198,
      "step": 844
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002535,
      "loss": 4.9454,
      "step": 845
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002538,
      "loss": 4.9462,
      "step": 846
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002541,
      "loss": 4.9342,
      "step": 847
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00025439999999999995,
      "loss": 4.9679,
      "step": 848
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00025469999999999996,
      "loss": 4.9111,
      "step": 849
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00025499999999999996,
      "loss": 4.9267,
      "step": 850
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00025529999999999997,
      "loss": 4.9339,
      "step": 851
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002556,
      "loss": 4.9389,
      "step": 852
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002559,
      "loss": 4.9759,
      "step": 853
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002562,
      "loss": 4.94,
      "step": 854
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00025649999999999995,
      "loss": 4.8969,
      "step": 855
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00025679999999999995,
      "loss": 4.8685,
      "step": 856
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00025709999999999996,
      "loss": 4.8964,
      "step": 857
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00025739999999999997,
      "loss": 4.9007,
      "step": 858
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002577,
      "loss": 4.8836,
      "step": 859
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.000258,
      "loss": 4.895,
      "step": 860
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002583,
      "loss": 4.8793,
      "step": 861
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002586,
      "loss": 4.8882,
      "step": 862
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00025889999999999995,
      "loss": 4.9012,
      "step": 863
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00025919999999999996,
      "loss": 4.8946,
      "step": 864
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00025949999999999997,
      "loss": 4.8595,
      "step": 865
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00025979999999999997,
      "loss": 4.8628,
      "step": 866
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002601,
      "loss": 4.9034,
      "step": 867
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002604,
      "loss": 4.927,
      "step": 868
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002607,
      "loss": 4.9085,
      "step": 869
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.000261,
      "loss": 4.8805,
      "step": 870
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00026129999999999995,
      "loss": 4.8798,
      "step": 871
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00026159999999999996,
      "loss": 4.9003,
      "step": 872
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00026189999999999997,
      "loss": 4.8542,
      "step": 873
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002622,
      "loss": 4.8432,
      "step": 874
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002625,
      "loss": 4.8796,
      "step": 875
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002628,
      "loss": 4.8879,
      "step": 876
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002631,
      "loss": 4.8702,
      "step": 877
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00026339999999999995,
      "loss": 4.8683,
      "step": 878
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00026369999999999996,
      "loss": 4.8386,
      "step": 879
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00026399999999999997,
      "loss": 4.8195,
      "step": 880
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002643,
      "loss": 4.873,
      "step": 881
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002646,
      "loss": 4.8662,
      "step": 882
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002649,
      "loss": 4.8346,
      "step": 883
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002652,
      "loss": 4.8827,
      "step": 884
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002655,
      "loss": 4.8859,
      "step": 885
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00026579999999999996,
      "loss": 4.8628,
      "step": 886
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00026609999999999996,
      "loss": 4.8311,
      "step": 887
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00026639999999999997,
      "loss": 4.8441,
      "step": 888
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002667,
      "loss": 4.8451,
      "step": 889
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.000267,
      "loss": 4.8425,
      "step": 890
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002673,
      "loss": 4.7954,
      "step": 891
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002676,
      "loss": 4.859,
      "step": 892
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002679,
      "loss": 4.8321,
      "step": 893
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00026819999999999996,
      "loss": 4.8303,
      "step": 894
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00026849999999999997,
      "loss": 4.8196,
      "step": 895
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002688,
      "loss": 4.8296,
      "step": 896
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002691,
      "loss": 4.8474,
      "step": 897
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002694,
      "loss": 4.8597,
      "step": 898
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002697,
      "loss": 4.8144,
      "step": 899
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00027,
      "loss": 4.84,
      "step": 900
    },
    {
      "epoch": 0.44,
      "eval_loss": 4.570080280303955,
      "eval_perplexity": 100.52706909179688,
      "eval_runtime": 194.1151,
      "eval_samples_per_second": 515.158,
      "eval_steps_per_second": 16.099,
      "step": 900
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00027029999999999996,
      "loss": 4.8321,
      "step": 901
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00027059999999999996,
      "loss": 4.8365,
      "step": 902
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00027089999999999997,
      "loss": 4.7744,
      "step": 903
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002712,
      "loss": 4.8448,
      "step": 904
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002715,
      "loss": 4.8124,
      "step": 905
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002718,
      "loss": 4.7964,
      "step": 906
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002721,
      "loss": 4.7892,
      "step": 907
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002724,
      "loss": 4.8744,
      "step": 908
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00027269999999999996,
      "loss": 4.8578,
      "step": 909
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00027299999999999997,
      "loss": 4.8504,
      "step": 910
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002733,
      "loss": 4.8396,
      "step": 911
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002736,
      "loss": 4.8377,
      "step": 912
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002739,
      "loss": 4.8548,
      "step": 913
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002742,
      "loss": 4.845,
      "step": 914
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002745,
      "loss": 4.7865,
      "step": 915
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002748,
      "loss": 4.8272,
      "step": 916
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00027509999999999996,
      "loss": 4.855,
      "step": 917
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00027539999999999997,
      "loss": 4.8344,
      "step": 918
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002757,
      "loss": 4.7945,
      "step": 919
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.000276,
      "loss": 4.7755,
      "step": 920
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002763,
      "loss": 4.8222,
      "step": 921
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002766,
      "loss": 4.8175,
      "step": 922
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002769,
      "loss": 4.8128,
      "step": 923
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002772,
      "loss": 4.7601,
      "step": 924
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00027749999999999997,
      "loss": 4.7856,
      "step": 925
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002778,
      "loss": 4.7842,
      "step": 926
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002781,
      "loss": 4.7607,
      "step": 927
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002784,
      "loss": 4.7637,
      "step": 928
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002787,
      "loss": 4.7758,
      "step": 929
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.000279,
      "loss": 4.8062,
      "step": 930
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002793,
      "loss": 4.776,
      "step": 931
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00027959999999999997,
      "loss": 4.8185,
      "step": 932
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002799,
      "loss": 4.8192,
      "step": 933
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002802,
      "loss": 4.8053,
      "step": 934
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002805,
      "loss": 4.7579,
      "step": 935
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002808,
      "loss": 4.7615,
      "step": 936
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002811,
      "loss": 4.7722,
      "step": 937
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00028139999999999996,
      "loss": 4.7742,
      "step": 938
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00028169999999999996,
      "loss": 4.7298,
      "step": 939
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00028199999999999997,
      "loss": 4.7981,
      "step": 940
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002823,
      "loss": 4.7678,
      "step": 941
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002826,
      "loss": 4.7884,
      "step": 942
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00028289999999999994,
      "loss": 4.7372,
      "step": 943
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00028319999999999994,
      "loss": 4.7507,
      "step": 944
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00028349999999999995,
      "loss": 4.7435,
      "step": 945
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00028379999999999996,
      "loss": 4.7501,
      "step": 946
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00028409999999999997,
      "loss": 4.8007,
      "step": 947
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002844,
      "loss": 4.7191,
      "step": 948
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002847,
      "loss": 4.7213,
      "step": 949
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.000285,
      "loss": 4.7674,
      "step": 950
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00028529999999999994,
      "loss": 4.7534,
      "step": 951
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00028559999999999995,
      "loss": 4.7947,
      "step": 952
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00028589999999999996,
      "loss": 4.8096,
      "step": 953
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00028619999999999996,
      "loss": 4.793,
      "step": 954
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00028649999999999997,
      "loss": 4.7386,
      "step": 955
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002868,
      "loss": 4.735,
      "step": 956
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002871,
      "loss": 4.7261,
      "step": 957
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00028739999999999994,
      "loss": 4.7441,
      "step": 958
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00028769999999999995,
      "loss": 4.7341,
      "step": 959
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00028799999999999995,
      "loss": 4.7379,
      "step": 960
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00028829999999999996,
      "loss": 4.7587,
      "step": 961
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00028859999999999997,
      "loss": 4.778,
      "step": 962
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002889,
      "loss": 4.748,
      "step": 963
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002892,
      "loss": 4.7715,
      "step": 964
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002895,
      "loss": 4.7615,
      "step": 965
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00028979999999999994,
      "loss": 4.7454,
      "step": 966
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00029009999999999995,
      "loss": 4.6778,
      "step": 967
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00029039999999999996,
      "loss": 4.7334,
      "step": 968
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00029069999999999996,
      "loss": 4.7587,
      "step": 969
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00029099999999999997,
      "loss": 4.7375,
      "step": 970
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002913,
      "loss": 4.7211,
      "step": 971
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002916,
      "loss": 4.7369,
      "step": 972
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002919,
      "loss": 4.755,
      "step": 973
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00029219999999999995,
      "loss": 4.7683,
      "step": 974
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00029249999999999995,
      "loss": 4.7617,
      "step": 975
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00029279999999999996,
      "loss": 4.713,
      "step": 976
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00029309999999999997,
      "loss": 4.723,
      "step": 977
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002934,
      "loss": 4.7351,
      "step": 978
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002937,
      "loss": 4.7132,
      "step": 979
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.000294,
      "loss": 4.745,
      "step": 980
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00029429999999999994,
      "loss": 4.752,
      "step": 981
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00029459999999999995,
      "loss": 4.7197,
      "step": 982
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00029489999999999996,
      "loss": 4.6821,
      "step": 983
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00029519999999999997,
      "loss": 4.6966,
      "step": 984
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00029549999999999997,
      "loss": 4.6908,
      "step": 985
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002958,
      "loss": 4.6917,
      "step": 986
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002961,
      "loss": 4.6559,
      "step": 987
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002964,
      "loss": 4.69,
      "step": 988
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00029669999999999995,
      "loss": 4.6989,
      "step": 989
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00029699999999999996,
      "loss": 4.7157,
      "step": 990
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00029729999999999996,
      "loss": 4.7251,
      "step": 991
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00029759999999999997,
      "loss": 4.6688,
      "step": 992
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002979,
      "loss": 4.6792,
      "step": 993
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002982,
      "loss": 4.7025,
      "step": 994
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002985,
      "loss": 4.7192,
      "step": 995
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002988,
      "loss": 4.7177,
      "step": 996
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00029909999999999995,
      "loss": 4.7192,
      "step": 997
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00029939999999999996,
      "loss": 4.6799,
      "step": 998
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00029969999999999997,
      "loss": 4.7056,
      "step": 999
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0003,
      "loss": 4.7068,
      "step": 1000
    },
    {
      "epoch": 0.49,
      "eval_loss": 4.45216178894043,
      "eval_perplexity": 89.42424774169922,
      "eval_runtime": 47.3036,
      "eval_samples_per_second": 2114.005,
      "eval_steps_per_second": 66.063,
      "step": 1000
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0003002999999999999,
      "loss": 4.6864,
      "step": 1001
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0003006,
      "loss": 4.618,
      "step": 1002
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00030089999999999994,
      "loss": 4.6605,
      "step": 1003
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00030119999999999995,
      "loss": 4.7112,
      "step": 1004
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00030149999999999996,
      "loss": 4.7177,
      "step": 1005
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00030179999999999996,
      "loss": 4.7397,
      "step": 1006
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0003020999999999999,
      "loss": 4.7121,
      "step": 1007
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0003024,
      "loss": 4.7259,
      "step": 1008
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00030269999999999993,
      "loss": 4.7203,
      "step": 1009
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.000303,
      "loss": 4.7053,
      "step": 1010
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00030329999999999995,
      "loss": 4.702,
      "step": 1011
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00030359999999999995,
      "loss": 4.6908,
      "step": 1012
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00030389999999999996,
      "loss": 4.6777,
      "step": 1013
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00030419999999999997,
      "loss": 4.6543,
      "step": 1014
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0003044999999999999,
      "loss": 4.6555,
      "step": 1015
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0003048,
      "loss": 4.6516,
      "step": 1016
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00030509999999999994,
      "loss": 4.6559,
      "step": 1017
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0003054,
      "loss": 4.6827,
      "step": 1018
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00030569999999999995,
      "loss": 4.6688,
      "step": 1019
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00030599999999999996,
      "loss": 4.6383,
      "step": 1020
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00030629999999999996,
      "loss": 4.6231,
      "step": 1021
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00030659999999999997,
      "loss": 4.6663,
      "step": 1022
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0003068999999999999,
      "loss": 4.6159,
      "step": 1023
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0003072,
      "loss": 4.641,
      "step": 1024
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00030749999999999994,
      "loss": 4.6437,
      "step": 1025
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0003078,
      "loss": 4.606,
      "step": 1026
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00030809999999999995,
      "loss": 4.7025,
      "step": 1027
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00030839999999999996,
      "loss": 4.7058,
      "step": 1028
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00030869999999999997,
      "loss": 4.6772,
      "step": 1029
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.000309,
      "loss": 4.719,
      "step": 1030
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00030929999999999993,
      "loss": 4.7004,
      "step": 1031
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0003096,
      "loss": 4.6517,
      "step": 1032
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00030989999999999994,
      "loss": 4.6574,
      "step": 1033
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0003102,
      "loss": 4.6242,
      "step": 1034
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00031049999999999996,
      "loss": 4.7013,
      "step": 1035
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00031079999999999997,
      "loss": 4.6954,
      "step": 1036
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00031109999999999997,
      "loss": 4.6558,
      "step": 1037
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0003114,
      "loss": 4.7032,
      "step": 1038
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00031169999999999993,
      "loss": 4.6547,
      "step": 1039
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.000312,
      "loss": 4.6887,
      "step": 1040
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00031229999999999995,
      "loss": 4.673,
      "step": 1041
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0003126,
      "loss": 4.6467,
      "step": 1042
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00031289999999999996,
      "loss": 4.6931,
      "step": 1043
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00031319999999999997,
      "loss": 4.6599,
      "step": 1044
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0003135,
      "loss": 4.6357,
      "step": 1045
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0003138,
      "loss": 4.638,
      "step": 1046
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00031409999999999994,
      "loss": 4.6345,
      "step": 1047
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0003144,
      "loss": 4.6414,
      "step": 1048
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00031469999999999995,
      "loss": 4.6273,
      "step": 1049
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00031499999999999996,
      "loss": 4.6096,
      "step": 1050
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00031529999999999997,
      "loss": 4.6274,
      "step": 1051
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0003156,
      "loss": 4.6193,
      "step": 1052
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0003158999999999999,
      "loss": 4.6152,
      "step": 1053
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0003162,
      "loss": 4.643,
      "step": 1054
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00031649999999999994,
      "loss": 4.5823,
      "step": 1055
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0003168,
      "loss": 4.5882,
      "step": 1056
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00031709999999999996,
      "loss": 4.6284,
      "step": 1057
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00031739999999999996,
      "loss": 4.5641,
      "step": 1058
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00031769999999999997,
      "loss": 4.6227,
      "step": 1059
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.000318,
      "loss": 4.6474,
      "step": 1060
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00031829999999999993,
      "loss": 4.6791,
      "step": 1061
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0003186,
      "loss": 4.5988,
      "step": 1062
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00031889999999999995,
      "loss": 4.6018,
      "step": 1063
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0003192,
      "loss": 4.5976,
      "step": 1064
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00031949999999999996,
      "loss": 4.5839,
      "step": 1065
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00031979999999999997,
      "loss": 4.6696,
      "step": 1066
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0003201,
      "loss": 4.6214,
      "step": 1067
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0003204,
      "loss": 4.599,
      "step": 1068
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00032069999999999993,
      "loss": 4.5393,
      "step": 1069
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.000321,
      "loss": 4.6813,
      "step": 1070
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00032129999999999995,
      "loss": 4.6642,
      "step": 1071
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0003216,
      "loss": 4.6548,
      "step": 1072
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00032189999999999996,
      "loss": 4.6532,
      "step": 1073
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00032219999999999997,
      "loss": 4.6129,
      "step": 1074
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0003225,
      "loss": 4.6175,
      "step": 1075
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0003228,
      "loss": 4.6039,
      "step": 1076
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00032309999999999994,
      "loss": 4.6161,
      "step": 1077
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0003234,
      "loss": 4.6652,
      "step": 1078
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00032369999999999995,
      "loss": 4.6137,
      "step": 1079
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.000324,
      "loss": 4.615,
      "step": 1080
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00032429999999999997,
      "loss": 4.5687,
      "step": 1081
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0003246,
      "loss": 4.6799,
      "step": 1082
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0003249,
      "loss": 4.6248,
      "step": 1083
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003252,
      "loss": 4.602,
      "step": 1084
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00032549999999999994,
      "loss": 4.6051,
      "step": 1085
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003258,
      "loss": 4.4821,
      "step": 1086
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00032609999999999996,
      "loss": 4.5486,
      "step": 1087
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003264,
      "loss": 4.6062,
      "step": 1088
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00032669999999999997,
      "loss": 4.626,
      "step": 1089
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.000327,
      "loss": 4.5474,
      "step": 1090
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003273,
      "loss": 4.5487,
      "step": 1091
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003276,
      "loss": 4.5813,
      "step": 1092
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00032789999999999995,
      "loss": 4.5764,
      "step": 1093
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003282,
      "loss": 4.6078,
      "step": 1094
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00032849999999999996,
      "loss": 4.5834,
      "step": 1095
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003288,
      "loss": 4.5755,
      "step": 1096
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003291,
      "loss": 4.6358,
      "step": 1097
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003294,
      "loss": 4.5996,
      "step": 1098
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003297,
      "loss": 4.5649,
      "step": 1099
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00033,
      "loss": 4.5655,
      "step": 1100
    },
    {
      "epoch": 0.54,
      "eval_loss": 4.351374626159668,
      "eval_perplexity": 80.82051086425781,
      "eval_runtime": 47.5538,
      "eval_samples_per_second": 2102.881,
      "eval_steps_per_second": 65.715,
      "step": 1100
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00033029999999999995,
      "loss": 4.5765,
      "step": 1101
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003306,
      "loss": 4.5457,
      "step": 1102
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00033089999999999997,
      "loss": 4.5068,
      "step": 1103
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003312,
      "loss": 4.5011,
      "step": 1104
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003315,
      "loss": 4.594,
      "step": 1105
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003318,
      "loss": 4.5603,
      "step": 1106
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00033209999999999994,
      "loss": 4.58,
      "step": 1107
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003324,
      "loss": 4.6267,
      "step": 1108
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00033269999999999996,
      "loss": 4.57,
      "step": 1109
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.000333,
      "loss": 4.5409,
      "step": 1110
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00033329999999999997,
      "loss": 4.603,
      "step": 1111
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003336,
      "loss": 4.6185,
      "step": 1112
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003339,
      "loss": 4.5293,
      "step": 1113
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003342,
      "loss": 4.5898,
      "step": 1114
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00033449999999999994,
      "loss": 4.5894,
      "step": 1115
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003348,
      "loss": 4.6179,
      "step": 1116
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00033509999999999996,
      "loss": 4.6191,
      "step": 1117
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003354,
      "loss": 4.6324,
      "step": 1118
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003357,
      "loss": 4.5903,
      "step": 1119
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.000336,
      "loss": 4.5847,
      "step": 1120
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003363,
      "loss": 4.586,
      "step": 1121
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003366,
      "loss": 4.6117,
      "step": 1122
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00033689999999999995,
      "loss": 4.5785,
      "step": 1123
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003372,
      "loss": 4.573,
      "step": 1124
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00033749999999999996,
      "loss": 4.6088,
      "step": 1125
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0003377999999999999,
      "loss": 4.5901,
      "step": 1126
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0003381,
      "loss": 4.5562,
      "step": 1127
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00033839999999999993,
      "loss": 4.5683,
      "step": 1128
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0003387,
      "loss": 4.5182,
      "step": 1129
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00033899999999999995,
      "loss": 4.5252,
      "step": 1130
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00033929999999999995,
      "loss": 4.5544,
      "step": 1131
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00033959999999999996,
      "loss": 4.6067,
      "step": 1132
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00033989999999999997,
      "loss": 4.5291,
      "step": 1133
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0003401999999999999,
      "loss": 4.5222,
      "step": 1134
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0003405,
      "loss": 4.5223,
      "step": 1135
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00034079999999999994,
      "loss": 4.5112,
      "step": 1136
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0003411,
      "loss": 4.5742,
      "step": 1137
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00034139999999999995,
      "loss": 4.5287,
      "step": 1138
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00034169999999999996,
      "loss": 4.561,
      "step": 1139
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00034199999999999996,
      "loss": 4.5672,
      "step": 1140
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00034229999999999997,
      "loss": 4.552,
      "step": 1141
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0003425999999999999,
      "loss": 4.5356,
      "step": 1142
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0003429,
      "loss": 4.5389,
      "step": 1143
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00034319999999999994,
      "loss": 4.5705,
      "step": 1144
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0003435,
      "loss": 4.5895,
      "step": 1145
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034379999999999995,
      "loss": 4.561,
      "step": 1146
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034409999999999996,
      "loss": 4.5845,
      "step": 1147
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034439999999999997,
      "loss": 4.6152,
      "step": 1148
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0003447,
      "loss": 4.5529,
      "step": 1149
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034499999999999993,
      "loss": 4.5271,
      "step": 1150
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0003453,
      "loss": 4.5254,
      "step": 1151
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034559999999999994,
      "loss": 4.5748,
      "step": 1152
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034589999999999995,
      "loss": 4.5536,
      "step": 1153
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034619999999999996,
      "loss": 4.487,
      "step": 1154
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034649999999999997,
      "loss": 4.4792,
      "step": 1155
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0003467999999999999,
      "loss": 4.4861,
      "step": 1156
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0003471,
      "loss": 4.5439,
      "step": 1157
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034739999999999993,
      "loss": 4.5244,
      "step": 1158
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0003477,
      "loss": 4.509,
      "step": 1159
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034799999999999995,
      "loss": 4.5349,
      "step": 1160
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034829999999999996,
      "loss": 4.5521,
      "step": 1161
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034859999999999996,
      "loss": 4.5471,
      "step": 1162
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00034889999999999997,
      "loss": 4.5182,
      "step": 1163
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0003491999999999999,
      "loss": 4.4779,
      "step": 1164
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0003495,
      "loss": 4.5219,
      "step": 1165
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00034979999999999994,
      "loss": 4.5242,
      "step": 1166
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0003501,
      "loss": 4.5013,
      "step": 1167
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00035039999999999995,
      "loss": 4.4316,
      "step": 1168
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00035069999999999996,
      "loss": 4.4628,
      "step": 1169
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00035099999999999997,
      "loss": 4.502,
      "step": 1170
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0003513,
      "loss": 4.5184,
      "step": 1171
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0003515999999999999,
      "loss": 4.4745,
      "step": 1172
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0003519,
      "loss": 4.5545,
      "step": 1173
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00035219999999999994,
      "loss": 4.4844,
      "step": 1174
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0003525,
      "loss": 4.4978,
      "step": 1175
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00035279999999999996,
      "loss": 4.5331,
      "step": 1176
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00035309999999999996,
      "loss": 4.5795,
      "step": 1177
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00035339999999999997,
      "loss": 4.501,
      "step": 1178
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0003537,
      "loss": 4.4909,
      "step": 1179
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00035399999999999993,
      "loss": 4.5395,
      "step": 1180
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0003543,
      "loss": 4.4806,
      "step": 1181
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00035459999999999995,
      "loss": 4.4867,
      "step": 1182
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0003549,
      "loss": 4.4889,
      "step": 1183
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00035519999999999996,
      "loss": 4.5166,
      "step": 1184
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00035549999999999997,
      "loss": 4.5299,
      "step": 1185
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0003558,
      "loss": 4.5145,
      "step": 1186
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0003561,
      "loss": 4.4892,
      "step": 1187
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00035639999999999994,
      "loss": 4.4762,
      "step": 1188
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0003567,
      "loss": 4.5345,
      "step": 1189
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00035699999999999995,
      "loss": 4.5065,
      "step": 1190
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0003573,
      "loss": 4.4862,
      "step": 1191
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00035759999999999996,
      "loss": 4.5081,
      "step": 1192
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00035789999999999997,
      "loss": 4.4883,
      "step": 1193
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0003582,
      "loss": 4.4711,
      "step": 1194
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0003585,
      "loss": 4.4842,
      "step": 1195
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00035879999999999994,
      "loss": 4.5092,
      "step": 1196
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0003591,
      "loss": 4.4374,
      "step": 1197
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00035939999999999995,
      "loss": 4.4841,
      "step": 1198
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00035969999999999996,
      "loss": 4.4994,
      "step": 1199
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00035999999999999997,
      "loss": 4.4925,
      "step": 1200
    },
    {
      "epoch": 0.59,
      "eval_loss": 4.266104698181152,
      "eval_perplexity": 74.24376678466797,
      "eval_runtime": 47.4774,
      "eval_samples_per_second": 2106.267,
      "eval_steps_per_second": 65.821,
      "step": 1200
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0003603,
      "loss": 4.4752,
      "step": 1201
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00036059999999999993,
      "loss": 4.4774,
      "step": 1202
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0003609,
      "loss": 4.4671,
      "step": 1203
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00036119999999999994,
      "loss": 4.4223,
      "step": 1204
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0003615,
      "loss": 4.4674,
      "step": 1205
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00036179999999999996,
      "loss": 4.4731,
      "step": 1206
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00036209999999999997,
      "loss": 4.4934,
      "step": 1207
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00036239999999999997,
      "loss": 4.5261,
      "step": 1208
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0003627,
      "loss": 4.4543,
      "step": 1209
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00036299999999999993,
      "loss": 4.475,
      "step": 1210
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0003633,
      "loss": 4.4577,
      "step": 1211
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00036359999999999995,
      "loss": 4.4647,
      "step": 1212
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0003639,
      "loss": 4.4306,
      "step": 1213
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00036419999999999996,
      "loss": 4.4572,
      "step": 1214
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00036449999999999997,
      "loss": 4.4675,
      "step": 1215
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0003648,
      "loss": 4.4727,
      "step": 1216
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0003651,
      "loss": 4.4152,
      "step": 1217
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00036539999999999994,
      "loss": 4.4185,
      "step": 1218
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0003657,
      "loss": 4.4369,
      "step": 1219
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00036599999999999995,
      "loss": 4.4573,
      "step": 1220
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0003663,
      "loss": 4.4142,
      "step": 1221
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00036659999999999997,
      "loss": 4.3901,
      "step": 1222
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0003669,
      "loss": 4.3966,
      "step": 1223
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0003672,
      "loss": 4.4474,
      "step": 1224
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0003675,
      "loss": 4.5038,
      "step": 1225
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00036779999999999994,
      "loss": 4.4657,
      "step": 1226
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003681,
      "loss": 4.4685,
      "step": 1227
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00036839999999999996,
      "loss": 4.482,
      "step": 1228
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003687,
      "loss": 4.4792,
      "step": 1229
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00036899999999999997,
      "loss": 4.4737,
      "step": 1230
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003693,
      "loss": 4.4856,
      "step": 1231
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003696,
      "loss": 4.3583,
      "step": 1232
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003699,
      "loss": 4.4107,
      "step": 1233
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00037019999999999995,
      "loss": 4.4153,
      "step": 1234
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003705,
      "loss": 4.3843,
      "step": 1235
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00037079999999999996,
      "loss": 4.4708,
      "step": 1236
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003711,
      "loss": 4.469,
      "step": 1237
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003714,
      "loss": 4.4474,
      "step": 1238
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003717,
      "loss": 4.4386,
      "step": 1239
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.000372,
      "loss": 4.3878,
      "step": 1240
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003723,
      "loss": 4.4097,
      "step": 1241
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00037259999999999995,
      "loss": 4.4263,
      "step": 1242
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003729,
      "loss": 4.408,
      "step": 1243
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00037319999999999996,
      "loss": 4.4164,
      "step": 1244
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003735,
      "loss": 4.3984,
      "step": 1245
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0003738,
      "loss": 4.3884,
      "step": 1246
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0003741,
      "loss": 4.3917,
      "step": 1247
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0003744,
      "loss": 4.4581,
      "step": 1248
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0003747,
      "loss": 4.4041,
      "step": 1249
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00037499999999999995,
      "loss": 4.3876,
      "step": 1250
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00037529999999999996,
      "loss": 4.4544,
      "step": 1251
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00037559999999999997,
      "loss": 4.4068,
      "step": 1252
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0003758999999999999,
      "loss": 4.4135,
      "step": 1253
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0003762,
      "loss": 4.4277,
      "step": 1254
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00037649999999999994,
      "loss": 4.4297,
      "step": 1255
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00037679999999999994,
      "loss": 4.4283,
      "step": 1256
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00037709999999999995,
      "loss": 4.4247,
      "step": 1257
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00037739999999999996,
      "loss": 4.4339,
      "step": 1258
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0003776999999999999,
      "loss": 4.474,
      "step": 1259
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00037799999999999997,
      "loss": 4.3817,
      "step": 1260
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0003782999999999999,
      "loss": 4.3437,
      "step": 1261
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0003786,
      "loss": 4.3597,
      "step": 1262
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00037889999999999994,
      "loss": 4.4117,
      "step": 1263
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00037919999999999995,
      "loss": 4.42,
      "step": 1264
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00037949999999999995,
      "loss": 4.4225,
      "step": 1265
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00037979999999999996,
      "loss": 4.4193,
      "step": 1266
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0003800999999999999,
      "loss": 4.459,
      "step": 1267
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0003804,
      "loss": 4.3975,
      "step": 1268
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00038069999999999993,
      "loss": 4.4163,
      "step": 1269
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.000381,
      "loss": 4.4372,
      "step": 1270
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00038129999999999994,
      "loss": 4.4118,
      "step": 1271
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00038159999999999995,
      "loss": 4.4513,
      "step": 1272
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00038189999999999996,
      "loss": 4.4138,
      "step": 1273
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00038219999999999997,
      "loss": 4.3837,
      "step": 1274
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0003824999999999999,
      "loss": 4.4024,
      "step": 1275
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0003828,
      "loss": 4.3359,
      "step": 1276
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00038309999999999993,
      "loss": 4.328,
      "step": 1277
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0003834,
      "loss": 4.3736,
      "step": 1278
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00038369999999999995,
      "loss": 4.4215,
      "step": 1279
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00038399999999999996,
      "loss": 4.3837,
      "step": 1280
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00038429999999999996,
      "loss": 4.3836,
      "step": 1281
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00038459999999999997,
      "loss": 4.3748,
      "step": 1282
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0003848999999999999,
      "loss": 4.4528,
      "step": 1283
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0003852,
      "loss": 4.3899,
      "step": 1284
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00038549999999999994,
      "loss": 4.3914,
      "step": 1285
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0003858,
      "loss": 4.3836,
      "step": 1286
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00038609999999999995,
      "loss": 4.4,
      "step": 1287
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00038639999999999996,
      "loss": 4.3445,
      "step": 1288
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00038669999999999997,
      "loss": 4.409,
      "step": 1289
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.000387,
      "loss": 4.4229,
      "step": 1290
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00038729999999999993,
      "loss": 4.3936,
      "step": 1291
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0003876,
      "loss": 4.3624,
      "step": 1292
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00038789999999999994,
      "loss": 4.3644,
      "step": 1293
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0003882,
      "loss": 4.4258,
      "step": 1294
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00038849999999999996,
      "loss": 4.3803,
      "step": 1295
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00038879999999999996,
      "loss": 4.3614,
      "step": 1296
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00038909999999999997,
      "loss": 4.4007,
      "step": 1297
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0003894,
      "loss": 4.3788,
      "step": 1298
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00038969999999999993,
      "loss": 4.4074,
      "step": 1299
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00039,
      "loss": 4.4305,
      "step": 1300
    },
    {
      "epoch": 0.64,
      "eval_loss": 4.19749116897583,
      "eval_perplexity": 69.2829818725586,
      "eval_runtime": 47.3234,
      "eval_samples_per_second": 2113.12,
      "eval_steps_per_second": 66.035,
      "step": 1300
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00039029999999999995,
      "loss": 4.4318,
      "step": 1301
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00039059999999999995,
      "loss": 4.3893,
      "step": 1302
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00039089999999999996,
      "loss": 4.3879,
      "step": 1303
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00039119999999999997,
      "loss": 4.3795,
      "step": 1304
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0003914999999999999,
      "loss": 4.391,
      "step": 1305
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0003918,
      "loss": 4.3912,
      "step": 1306
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00039209999999999994,
      "loss": 4.4034,
      "step": 1307
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0003924,
      "loss": 4.3754,
      "step": 1308
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00039269999999999995,
      "loss": 4.3448,
      "step": 1309
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00039299999999999996,
      "loss": 4.3619,
      "step": 1310
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00039329999999999996,
      "loss": 4.3737,
      "step": 1311
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00039359999999999997,
      "loss": 4.4204,
      "step": 1312
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0003938999999999999,
      "loss": 4.4113,
      "step": 1313
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0003942,
      "loss": 4.3525,
      "step": 1314
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00039449999999999994,
      "loss": 4.3489,
      "step": 1315
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0003948,
      "loss": 4.3514,
      "step": 1316
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00039509999999999995,
      "loss": 4.3722,
      "step": 1317
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00039539999999999996,
      "loss": 4.3921,
      "step": 1318
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00039569999999999997,
      "loss": 4.3357,
      "step": 1319
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.000396,
      "loss": 4.3295,
      "step": 1320
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00039629999999999993,
      "loss": 4.3717,
      "step": 1321
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0003966,
      "loss": 4.3955,
      "step": 1322
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00039689999999999994,
      "loss": 4.3454,
      "step": 1323
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0003972,
      "loss": 4.3055,
      "step": 1324
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00039749999999999996,
      "loss": 4.3031,
      "step": 1325
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00039779999999999997,
      "loss": 4.3631,
      "step": 1326
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0003981,
      "loss": 4.3061,
      "step": 1327
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0003984,
      "loss": 4.3545,
      "step": 1328
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00039869999999999993,
      "loss": 4.374,
      "step": 1329
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.000399,
      "loss": 4.3521,
      "step": 1330
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00039929999999999995,
      "loss": 4.3433,
      "step": 1331
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0003996,
      "loss": 4.3368,
      "step": 1332
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00039989999999999996,
      "loss": 4.3675,
      "step": 1333
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00040019999999999997,
      "loss": 4.2883,
      "step": 1334
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004005,
      "loss": 4.3882,
      "step": 1335
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004008,
      "loss": 4.3892,
      "step": 1336
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00040109999999999994,
      "loss": 4.3825,
      "step": 1337
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004014,
      "loss": 4.352,
      "step": 1338
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00040169999999999995,
      "loss": 4.4569,
      "step": 1339
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.000402,
      "loss": 4.4267,
      "step": 1340
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00040229999999999997,
      "loss": 4.4132,
      "step": 1341
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004026,
      "loss": 4.3773,
      "step": 1342
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004029,
      "loss": 4.4141,
      "step": 1343
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004032,
      "loss": 4.3646,
      "step": 1344
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00040349999999999994,
      "loss": 4.3288,
      "step": 1345
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004038,
      "loss": 4.3187,
      "step": 1346
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00040409999999999996,
      "loss": 4.3758,
      "step": 1347
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004044,
      "loss": 4.393,
      "step": 1348
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00040469999999999997,
      "loss": 4.3761,
      "step": 1349
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.000405,
      "loss": 4.3576,
      "step": 1350
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00040529999999999993,
      "loss": 4.3425,
      "step": 1351
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004056,
      "loss": 4.3357,
      "step": 1352
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00040589999999999995,
      "loss": 4.3493,
      "step": 1353
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004062,
      "loss": 4.2972,
      "step": 1354
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00040649999999999996,
      "loss": 4.3686,
      "step": 1355
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00040679999999999997,
      "loss": 4.3656,
      "step": 1356
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004071,
      "loss": 4.3331,
      "step": 1357
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004074,
      "loss": 4.3408,
      "step": 1358
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00040769999999999994,
      "loss": 4.35,
      "step": 1359
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.000408,
      "loss": 4.3233,
      "step": 1360
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00040829999999999995,
      "loss": 4.2974,
      "step": 1361
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004086,
      "loss": 4.3523,
      "step": 1362
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00040889999999999996,
      "loss": 4.3312,
      "step": 1363
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00040919999999999997,
      "loss": 4.3195,
      "step": 1364
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004095,
      "loss": 4.3105,
      "step": 1365
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004098,
      "loss": 4.3185,
      "step": 1366
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00041009999999999994,
      "loss": 4.3244,
      "step": 1367
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004104,
      "loss": 4.3001,
      "step": 1368
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00041069999999999995,
      "loss": 4.3108,
      "step": 1369
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.000411,
      "loss": 4.3434,
      "step": 1370
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00041129999999999997,
      "loss": 4.2873,
      "step": 1371
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004116,
      "loss": 4.2662,
      "step": 1372
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004119,
      "loss": 4.2398,
      "step": 1373
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004122,
      "loss": 4.2566,
      "step": 1374
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00041249999999999994,
      "loss": 4.2572,
      "step": 1375
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00041279999999999995,
      "loss": 4.2793,
      "step": 1376
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00041309999999999996,
      "loss": 4.3525,
      "step": 1377
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004133999999999999,
      "loss": 4.366,
      "step": 1378
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00041369999999999997,
      "loss": 4.3135,
      "step": 1379
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004139999999999999,
      "loss": 4.2663,
      "step": 1380
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004143,
      "loss": 4.3015,
      "step": 1381
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00041459999999999994,
      "loss": 4.3394,
      "step": 1382
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00041489999999999995,
      "loss": 4.3388,
      "step": 1383
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00041519999999999995,
      "loss": 4.3474,
      "step": 1384
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00041549999999999996,
      "loss": 4.4316,
      "step": 1385
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004157999999999999,
      "loss": 4.3695,
      "step": 1386
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004161,
      "loss": 4.3367,
      "step": 1387
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00041639999999999993,
      "loss": 4.3627,
      "step": 1388
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004167,
      "loss": 4.407,
      "step": 1389
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00041699999999999994,
      "loss": 4.3075,
      "step": 1390
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00041729999999999995,
      "loss": 4.2913,
      "step": 1391
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00041759999999999996,
      "loss": 4.2836,
      "step": 1392
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00041789999999999997,
      "loss": 4.2262,
      "step": 1393
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004181999999999999,
      "loss": 4.2695,
      "step": 1394
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004185,
      "loss": 4.2722,
      "step": 1395
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00041879999999999993,
      "loss": 4.2611,
      "step": 1396
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004191,
      "loss": 4.2869,
      "step": 1397
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00041939999999999995,
      "loss": 4.3536,
      "step": 1398
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00041969999999999996,
      "loss": 4.2748,
      "step": 1399
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00041999999999999996,
      "loss": 4.269,
      "step": 1400
    },
    {
      "epoch": 0.69,
      "eval_loss": 4.127819061279297,
      "eval_perplexity": 64.63782501220703,
      "eval_runtime": 47.5238,
      "eval_samples_per_second": 2104.207,
      "eval_steps_per_second": 65.756,
      "step": 1400
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00042029999999999997,
      "loss": 4.3463,
      "step": 1401
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004205999999999999,
      "loss": 4.3116,
      "step": 1402
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004209,
      "loss": 4.3103,
      "step": 1403
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00042119999999999994,
      "loss": 4.3147,
      "step": 1404
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00042149999999999995,
      "loss": 4.2984,
      "step": 1405
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00042179999999999995,
      "loss": 4.316,
      "step": 1406
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00042209999999999996,
      "loss": 4.3241,
      "step": 1407
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004223999999999999,
      "loss": 4.3335,
      "step": 1408
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004227,
      "loss": 4.3387,
      "step": 1409
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00042299999999999993,
      "loss": 4.3371,
      "step": 1410
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004233,
      "loss": 4.291,
      "step": 1411
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00042359999999999994,
      "loss": 4.2576,
      "step": 1412
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00042389999999999995,
      "loss": 4.289,
      "step": 1413
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00042419999999999996,
      "loss": 4.2612,
      "step": 1414
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00042449999999999996,
      "loss": 4.2477,
      "step": 1415
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004247999999999999,
      "loss": 4.2565,
      "step": 1416
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004251,
      "loss": 4.2765,
      "step": 1417
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00042539999999999993,
      "loss": 4.3071,
      "step": 1418
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004257,
      "loss": 4.2764,
      "step": 1419
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00042599999999999995,
      "loss": 4.2813,
      "step": 1420
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00042629999999999995,
      "loss": 4.3219,
      "step": 1421
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00042659999999999996,
      "loss": 4.3004,
      "step": 1422
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00042689999999999997,
      "loss": 4.2958,
      "step": 1423
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004271999999999999,
      "loss": 4.2902,
      "step": 1424
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004275,
      "loss": 4.2833,
      "step": 1425
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00042779999999999994,
      "loss": 4.1878,
      "step": 1426
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004281,
      "loss": 4.2085,
      "step": 1427
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00042839999999999995,
      "loss": 4.2454,
      "step": 1428
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00042869999999999996,
      "loss": 4.2449,
      "step": 1429
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00042899999999999997,
      "loss": 4.2871,
      "step": 1430
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00042929999999999997,
      "loss": 4.2962,
      "step": 1431
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004295999999999999,
      "loss": 4.3264,
      "step": 1432
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004299,
      "loss": 4.2488,
      "step": 1433
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00043019999999999994,
      "loss": 4.2669,
      "step": 1434
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004305,
      "loss": 4.2653,
      "step": 1435
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00043079999999999995,
      "loss": 4.2534,
      "step": 1436
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00043109999999999996,
      "loss": 4.2853,
      "step": 1437
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00043139999999999997,
      "loss": 4.2546,
      "step": 1438
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004317,
      "loss": 4.216,
      "step": 1439
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00043199999999999993,
      "loss": 4.211,
      "step": 1440
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004323,
      "loss": 4.2115,
      "step": 1441
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00043259999999999994,
      "loss": 4.2311,
      "step": 1442
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004329,
      "loss": 4.2456,
      "step": 1443
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00043319999999999996,
      "loss": 4.2824,
      "step": 1444
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00043349999999999997,
      "loss": 4.232,
      "step": 1445
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004338,
      "loss": 4.2264,
      "step": 1446
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004341,
      "loss": 4.2656,
      "step": 1447
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00043439999999999993,
      "loss": 4.2896,
      "step": 1448
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004347,
      "loss": 4.3192,
      "step": 1449
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00043499999999999995,
      "loss": 4.2752,
      "step": 1450
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00043529999999999996,
      "loss": 4.2498,
      "step": 1451
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00043559999999999996,
      "loss": 4.2777,
      "step": 1452
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00043589999999999997,
      "loss": 4.2628,
      "step": 1453
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004361999999999999,
      "loss": 4.2303,
      "step": 1454
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004365,
      "loss": 4.2323,
      "step": 1455
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00043679999999999994,
      "loss": 4.2838,
      "step": 1456
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004371,
      "loss": 4.2471,
      "step": 1457
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00043739999999999995,
      "loss": 4.2344,
      "step": 1458
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00043769999999999996,
      "loss": 4.2465,
      "step": 1459
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00043799999999999997,
      "loss": 4.282,
      "step": 1460
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004383,
      "loss": 4.2863,
      "step": 1461
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00043859999999999993,
      "loss": 4.2765,
      "step": 1462
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004389,
      "loss": 4.2542,
      "step": 1463
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00043919999999999994,
      "loss": 4.2624,
      "step": 1464
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004395,
      "loss": 4.285,
      "step": 1465
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00043979999999999996,
      "loss": 4.2984,
      "step": 1466
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00044009999999999996,
      "loss": 4.2924,
      "step": 1467
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00044039999999999997,
      "loss": 4.2858,
      "step": 1468
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004407,
      "loss": 4.2836,
      "step": 1469
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00044099999999999993,
      "loss": 4.2721,
      "step": 1470
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004413,
      "loss": 4.2814,
      "step": 1471
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00044159999999999995,
      "loss": 4.3112,
      "step": 1472
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004419,
      "loss": 4.2799,
      "step": 1473
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00044219999999999996,
      "loss": 4.233,
      "step": 1474
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00044249999999999997,
      "loss": 4.2011,
      "step": 1475
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004428,
      "loss": 4.2513,
      "step": 1476
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004431,
      "loss": 4.2504,
      "step": 1477
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00044339999999999994,
      "loss": 4.237,
      "step": 1478
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004437,
      "loss": 4.2364,
      "step": 1479
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00044399999999999995,
      "loss": 4.2309,
      "step": 1480
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004443,
      "loss": 4.2785,
      "step": 1481
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00044459999999999996,
      "loss": 4.2236,
      "step": 1482
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00044489999999999997,
      "loss": 4.198,
      "step": 1483
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004452,
      "loss": 4.23,
      "step": 1484
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004455,
      "loss": 4.2366,
      "step": 1485
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00044579999999999994,
      "loss": 4.228,
      "step": 1486
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004461,
      "loss": 4.2359,
      "step": 1487
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00044639999999999995,
      "loss": 4.2234,
      "step": 1488
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004467,
      "loss": 4.2333,
      "step": 1489
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00044699999999999997,
      "loss": 4.2397,
      "step": 1490
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004473,
      "loss": 4.2447,
      "step": 1491
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004476,
      "loss": 4.2158,
      "step": 1492
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004479,
      "loss": 4.2602,
      "step": 1493
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00044819999999999994,
      "loss": 4.2082,
      "step": 1494
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004485,
      "loss": 4.19,
      "step": 1495
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00044879999999999996,
      "loss": 4.2625,
      "step": 1496
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004491,
      "loss": 4.2156,
      "step": 1497
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004494,
      "loss": 4.23,
      "step": 1498
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004497,
      "loss": 4.2609,
      "step": 1499
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00045,
      "loss": 4.2478,
      "step": 1500
    },
    {
      "epoch": 0.74,
      "eval_loss": 4.059741020202637,
      "eval_perplexity": 60.37553405761719,
      "eval_runtime": 192.4188,
      "eval_samples_per_second": 519.7,
      "eval_steps_per_second": 16.241,
      "step": 1500
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00045029999999999994,
      "loss": 4.2496,
      "step": 1501
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00045059999999999995,
      "loss": 4.2064,
      "step": 1502
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004508999999999999,
      "loss": 4.1808,
      "step": 1503
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00045119999999999996,
      "loss": 4.23,
      "step": 1504
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004514999999999999,
      "loss": 4.2406,
      "step": 1505
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004518,
      "loss": 4.2504,
      "step": 1506
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00045209999999999993,
      "loss": 4.2273,
      "step": 1507
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00045239999999999994,
      "loss": 4.2055,
      "step": 1508
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00045269999999999994,
      "loss": 4.1547,
      "step": 1509
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00045299999999999995,
      "loss": 4.1447,
      "step": 1510
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004532999999999999,
      "loss": 4.1647,
      "step": 1511
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00045359999999999997,
      "loss": 4.2895,
      "step": 1512
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004538999999999999,
      "loss": 4.2186,
      "step": 1513
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004542,
      "loss": 4.1997,
      "step": 1514
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00045449999999999993,
      "loss": 4.2314,
      "step": 1515
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00045479999999999994,
      "loss": 4.2325,
      "step": 1516
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00045509999999999995,
      "loss": 4.2066,
      "step": 1517
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00045539999999999996,
      "loss": 4.1674,
      "step": 1518
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004556999999999999,
      "loss": 4.1443,
      "step": 1519
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00045599999999999997,
      "loss": 4.2458,
      "step": 1520
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004562999999999999,
      "loss": 4.2233,
      "step": 1521
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004566,
      "loss": 4.2338,
      "step": 1522
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00045689999999999994,
      "loss": 4.2527,
      "step": 1523
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00045719999999999995,
      "loss": 4.2469,
      "step": 1524
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00045749999999999995,
      "loss": 4.2502,
      "step": 1525
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00045779999999999996,
      "loss": 4.2289,
      "step": 1526
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004580999999999999,
      "loss": 4.1875,
      "step": 1527
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004584,
      "loss": 4.2387,
      "step": 1528
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00045869999999999993,
      "loss": 4.196,
      "step": 1529
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.000459,
      "loss": 4.188,
      "step": 1530
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00045929999999999994,
      "loss": 4.2097,
      "step": 1531
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00045959999999999995,
      "loss": 4.2169,
      "step": 1532
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00045989999999999996,
      "loss": 4.1897,
      "step": 1533
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00046019999999999996,
      "loss": 4.1648,
      "step": 1534
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004604999999999999,
      "loss": 4.1818,
      "step": 1535
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004608,
      "loss": 4.1785,
      "step": 1536
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00046109999999999993,
      "loss": 4.1744,
      "step": 1537
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004614,
      "loss": 4.1694,
      "step": 1538
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00046169999999999995,
      "loss": 4.1242,
      "step": 1539
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00046199999999999995,
      "loss": 4.1495,
      "step": 1540
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00046229999999999996,
      "loss": 4.1911,
      "step": 1541
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00046259999999999997,
      "loss": 4.1961,
      "step": 1542
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004628999999999999,
      "loss": 4.21,
      "step": 1543
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004632,
      "loss": 4.2411,
      "step": 1544
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00046349999999999994,
      "loss": 4.2263,
      "step": 1545
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004638,
      "loss": 4.2381,
      "step": 1546
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00046409999999999995,
      "loss": 4.2379,
      "step": 1547
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00046439999999999996,
      "loss": 4.1615,
      "step": 1548
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00046469999999999997,
      "loss": 4.2175,
      "step": 1549
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00046499999999999997,
      "loss": 4.2345,
      "step": 1550
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004652999999999999,
      "loss": 4.177,
      "step": 1551
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004656,
      "loss": 4.184,
      "step": 1552
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00046589999999999994,
      "loss": 4.1467,
      "step": 1553
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00046619999999999995,
      "loss": 4.1463,
      "step": 1554
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00046649999999999996,
      "loss": 4.1875,
      "step": 1555
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00046679999999999996,
      "loss": 4.1324,
      "step": 1556
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004670999999999999,
      "loss": 4.1612,
      "step": 1557
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004674,
      "loss": 4.1324,
      "step": 1558
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00046769999999999993,
      "loss": 4.1174,
      "step": 1559
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.000468,
      "loss": 4.2276,
      "step": 1560
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00046829999999999994,
      "loss": 4.2369,
      "step": 1561
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00046859999999999995,
      "loss": 4.1917,
      "step": 1562
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00046889999999999996,
      "loss": 4.1791,
      "step": 1563
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00046919999999999997,
      "loss": 4.1654,
      "step": 1564
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004694999999999999,
      "loss": 4.1479,
      "step": 1565
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004698,
      "loss": 4.1345,
      "step": 1566
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00047009999999999993,
      "loss": 4.1796,
      "step": 1567
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004704,
      "loss": 4.1389,
      "step": 1568
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00047069999999999995,
      "loss": 4.1381,
      "step": 1569
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00047099999999999996,
      "loss": 4.1623,
      "step": 1570
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00047129999999999996,
      "loss": 4.1374,
      "step": 1571
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00047159999999999997,
      "loss": 4.1221,
      "step": 1572
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004718999999999999,
      "loss": 4.1624,
      "step": 1573
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004722,
      "loss": 4.1843,
      "step": 1574
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00047249999999999994,
      "loss": 4.0856,
      "step": 1575
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004728,
      "loss": 4.2118,
      "step": 1576
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00047309999999999995,
      "loss": 4.2305,
      "step": 1577
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00047339999999999996,
      "loss": 4.2006,
      "step": 1578
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00047369999999999997,
      "loss": 4.1392,
      "step": 1579
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.000474,
      "loss": 4.1481,
      "step": 1580
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00047429999999999993,
      "loss": 4.1634,
      "step": 1581
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0004746,
      "loss": 4.164,
      "step": 1582
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00047489999999999994,
      "loss": 4.1788,
      "step": 1583
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0004752,
      "loss": 4.1968,
      "step": 1584
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00047549999999999996,
      "loss": 4.1835,
      "step": 1585
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00047579999999999996,
      "loss": 4.1788,
      "step": 1586
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00047609999999999997,
      "loss": 4.2404,
      "step": 1587
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0004764,
      "loss": 4.1942,
      "step": 1588
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00047669999999999993,
      "loss": 4.1555,
      "step": 1589
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.000477,
      "loss": 4.1332,
      "step": 1590
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00047729999999999995,
      "loss": 4.1729,
      "step": 1591
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0004776,
      "loss": 4.1725,
      "step": 1592
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00047789999999999996,
      "loss": 4.1628,
      "step": 1593
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00047819999999999997,
      "loss": 4.1641,
      "step": 1594
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0004785,
      "loss": 4.1836,
      "step": 1595
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0004788,
      "loss": 4.2481,
      "step": 1596
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00047909999999999994,
      "loss": 4.1991,
      "step": 1597
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0004794,
      "loss": 4.1539,
      "step": 1598
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00047969999999999995,
      "loss": 4.217,
      "step": 1599
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00047999999999999996,
      "loss": 4.2349,
      "step": 1600
    },
    {
      "epoch": 1.01,
      "eval_loss": 4.016582012176514,
      "eval_perplexity": 57.84401321411133,
      "eval_runtime": 47.3892,
      "eval_samples_per_second": 2110.187,
      "eval_steps_per_second": 65.943,
      "step": 1600
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00048029999999999997,
      "loss": 4.1846,
      "step": 1601
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00048059999999999997,
      "loss": 4.1605,
      "step": 1602
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0004808999999999999,
      "loss": 4.1539,
      "step": 1603
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0004812,
      "loss": 4.1672,
      "step": 1604
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00048149999999999994,
      "loss": 4.1658,
      "step": 1605
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0004818,
      "loss": 4.1666,
      "step": 1606
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00048209999999999995,
      "loss": 4.1711,
      "step": 1607
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00048239999999999996,
      "loss": 4.1503,
      "step": 1608
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00048269999999999997,
      "loss": 4.1411,
      "step": 1609
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.000483,
      "loss": 4.1677,
      "step": 1610
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00048329999999999993,
      "loss": 4.1644,
      "step": 1611
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0004836,
      "loss": 4.1702,
      "step": 1612
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00048389999999999994,
      "loss": 4.1356,
      "step": 1613
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0004842,
      "loss": 4.1263,
      "step": 1614
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00048449999999999996,
      "loss": 4.0946,
      "step": 1615
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00048479999999999997,
      "loss": 4.1319,
      "step": 1616
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0004851,
      "loss": 4.1645,
      "step": 1617
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0004854,
      "loss": 4.1869,
      "step": 1618
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00048569999999999993,
      "loss": 4.1873,
      "step": 1619
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.000486,
      "loss": 4.145,
      "step": 1620
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00048629999999999995,
      "loss": 4.1321,
      "step": 1621
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0004866,
      "loss": 4.1557,
      "step": 1622
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00048689999999999996,
      "loss": 4.214,
      "step": 1623
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00048719999999999997,
      "loss": 4.1804,
      "step": 1624
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0004875,
      "loss": 4.1447,
      "step": 1625
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00048779999999999993,
      "loss": 4.1524,
      "step": 1626
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00048809999999999994,
      "loss": 4.1978,
      "step": 1627
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0004883999999999999,
      "loss": 4.2138,
      "step": 1628
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0004887,
      "loss": 4.1659,
      "step": 1629
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.000489,
      "loss": 4.1183,
      "step": 1630
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0004892999999999999,
      "loss": 4.0899,
      "step": 1631
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0004896,
      "loss": 4.1175,
      "step": 1632
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004898999999999999,
      "loss": 4.1305,
      "step": 1633
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004901999999999999,
      "loss": 4.1584,
      "step": 1634
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004904999999999999,
      "loss": 4.111,
      "step": 1635
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004907999999999999,
      "loss": 4.1134,
      "step": 1636
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004911,
      "loss": 4.1254,
      "step": 1637
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004913999999999999,
      "loss": 4.087,
      "step": 1638
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004917,
      "loss": 4.106,
      "step": 1639
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004919999999999999,
      "loss": 4.1535,
      "step": 1640
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004923,
      "loss": 4.1558,
      "step": 1641
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004925999999999999,
      "loss": 4.1393,
      "step": 1642
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004929,
      "loss": 4.1519,
      "step": 1643
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004932,
      "loss": 4.1432,
      "step": 1644
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004935,
      "loss": 4.1536,
      "step": 1645
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004938,
      "loss": 4.1741,
      "step": 1646
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004940999999999999,
      "loss": 4.1448,
      "step": 1647
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004944,
      "loss": 4.181,
      "step": 1648
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004946999999999999,
      "loss": 4.1441,
      "step": 1649
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004949999999999999,
      "loss": 4.1232,
      "step": 1650
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004953,
      "loss": 4.1713,
      "step": 1651
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0004955999999999999,
      "loss": 4.1387,
      "step": 1652
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0004959,
      "loss": 4.1038,
      "step": 1653
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0004961999999999999,
      "loss": 4.1167,
      "step": 1654
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0004965,
      "loss": 4.1081,
      "step": 1655
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0004967999999999999,
      "loss": 4.1105,
      "step": 1656
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0004971,
      "loss": 4.1325,
      "step": 1657
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0004974,
      "loss": 4.1518,
      "step": 1658
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0004977,
      "loss": 4.1164,
      "step": 1659
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.000498,
      "loss": 4.1201,
      "step": 1660
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0004982999999999999,
      "loss": 4.1248,
      "step": 1661
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0004986,
      "loss": 4.1087,
      "step": 1662
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0004988999999999999,
      "loss": 4.1604,
      "step": 1663
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0004991999999999999,
      "loss": 4.1061,
      "step": 1664
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0004994999999999999,
      "loss": 4.1308,
      "step": 1665
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0004997999999999999,
      "loss": 4.1474,
      "step": 1666
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0005001,
      "loss": 4.133,
      "step": 1667
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0005003999999999999,
      "loss": 4.1212,
      "step": 1668
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0005007,
      "loss": 4.1132,
      "step": 1669
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0005009999999999999,
      "loss": 4.1191,
      "step": 1670
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0005013,
      "loss": 4.1681,
      "step": 1671
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0005015999999999999,
      "loss": 4.1444,
      "step": 1672
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005019,
      "loss": 4.1213,
      "step": 1673
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005022,
      "loss": 4.121,
      "step": 1674
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005025,
      "loss": 4.1413,
      "step": 1675
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005028,
      "loss": 4.1375,
      "step": 1676
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005030999999999999,
      "loss": 4.1076,
      "step": 1677
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005034,
      "loss": 4.0719,
      "step": 1678
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005036999999999999,
      "loss": 4.1288,
      "step": 1679
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005039999999999999,
      "loss": 4.1023,
      "step": 1680
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005043,
      "loss": 4.1073,
      "step": 1681
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005045999999999999,
      "loss": 4.1264,
      "step": 1682
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005049,
      "loss": 4.072,
      "step": 1683
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005051999999999999,
      "loss": 4.0908,
      "step": 1684
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005055,
      "loss": 4.0811,
      "step": 1685
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005057999999999999,
      "loss": 4.0719,
      "step": 1686
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005061,
      "loss": 4.087,
      "step": 1687
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005064,
      "loss": 4.096,
      "step": 1688
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005067,
      "loss": 4.083,
      "step": 1689
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.000507,
      "loss": 4.0816,
      "step": 1690
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005073,
      "loss": 4.0978,
      "step": 1691
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005076,
      "loss": 4.0643,
      "step": 1692
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005078999999999999,
      "loss": 4.0676,
      "step": 1693
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005082,
      "loss": 4.0901,
      "step": 1694
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005085,
      "loss": 4.11,
      "step": 1695
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005087999999999999,
      "loss": 4.1256,
      "step": 1696
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005091,
      "loss": 4.1302,
      "step": 1697
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005093999999999999,
      "loss": 4.1601,
      "step": 1698
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005097,
      "loss": 4.1169,
      "step": 1699
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005099999999999999,
      "loss": 4.149,
      "step": 1700
    },
    {
      "epoch": 1.06,
      "eval_loss": 3.953185558319092,
      "eval_perplexity": 54.27214813232422,
      "eval_runtime": 212.2517,
      "eval_samples_per_second": 471.139,
      "eval_steps_per_second": 14.723,
      "step": 1700
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005103,
      "loss": 4.1239,
      "step": 1701
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005105999999999999,
      "loss": 4.0859,
      "step": 1702
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005109,
      "loss": 4.1329,
      "step": 1703
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005112,
      "loss": 4.0936,
      "step": 1704
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005115,
      "loss": 4.0897,
      "step": 1705
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005118,
      "loss": 4.1053,
      "step": 1706
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005120999999999999,
      "loss": 4.0769,
      "step": 1707
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005124,
      "loss": 4.0893,
      "step": 1708
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005126999999999999,
      "loss": 4.1031,
      "step": 1709
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005129999999999999,
      "loss": 4.0628,
      "step": 1710
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005133,
      "loss": 4.1069,
      "step": 1711
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005135999999999999,
      "loss": 4.1122,
      "step": 1712
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0005139,
      "loss": 4.0747,
      "step": 1713
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005141999999999999,
      "loss": 4.0583,
      "step": 1714
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005145,
      "loss": 3.9964,
      "step": 1715
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005147999999999999,
      "loss": 4.0403,
      "step": 1716
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005151,
      "loss": 4.0748,
      "step": 1717
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005154,
      "loss": 4.094,
      "step": 1718
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005157,
      "loss": 4.0859,
      "step": 1719
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.000516,
      "loss": 4.1243,
      "step": 1720
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005163,
      "loss": 4.109,
      "step": 1721
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005166,
      "loss": 4.0389,
      "step": 1722
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005168999999999999,
      "loss": 4.0741,
      "step": 1723
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005172,
      "loss": 4.0563,
      "step": 1724
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005175,
      "loss": 4.0604,
      "step": 1725
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005177999999999999,
      "loss": 4.0593,
      "step": 1726
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005181,
      "loss": 4.1176,
      "step": 1727
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005183999999999999,
      "loss": 4.0916,
      "step": 1728
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005187,
      "loss": 4.061,
      "step": 1729
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005189999999999999,
      "loss": 4.0624,
      "step": 1730
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005193,
      "loss": 3.9896,
      "step": 1731
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005195999999999999,
      "loss": 4.0311,
      "step": 1732
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0005199,
      "loss": 4.0439,
      "step": 1733
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005202,
      "loss": 4.0345,
      "step": 1734
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005205,
      "loss": 4.0934,
      "step": 1735
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005208,
      "loss": 4.0757,
      "step": 1736
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005211,
      "loss": 4.0626,
      "step": 1737
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005214,
      "loss": 4.0787,
      "step": 1738
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005216999999999999,
      "loss": 4.1113,
      "step": 1739
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.000522,
      "loss": 4.0523,
      "step": 1740
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005223,
      "loss": 4.0353,
      "step": 1741
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005225999999999999,
      "loss": 4.1152,
      "step": 1742
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005229,
      "loss": 4.0616,
      "step": 1743
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005231999999999999,
      "loss": 4.1035,
      "step": 1744
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005235,
      "loss": 4.1447,
      "step": 1745
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005237999999999999,
      "loss": 4.0961,
      "step": 1746
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005241,
      "loss": 4.0651,
      "step": 1747
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005244,
      "loss": 4.0655,
      "step": 1748
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005247,
      "loss": 4.042,
      "step": 1749
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.000525,
      "loss": 4.0667,
      "step": 1750
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005252999999999999,
      "loss": 4.1241,
      "step": 1751
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005256,
      "loss": 4.0558,
      "step": 1752
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0005258999999999999,
      "loss": 4.0555,
      "step": 1753
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005262,
      "loss": 4.0447,
      "step": 1754
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005265,
      "loss": 4.0933,
      "step": 1755
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005267999999999999,
      "loss": 4.0629,
      "step": 1756
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005271,
      "loss": 4.0357,
      "step": 1757
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005273999999999999,
      "loss": 3.9931,
      "step": 1758
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005276999999999999,
      "loss": 4.1122,
      "step": 1759
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005279999999999999,
      "loss": 4.1082,
      "step": 1760
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005282999999999999,
      "loss": 4.0722,
      "step": 1761
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005286,
      "loss": 4.0686,
      "step": 1762
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005288999999999999,
      "loss": 4.0485,
      "step": 1763
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005292,
      "loss": 4.0805,
      "step": 1764
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005294999999999999,
      "loss": 4.1194,
      "step": 1765
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005298,
      "loss": 4.0556,
      "step": 1766
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005300999999999999,
      "loss": 4.078,
      "step": 1767
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005304,
      "loss": 4.0612,
      "step": 1768
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005306999999999999,
      "loss": 4.0465,
      "step": 1769
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.000531,
      "loss": 4.0688,
      "step": 1770
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005313,
      "loss": 4.0753,
      "step": 1771
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005315999999999999,
      "loss": 4.0786,
      "step": 1772
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005319,
      "loss": 4.037,
      "step": 1773
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005321999999999999,
      "loss": 4.0109,
      "step": 1774
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005324999999999999,
      "loss": 4.0237,
      "step": 1775
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005327999999999999,
      "loss": 4.0473,
      "step": 1776
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005330999999999999,
      "loss": 4.0546,
      "step": 1777
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005334,
      "loss": 4.0889,
      "step": 1778
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005336999999999999,
      "loss": 4.0833,
      "step": 1779
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.000534,
      "loss": 4.0634,
      "step": 1780
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005342999999999999,
      "loss": 4.0374,
      "step": 1781
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005346,
      "loss": 4.0549,
      "step": 1782
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005348999999999999,
      "loss": 4.0296,
      "step": 1783
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005352,
      "loss": 4.0487,
      "step": 1784
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005355,
      "loss": 4.0686,
      "step": 1785
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005358,
      "loss": 4.0429,
      "step": 1786
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005361,
      "loss": 4.0074,
      "step": 1787
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005363999999999999,
      "loss": 3.9916,
      "step": 1788
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005367,
      "loss": 4.0457,
      "step": 1789
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005369999999999999,
      "loss": 4.0612,
      "step": 1790
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005372999999999999,
      "loss": 4.0245,
      "step": 1791
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005376,
      "loss": 4.0093,
      "step": 1792
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005378999999999999,
      "loss": 3.9888,
      "step": 1793
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0005382,
      "loss": 4.0003,
      "step": 1794
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005384999999999999,
      "loss": 4.0487,
      "step": 1795
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005388,
      "loss": 4.0915,
      "step": 1796
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005390999999999999,
      "loss": 4.0433,
      "step": 1797
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005394,
      "loss": 4.0353,
      "step": 1798
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005396999999999999,
      "loss": 4.0241,
      "step": 1799
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00054,
      "loss": 4.0502,
      "step": 1800
    },
    {
      "epoch": 1.11,
      "eval_loss": 3.9064841270446777,
      "eval_perplexity": 51.802711486816406,
      "eval_runtime": 47.2428,
      "eval_samples_per_second": 2116.723,
      "eval_steps_per_second": 66.148,
      "step": 1800
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005403,
      "loss": 4.0851,
      "step": 1801
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005405999999999999,
      "loss": 4.0724,
      "step": 1802
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005409,
      "loss": 4.0191,
      "step": 1803
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005411999999999999,
      "loss": 4.0103,
      "step": 1804
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005414999999999999,
      "loss": 4.0705,
      "step": 1805
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005417999999999999,
      "loss": 4.0584,
      "step": 1806
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005420999999999999,
      "loss": 4.0455,
      "step": 1807
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005424,
      "loss": 4.0395,
      "step": 1808
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005426999999999999,
      "loss": 3.985,
      "step": 1809
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.000543,
      "loss": 4.005,
      "step": 1810
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005432999999999999,
      "loss": 3.9954,
      "step": 1811
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005436,
      "loss": 4.003,
      "step": 1812
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005438999999999999,
      "loss": 4.0335,
      "step": 1813
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0005442,
      "loss": 3.9954,
      "step": 1814
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005445,
      "loss": 3.9847,
      "step": 1815
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005448,
      "loss": 3.9932,
      "step": 1816
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005451,
      "loss": 4.0449,
      "step": 1817
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005453999999999999,
      "loss": 4.0288,
      "step": 1818
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005457,
      "loss": 4.0496,
      "step": 1819
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005459999999999999,
      "loss": 4.0496,
      "step": 1820
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005462999999999999,
      "loss": 4.0436,
      "step": 1821
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005466,
      "loss": 4.0036,
      "step": 1822
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005468999999999999,
      "loss": 4.0052,
      "step": 1823
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005472,
      "loss": 4.0004,
      "step": 1824
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005474999999999999,
      "loss": 3.9788,
      "step": 1825
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005478,
      "loss": 4.0164,
      "step": 1826
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005480999999999999,
      "loss": 4.0315,
      "step": 1827
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005484,
      "loss": 4.0499,
      "step": 1828
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005487,
      "loss": 4.0708,
      "step": 1829
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.000549,
      "loss": 4.0634,
      "step": 1830
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005493,
      "loss": 4.0527,
      "step": 1831
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005496,
      "loss": 4.088,
      "step": 1832
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005499,
      "loss": 4.0516,
      "step": 1833
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005501999999999999,
      "loss": 4.0492,
      "step": 1834
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0005505,
      "loss": 4.0484,
      "step": 1835
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005507999999999999,
      "loss": 4.0488,
      "step": 1836
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005510999999999999,
      "loss": 4.0451,
      "step": 1837
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005514,
      "loss": 4.0114,
      "step": 1838
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005516999999999999,
      "loss": 3.9997,
      "step": 1839
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.000552,
      "loss": 4.028,
      "step": 1840
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005522999999999999,
      "loss": 4.0499,
      "step": 1841
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005526,
      "loss": 3.9892,
      "step": 1842
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005528999999999999,
      "loss": 3.983,
      "step": 1843
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005532,
      "loss": 4.015,
      "step": 1844
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005535,
      "loss": 3.9778,
      "step": 1845
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005538,
      "loss": 4.049,
      "step": 1846
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005541,
      "loss": 4.0412,
      "step": 1847
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005544,
      "loss": 4.0207,
      "step": 1848
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005547,
      "loss": 3.9863,
      "step": 1849
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005549999999999999,
      "loss": 3.9847,
      "step": 1850
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005552999999999999,
      "loss": 4.0218,
      "step": 1851
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005556,
      "loss": 4.0668,
      "step": 1852
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005558999999999999,
      "loss": 4.0213,
      "step": 1853
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005562,
      "loss": 4.0252,
      "step": 1854
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0005564999999999999,
      "loss": 3.9635,
      "step": 1855
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005568,
      "loss": 3.9438,
      "step": 1856
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005570999999999999,
      "loss": 4.0265,
      "step": 1857
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005574,
      "loss": 4.0345,
      "step": 1858
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005577,
      "loss": 4.0391,
      "step": 1859
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.000558,
      "loss": 4.0274,
      "step": 1860
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005583,
      "loss": 4.0266,
      "step": 1861
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005586,
      "loss": 4.0132,
      "step": 1862
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005589,
      "loss": 3.9906,
      "step": 1863
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005591999999999999,
      "loss": 3.9424,
      "step": 1864
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005595,
      "loss": 4.067,
      "step": 1865
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005598,
      "loss": 4.0064,
      "step": 1866
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005600999999999999,
      "loss": 3.9963,
      "step": 1867
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005604,
      "loss": 4.0314,
      "step": 1868
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005606999999999999,
      "loss": 3.9558,
      "step": 1869
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.000561,
      "loss": 4.0184,
      "step": 1870
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005612999999999999,
      "loss": 4.0283,
      "step": 1871
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005616,
      "loss": 3.9914,
      "step": 1872
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005618999999999999,
      "loss": 4.0078,
      "step": 1873
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005622,
      "loss": 3.9864,
      "step": 1874
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005625,
      "loss": 4.01,
      "step": 1875
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005627999999999999,
      "loss": 4.0438,
      "step": 1876
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005631,
      "loss": 4.0354,
      "step": 1877
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005633999999999999,
      "loss": 4.0449,
      "step": 1878
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005637,
      "loss": 4.0009,
      "step": 1879
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005639999999999999,
      "loss": 3.9997,
      "step": 1880
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005643,
      "loss": 4.0594,
      "step": 1881
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005646,
      "loss": 4.011,
      "step": 1882
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005648999999999999,
      "loss": 3.9833,
      "step": 1883
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005652,
      "loss": 4.0203,
      "step": 1884
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005654999999999999,
      "loss": 4.0562,
      "step": 1885
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005657999999999999,
      "loss": 4.0371,
      "step": 1886
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005660999999999999,
      "loss": 4.0104,
      "step": 1887
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005663999999999999,
      "loss": 4.0365,
      "step": 1888
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005667,
      "loss": 3.9781,
      "step": 1889
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005669999999999999,
      "loss": 4.0389,
      "step": 1890
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005673,
      "loss": 4.0324,
      "step": 1891
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005675999999999999,
      "loss": 3.988,
      "step": 1892
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005679,
      "loss": 3.9798,
      "step": 1893
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005681999999999999,
      "loss": 3.9834,
      "step": 1894
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0005685,
      "loss": 3.9745,
      "step": 1895
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005688,
      "loss": 3.9758,
      "step": 1896
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005691,
      "loss": 4.014,
      "step": 1897
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005694,
      "loss": 3.998,
      "step": 1898
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005696999999999999,
      "loss": 4.0131,
      "step": 1899
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00057,
      "loss": 4.0128,
      "step": 1900
    },
    {
      "epoch": 1.16,
      "eval_loss": 3.874253749847412,
      "eval_perplexity": 50.187015533447266,
      "eval_runtime": 47.6157,
      "eval_samples_per_second": 2100.149,
      "eval_steps_per_second": 65.63,
      "step": 1900
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005702999999999999,
      "loss": 4.032,
      "step": 1901
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005705999999999999,
      "loss": 3.9977,
      "step": 1902
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005708999999999999,
      "loss": 3.9852,
      "step": 1903
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005711999999999999,
      "loss": 4.0102,
      "step": 1904
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005715,
      "loss": 3.9744,
      "step": 1905
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005717999999999999,
      "loss": 3.994,
      "step": 1906
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005721,
      "loss": 3.9766,
      "step": 1907
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005723999999999999,
      "loss": 4.0052,
      "step": 1908
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005727,
      "loss": 4.0284,
      "step": 1909
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005729999999999999,
      "loss": 4.002,
      "step": 1910
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005733,
      "loss": 3.9853,
      "step": 1911
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005736,
      "loss": 3.9861,
      "step": 1912
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005738999999999999,
      "loss": 3.9658,
      "step": 1913
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005742,
      "loss": 3.945,
      "step": 1914
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005744999999999999,
      "loss": 3.9531,
      "step": 1915
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005747999999999999,
      "loss": 4.0037,
      "step": 1916
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005750999999999999,
      "loss": 3.961,
      "step": 1917
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005753999999999999,
      "loss": 3.9819,
      "step": 1918
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005757,
      "loss": 3.9919,
      "step": 1919
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005759999999999999,
      "loss": 3.9567,
      "step": 1920
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005763,
      "loss": 3.9865,
      "step": 1921
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005765999999999999,
      "loss": 3.986,
      "step": 1922
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005769,
      "loss": 3.9941,
      "step": 1923
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005771999999999999,
      "loss": 3.9946,
      "step": 1924
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005775,
      "loss": 3.9909,
      "step": 1925
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005778,
      "loss": 4.0136,
      "step": 1926
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005781,
      "loss": 4.0018,
      "step": 1927
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005784,
      "loss": 3.9364,
      "step": 1928
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005786999999999999,
      "loss": 3.9154,
      "step": 1929
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.000579,
      "loss": 3.948,
      "step": 1930
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005792999999999999,
      "loss": 3.9948,
      "step": 1931
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005795999999999999,
      "loss": 3.9623,
      "step": 1932
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005799,
      "loss": 3.9731,
      "step": 1933
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005801999999999999,
      "loss": 3.9725,
      "step": 1934
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005805,
      "loss": 3.9708,
      "step": 1935
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0005807999999999999,
      "loss": 3.9157,
      "step": 1936
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005811,
      "loss": 3.9424,
      "step": 1937
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005813999999999999,
      "loss": 3.9781,
      "step": 1938
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005817,
      "loss": 4.0248,
      "step": 1939
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005819999999999999,
      "loss": 3.9873,
      "step": 1940
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005823,
      "loss": 3.9777,
      "step": 1941
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005826,
      "loss": 3.9033,
      "step": 1942
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005829,
      "loss": 3.8921,
      "step": 1943
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005832,
      "loss": 3.9986,
      "step": 1944
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005834999999999999,
      "loss": 3.9923,
      "step": 1945
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005838,
      "loss": 3.9622,
      "step": 1946
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005840999999999999,
      "loss": 3.9483,
      "step": 1947
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005843999999999999,
      "loss": 3.9998,
      "step": 1948
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005847,
      "loss": 3.9631,
      "step": 1949
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005849999999999999,
      "loss": 3.9638,
      "step": 1950
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005853,
      "loss": 3.9566,
      "step": 1951
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005855999999999999,
      "loss": 3.9082,
      "step": 1952
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005859,
      "loss": 3.994,
      "step": 1953
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005861999999999999,
      "loss": 4.0098,
      "step": 1954
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005865,
      "loss": 3.9609,
      "step": 1955
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005868,
      "loss": 3.9165,
      "step": 1956
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005871,
      "loss": 4.0147,
      "step": 1957
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005874,
      "loss": 3.9717,
      "step": 1958
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005876999999999999,
      "loss": 3.9311,
      "step": 1959
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.000588,
      "loss": 3.9642,
      "step": 1960
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005882999999999999,
      "loss": 3.9173,
      "step": 1961
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005885999999999999,
      "loss": 3.9033,
      "step": 1962
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005889,
      "loss": 3.9325,
      "step": 1963
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005891999999999999,
      "loss": 3.954,
      "step": 1964
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005895,
      "loss": 3.9876,
      "step": 1965
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005897999999999999,
      "loss": 3.9586,
      "step": 1966
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005901,
      "loss": 3.9561,
      "step": 1967
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005903999999999999,
      "loss": 3.9322,
      "step": 1968
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005907,
      "loss": 3.9493,
      "step": 1969
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005909999999999999,
      "loss": 3.9936,
      "step": 1970
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005913,
      "loss": 4.0402,
      "step": 1971
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005916,
      "loss": 3.9784,
      "step": 1972
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005919,
      "loss": 4.0273,
      "step": 1973
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005922,
      "loss": 4.0071,
      "step": 1974
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005924999999999999,
      "loss": 3.9898,
      "step": 1975
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0005928,
      "loss": 3.966,
      "step": 1976
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005930999999999999,
      "loss": 4.0065,
      "step": 1977
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005933999999999999,
      "loss": 3.9651,
      "step": 1978
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005937,
      "loss": 3.9377,
      "step": 1979
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005939999999999999,
      "loss": 3.9567,
      "step": 1980
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005943,
      "loss": 3.9288,
      "step": 1981
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005945999999999999,
      "loss": 3.9288,
      "step": 1982
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005949,
      "loss": 3.966,
      "step": 1983
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005951999999999999,
      "loss": 3.9475,
      "step": 1984
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005955,
      "loss": 3.9572,
      "step": 1985
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005958,
      "loss": 3.9683,
      "step": 1986
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005961,
      "loss": 3.9717,
      "step": 1987
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005964,
      "loss": 3.9286,
      "step": 1988
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005967,
      "loss": 3.9624,
      "step": 1989
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.000597,
      "loss": 3.9477,
      "step": 1990
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005972999999999999,
      "loss": 3.9115,
      "step": 1991
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005976,
      "loss": 4.0015,
      "step": 1992
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005979,
      "loss": 3.9921,
      "step": 1993
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005981999999999999,
      "loss": 3.9782,
      "step": 1994
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005985,
      "loss": 3.9871,
      "step": 1995
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005987999999999999,
      "loss": 3.9431,
      "step": 1996
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005991,
      "loss": 3.9612,
      "step": 1997
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0005993999999999999,
      "loss": 3.9456,
      "step": 1998
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0005997,
      "loss": 3.915,
      "step": 1999
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.9405,
      "step": 2000
    },
    {
      "epoch": 1.21,
      "eval_loss": 3.833540201187134,
      "eval_perplexity": 48.174896240234375,
      "eval_runtime": 46.9018,
      "eval_samples_per_second": 2132.115,
      "eval_steps_per_second": 66.629,
      "step": 2000
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.954,
      "step": 2001
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.9501,
      "step": 2002
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.9747,
      "step": 2003
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 4.0305,
      "step": 2004
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.9725,
      "step": 2005
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.9359,
      "step": 2006
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.9296,
      "step": 2007
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.9366,
      "step": 2008
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.9849,
      "step": 2009
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.9645,
      "step": 2010
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.9246,
      "step": 2011
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.9514,
      "step": 2012
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.967,
      "step": 2013
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.9812,
      "step": 2014
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.9503,
      "step": 2015
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.9712,
      "step": 2016
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0006,
      "loss": 3.9501,
      "step": 2017
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 3.92,
      "step": 2018
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 3.9139,
      "step": 2019
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 3.9152,
      "step": 2020
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 3.9344,
      "step": 2021
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 3.9316,
      "step": 2022
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 3.8758,
      "step": 2023
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 3.9613,
      "step": 2024
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 3.9125,
      "step": 2025
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 3.9012,
      "step": 2026
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 3.962,
      "step": 2027
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 3.9574,
      "step": 2028
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 4.0588,
      "step": 2029
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 3.984,
      "step": 2030
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 4.1151,
      "step": 2031
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 4.073,
      "step": 2032
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 4.203,
      "step": 2033
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 4.1915,
      "step": 2034
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 4.0958,
      "step": 2035
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 4.1476,
      "step": 2036
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0006,
      "loss": 4.1386,
      "step": 2037
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 4.1561,
      "step": 2038
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 4.1215,
      "step": 2039
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 4.0955,
      "step": 2040
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 4.0759,
      "step": 2041
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 4.0871,
      "step": 2042
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 4.0731,
      "step": 2043
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 4.0425,
      "step": 2044
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 4.0208,
      "step": 2045
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 4.0064,
      "step": 2046
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 4.0133,
      "step": 2047
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 4.0371,
      "step": 2048
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 4.0212,
      "step": 2049
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 3.986,
      "step": 2050
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 4.0076,
      "step": 2051
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 3.9787,
      "step": 2052
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 3.9419,
      "step": 2053
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 3.9339,
      "step": 2054
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 3.9868,
      "step": 2055
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 3.926,
      "step": 2056
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0006,
      "loss": 3.9599,
      "step": 2057
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 4.0244,
      "step": 2058
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.9741,
      "step": 2059
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.9664,
      "step": 2060
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.9516,
      "step": 2061
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.9184,
      "step": 2062
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.95,
      "step": 2063
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.9627,
      "step": 2064
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.9566,
      "step": 2065
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.9242,
      "step": 2066
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.9795,
      "step": 2067
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.9555,
      "step": 2068
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.9361,
      "step": 2069
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.9588,
      "step": 2070
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.9766,
      "step": 2071
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 4.0255,
      "step": 2072
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 4.0336,
      "step": 2073
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 4.0691,
      "step": 2074
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.9602,
      "step": 2075
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.9717,
      "step": 2076
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.996,
      "step": 2077
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0006,
      "loss": 3.9848,
      "step": 2078
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9884,
      "step": 2079
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9655,
      "step": 2080
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9359,
      "step": 2081
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9122,
      "step": 2082
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 4.0065,
      "step": 2083
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9542,
      "step": 2084
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9376,
      "step": 2085
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.8986,
      "step": 2086
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9599,
      "step": 2087
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9625,
      "step": 2088
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9646,
      "step": 2089
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9169,
      "step": 2090
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9083,
      "step": 2091
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9194,
      "step": 2092
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9077,
      "step": 2093
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9971,
      "step": 2094
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9168,
      "step": 2095
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.9391,
      "step": 2096
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 3.978,
      "step": 2097
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0006,
      "loss": 4.0025,
      "step": 2098
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.9299,
      "step": 2099
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.923,
      "step": 2100
    },
    {
      "epoch": 1.26,
      "eval_loss": 3.8476433753967285,
      "eval_perplexity": 48.79921340942383,
      "eval_runtime": 47.5286,
      "eval_samples_per_second": 2103.997,
      "eval_steps_per_second": 65.75,
      "step": 2100
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.9356,
      "step": 2101
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 4.0006,
      "step": 2102
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.9291,
      "step": 2103
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.8929,
      "step": 2104
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.9015,
      "step": 2105
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.9651,
      "step": 2106
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.924,
      "step": 2107
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.9215,
      "step": 2108
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.9188,
      "step": 2109
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.9302,
      "step": 2110
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.9167,
      "step": 2111
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.8914,
      "step": 2112
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.8986,
      "step": 2113
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.9389,
      "step": 2114
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.9292,
      "step": 2115
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.9152,
      "step": 2116
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.8982,
      "step": 2117
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0006,
      "loss": 3.9148,
      "step": 2118
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.9253,
      "step": 2119
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.9638,
      "step": 2120
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.964,
      "step": 2121
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.9406,
      "step": 2122
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.9193,
      "step": 2123
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.9308,
      "step": 2124
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.9695,
      "step": 2125
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.9574,
      "step": 2126
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.9092,
      "step": 2127
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.8779,
      "step": 2128
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.8369,
      "step": 2129
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.9209,
      "step": 2130
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.9543,
      "step": 2131
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.9314,
      "step": 2132
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.8864,
      "step": 2133
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.9372,
      "step": 2134
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.9228,
      "step": 2135
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.9113,
      "step": 2136
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.9125,
      "step": 2137
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0006,
      "loss": 3.8951,
      "step": 2138
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.8895,
      "step": 2139
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.8555,
      "step": 2140
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.8535,
      "step": 2141
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.9129,
      "step": 2142
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.9056,
      "step": 2143
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.9002,
      "step": 2144
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.9035,
      "step": 2145
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.8851,
      "step": 2146
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.8923,
      "step": 2147
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.874,
      "step": 2148
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.8854,
      "step": 2149
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.8968,
      "step": 2150
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.8792,
      "step": 2151
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.8971,
      "step": 2152
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.9292,
      "step": 2153
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.889,
      "step": 2154
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.8408,
      "step": 2155
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.8557,
      "step": 2156
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.8914,
      "step": 2157
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.8542,
      "step": 2158
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0006,
      "loss": 3.8458,
      "step": 2159
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.8486,
      "step": 2160
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.919,
      "step": 2161
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.9785,
      "step": 2162
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.9414,
      "step": 2163
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.909,
      "step": 2164
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.8953,
      "step": 2165
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.8647,
      "step": 2166
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.8816,
      "step": 2167
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.8843,
      "step": 2168
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.891,
      "step": 2169
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.8849,
      "step": 2170
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.868,
      "step": 2171
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.8573,
      "step": 2172
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.8504,
      "step": 2173
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.9284,
      "step": 2174
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.8981,
      "step": 2175
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.8917,
      "step": 2176
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.8533,
      "step": 2177
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.8931,
      "step": 2178
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0006,
      "loss": 3.889,
      "step": 2179
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.8792,
      "step": 2180
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.8559,
      "step": 2181
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.8714,
      "step": 2182
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.8962,
      "step": 2183
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.8944,
      "step": 2184
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.8599,
      "step": 2185
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.906,
      "step": 2186
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.9407,
      "step": 2187
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.9155,
      "step": 2188
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.8677,
      "step": 2189
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.8774,
      "step": 2190
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.8636,
      "step": 2191
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.8477,
      "step": 2192
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.8799,
      "step": 2193
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.9029,
      "step": 2194
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.841,
      "step": 2195
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.8256,
      "step": 2196
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.8854,
      "step": 2197
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.9012,
      "step": 2198
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0006,
      "loss": 3.8832,
      "step": 2199
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.9089,
      "step": 2200
    },
    {
      "epoch": 1.31,
      "eval_loss": 3.755662441253662,
      "eval_perplexity": 44.56911087036133,
      "eval_runtime": 47.5448,
      "eval_samples_per_second": 2103.281,
      "eval_steps_per_second": 65.728,
      "step": 2200
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8654,
      "step": 2201
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8842,
      "step": 2202
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8692,
      "step": 2203
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8475,
      "step": 2204
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8672,
      "step": 2205
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8773,
      "step": 2206
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8568,
      "step": 2207
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8664,
      "step": 2208
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8798,
      "step": 2209
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8637,
      "step": 2210
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.9074,
      "step": 2211
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8838,
      "step": 2212
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8246,
      "step": 2213
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8556,
      "step": 2214
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8543,
      "step": 2215
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8848,
      "step": 2216
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8845,
      "step": 2217
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.9068,
      "step": 2218
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0006,
      "loss": 3.8927,
      "step": 2219
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8621,
      "step": 2220
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.7808,
      "step": 2221
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8373,
      "step": 2222
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8356,
      "step": 2223
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8539,
      "step": 2224
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8787,
      "step": 2225
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8232,
      "step": 2226
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8309,
      "step": 2227
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8441,
      "step": 2228
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8115,
      "step": 2229
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8581,
      "step": 2230
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8943,
      "step": 2231
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8995,
      "step": 2232
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.9041,
      "step": 2233
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.9187,
      "step": 2234
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8413,
      "step": 2235
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8086,
      "step": 2236
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.9175,
      "step": 2237
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.9036,
      "step": 2238
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8759,
      "step": 2239
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0006,
      "loss": 3.8257,
      "step": 2240
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8473,
      "step": 2241
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8429,
      "step": 2242
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8272,
      "step": 2243
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8206,
      "step": 2244
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8362,
      "step": 2245
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8487,
      "step": 2246
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8414,
      "step": 2247
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8355,
      "step": 2248
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.9082,
      "step": 2249
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8121,
      "step": 2250
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.7901,
      "step": 2251
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8336,
      "step": 2252
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8836,
      "step": 2253
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8457,
      "step": 2254
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8294,
      "step": 2255
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8537,
      "step": 2256
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8676,
      "step": 2257
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.858,
      "step": 2258
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.8126,
      "step": 2259
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0006,
      "loss": 3.787,
      "step": 2260
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8868,
      "step": 2261
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8357,
      "step": 2262
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.819,
      "step": 2263
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8075,
      "step": 2264
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8661,
      "step": 2265
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8661,
      "step": 2266
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8704,
      "step": 2267
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8289,
      "step": 2268
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.872,
      "step": 2269
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8553,
      "step": 2270
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8354,
      "step": 2271
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.847,
      "step": 2272
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8396,
      "step": 2273
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8159,
      "step": 2274
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8265,
      "step": 2275
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8937,
      "step": 2276
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8691,
      "step": 2277
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8304,
      "step": 2278
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8114,
      "step": 2279
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0006,
      "loss": 3.8545,
      "step": 2280
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8812,
      "step": 2281
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8779,
      "step": 2282
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8248,
      "step": 2283
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8245,
      "step": 2284
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8579,
      "step": 2285
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8105,
      "step": 2286
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8097,
      "step": 2287
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8894,
      "step": 2288
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8681,
      "step": 2289
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8761,
      "step": 2290
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8833,
      "step": 2291
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8525,
      "step": 2292
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8656,
      "step": 2293
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8734,
      "step": 2294
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8981,
      "step": 2295
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.9054,
      "step": 2296
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8639,
      "step": 2297
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8661,
      "step": 2298
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8904,
      "step": 2299
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0006,
      "loss": 3.8391,
      "step": 2300
    },
    {
      "epoch": 1.35,
      "eval_loss": 3.717740058898926,
      "eval_perplexity": 42.939910888671875,
      "eval_runtime": 49.2526,
      "eval_samples_per_second": 2030.348,
      "eval_steps_per_second": 63.448,
      "step": 2300
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8022,
      "step": 2301
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8464,
      "step": 2302
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8527,
      "step": 2303
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8207,
      "step": 2304
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8118,
      "step": 2305
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8319,
      "step": 2306
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8076,
      "step": 2307
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8102,
      "step": 2308
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8265,
      "step": 2309
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.867,
      "step": 2310
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8826,
      "step": 2311
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8243,
      "step": 2312
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8198,
      "step": 2313
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8109,
      "step": 2314
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.7867,
      "step": 2315
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.7756,
      "step": 2316
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8261,
      "step": 2317
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.81,
      "step": 2318
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8234,
      "step": 2319
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.8216,
      "step": 2320
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0006,
      "loss": 3.7676,
      "step": 2321
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.7534,
      "step": 2322
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.7628,
      "step": 2323
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.8358,
      "step": 2324
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.8708,
      "step": 2325
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.8294,
      "step": 2326
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.8117,
      "step": 2327
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.8207,
      "step": 2328
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.8248,
      "step": 2329
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.8476,
      "step": 2330
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.8586,
      "step": 2331
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.8203,
      "step": 2332
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.8557,
      "step": 2333
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.8342,
      "step": 2334
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.8106,
      "step": 2335
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.8226,
      "step": 2336
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.7961,
      "step": 2337
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.7902,
      "step": 2338
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.7792,
      "step": 2339
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.9061,
      "step": 2340
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0006,
      "loss": 3.857,
      "step": 2341
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8066,
      "step": 2342
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.7933,
      "step": 2343
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.814,
      "step": 2344
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8354,
      "step": 2345
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8359,
      "step": 2346
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8178,
      "step": 2347
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8006,
      "step": 2348
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8483,
      "step": 2349
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8707,
      "step": 2350
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8589,
      "step": 2351
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8469,
      "step": 2352
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.7765,
      "step": 2353
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.7681,
      "step": 2354
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8363,
      "step": 2355
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8406,
      "step": 2356
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8169,
      "step": 2357
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8225,
      "step": 2358
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8052,
      "step": 2359
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8106,
      "step": 2360
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0006,
      "loss": 3.8086,
      "step": 2361
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.825,
      "step": 2362
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.8255,
      "step": 2363
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.7785,
      "step": 2364
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.7801,
      "step": 2365
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.7999,
      "step": 2366
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.8548,
      "step": 2367
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.7921,
      "step": 2368
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.7592,
      "step": 2369
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.7672,
      "step": 2370
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.7629,
      "step": 2371
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.7827,
      "step": 2372
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.7838,
      "step": 2373
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.7655,
      "step": 2374
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.7854,
      "step": 2375
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.8355,
      "step": 2376
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.8396,
      "step": 2377
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.8285,
      "step": 2378
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.7436,
      "step": 2379
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.8065,
      "step": 2380
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0006,
      "loss": 3.804,
      "step": 2381
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.7966,
      "step": 2382
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.7987,
      "step": 2383
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.7913,
      "step": 2384
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.812,
      "step": 2385
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.8143,
      "step": 2386
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.7455,
      "step": 2387
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.7292,
      "step": 2388
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.7553,
      "step": 2389
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.8247,
      "step": 2390
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.8642,
      "step": 2391
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.8044,
      "step": 2392
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.7744,
      "step": 2393
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.7629,
      "step": 2394
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.7722,
      "step": 2395
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.7597,
      "step": 2396
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.7753,
      "step": 2397
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.8419,
      "step": 2398
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.7823,
      "step": 2399
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.7952,
      "step": 2400
    },
    {
      "epoch": 1.4,
      "eval_loss": 3.6893577575683594,
      "eval_perplexity": 41.76032638549805,
      "eval_runtime": 47.441,
      "eval_samples_per_second": 2107.883,
      "eval_steps_per_second": 65.871,
      "step": 2400
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.8083,
      "step": 2401
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0006,
      "loss": 3.8251,
      "step": 2402
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.8014,
      "step": 2403
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.8475,
      "step": 2404
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.8115,
      "step": 2405
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.8004,
      "step": 2406
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.8088,
      "step": 2407
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.7496,
      "step": 2408
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.7417,
      "step": 2409
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.8227,
      "step": 2410
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.8632,
      "step": 2411
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.8601,
      "step": 2412
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.8192,
      "step": 2413
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.7736,
      "step": 2414
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.8283,
      "step": 2415
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.8176,
      "step": 2416
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.7916,
      "step": 2417
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.7805,
      "step": 2418
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.7899,
      "step": 2419
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.8125,
      "step": 2420
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.8076,
      "step": 2421
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0006,
      "loss": 3.8167,
      "step": 2422
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.7848,
      "step": 2423
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.7686,
      "step": 2424
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.7684,
      "step": 2425
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.7893,
      "step": 2426
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.8266,
      "step": 2427
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.8179,
      "step": 2428
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.8099,
      "step": 2429
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.8528,
      "step": 2430
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.7737,
      "step": 2431
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.8218,
      "step": 2432
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.8321,
      "step": 2433
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.8111,
      "step": 2434
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.8791,
      "step": 2435
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.8011,
      "step": 2436
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.7536,
      "step": 2437
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.7421,
      "step": 2438
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.7959,
      "step": 2439
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.8016,
      "step": 2440
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.7806,
      "step": 2441
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0006,
      "loss": 3.7713,
      "step": 2442
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7825,
      "step": 2443
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7961,
      "step": 2444
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7967,
      "step": 2445
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7891,
      "step": 2446
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7878,
      "step": 2447
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7719,
      "step": 2448
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7856,
      "step": 2449
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.8235,
      "step": 2450
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7973,
      "step": 2451
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7881,
      "step": 2452
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.765,
      "step": 2453
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.8111,
      "step": 2454
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7506,
      "step": 2455
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7717,
      "step": 2456
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7846,
      "step": 2457
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7976,
      "step": 2458
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.8062,
      "step": 2459
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7639,
      "step": 2460
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7397,
      "step": 2461
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0006,
      "loss": 3.7596,
      "step": 2462
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.8248,
      "step": 2463
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7761,
      "step": 2464
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7593,
      "step": 2465
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7975,
      "step": 2466
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.8142,
      "step": 2467
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7757,
      "step": 2468
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7601,
      "step": 2469
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7519,
      "step": 2470
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.786,
      "step": 2471
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7523,
      "step": 2472
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7265,
      "step": 2473
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7815,
      "step": 2474
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7703,
      "step": 2475
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7551,
      "step": 2476
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7374,
      "step": 2477
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7486,
      "step": 2478
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.8185,
      "step": 2479
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7756,
      "step": 2480
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7516,
      "step": 2481
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7655,
      "step": 2482
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0006,
      "loss": 3.7737,
      "step": 2483
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7552,
      "step": 2484
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7197,
      "step": 2485
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7964,
      "step": 2486
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7427,
      "step": 2487
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7456,
      "step": 2488
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7417,
      "step": 2489
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.8041,
      "step": 2490
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7992,
      "step": 2491
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7888,
      "step": 2492
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.8085,
      "step": 2493
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7831,
      "step": 2494
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.8231,
      "step": 2495
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7803,
      "step": 2496
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7539,
      "step": 2497
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7849,
      "step": 2498
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.8222,
      "step": 2499
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7896,
      "step": 2500
    },
    {
      "epoch": 1.45,
      "eval_loss": 3.654545307159424,
      "eval_perplexity": 40.33282470703125,
      "eval_runtime": 47.5132,
      "eval_samples_per_second": 2104.677,
      "eval_steps_per_second": 65.771,
      "step": 2500
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7654,
      "step": 2501
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7417,
      "step": 2502
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0006,
      "loss": 3.7731,
      "step": 2503
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7711,
      "step": 2504
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7421,
      "step": 2505
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7354,
      "step": 2506
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.788,
      "step": 2507
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7567,
      "step": 2508
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7331,
      "step": 2509
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7455,
      "step": 2510
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7668,
      "step": 2511
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7744,
      "step": 2512
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.76,
      "step": 2513
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7779,
      "step": 2514
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7935,
      "step": 2515
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7609,
      "step": 2516
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.74,
      "step": 2517
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7651,
      "step": 2518
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7733,
      "step": 2519
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7529,
      "step": 2520
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7345,
      "step": 2521
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.8016,
      "step": 2522
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0006,
      "loss": 3.7867,
      "step": 2523
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7619,
      "step": 2524
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7305,
      "step": 2525
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7456,
      "step": 2526
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7368,
      "step": 2527
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7506,
      "step": 2528
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7734,
      "step": 2529
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.6744,
      "step": 2530
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7244,
      "step": 2531
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7554,
      "step": 2532
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7385,
      "step": 2533
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.8164,
      "step": 2534
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.798,
      "step": 2535
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7646,
      "step": 2536
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7331,
      "step": 2537
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7389,
      "step": 2538
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7464,
      "step": 2539
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.741,
      "step": 2540
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.748,
      "step": 2541
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7477,
      "step": 2542
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7583,
      "step": 2543
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0006,
      "loss": 3.7732,
      "step": 2544
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.7722,
      "step": 2545
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.7775,
      "step": 2546
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.7591,
      "step": 2547
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.7333,
      "step": 2548
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.6916,
      "step": 2549
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.772,
      "step": 2550
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.7576,
      "step": 2551
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.7364,
      "step": 2552
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.7373,
      "step": 2553
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.756,
      "step": 2554
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.7793,
      "step": 2555
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.7845,
      "step": 2556
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.7619,
      "step": 2557
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.762,
      "step": 2558
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.7392,
      "step": 2559
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.7693,
      "step": 2560
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.729,
      "step": 2561
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.7857,
      "step": 2562
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.7683,
      "step": 2563
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0006,
      "loss": 3.7331,
      "step": 2564
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7273,
      "step": 2565
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7368,
      "step": 2566
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7197,
      "step": 2567
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7023,
      "step": 2568
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.6947,
      "step": 2569
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.72,
      "step": 2570
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7349,
      "step": 2571
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7396,
      "step": 2572
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7197,
      "step": 2573
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7184,
      "step": 2574
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7239,
      "step": 2575
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7273,
      "step": 2576
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7524,
      "step": 2577
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7589,
      "step": 2578
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7464,
      "step": 2579
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7382,
      "step": 2580
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7655,
      "step": 2581
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7616,
      "step": 2582
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.7032,
      "step": 2583
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0006,
      "loss": 3.6628,
      "step": 2584
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7479,
      "step": 2585
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7653,
      "step": 2586
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7882,
      "step": 2587
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7645,
      "step": 2588
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7784,
      "step": 2589
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7814,
      "step": 2590
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7466,
      "step": 2591
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7395,
      "step": 2592
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.761,
      "step": 2593
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7338,
      "step": 2594
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.706,
      "step": 2595
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7165,
      "step": 2596
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7056,
      "step": 2597
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7097,
      "step": 2598
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.732,
      "step": 2599
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7478,
      "step": 2600
    },
    {
      "epoch": 1.5,
      "eval_loss": 3.626054525375366,
      "eval_perplexity": 39.213687896728516,
      "eval_runtime": 179.8603,
      "eval_samples_per_second": 555.987,
      "eval_steps_per_second": 17.375,
      "step": 2600
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7518,
      "step": 2601
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.705,
      "step": 2602
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7164,
      "step": 2603
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0006,
      "loss": 3.7476,
      "step": 2604
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7169,
      "step": 2605
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7142,
      "step": 2606
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7062,
      "step": 2607
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.6788,
      "step": 2608
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.8054,
      "step": 2609
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7572,
      "step": 2610
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7519,
      "step": 2611
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7718,
      "step": 2612
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7718,
      "step": 2613
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7089,
      "step": 2614
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.6927,
      "step": 2615
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7037,
      "step": 2616
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7823,
      "step": 2617
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7238,
      "step": 2618
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7481,
      "step": 2619
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7676,
      "step": 2620
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.6991,
      "step": 2621
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7573,
      "step": 2622
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7371,
      "step": 2623
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7255,
      "step": 2624
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0006,
      "loss": 3.7647,
      "step": 2625
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.7333,
      "step": 2626
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.7032,
      "step": 2627
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.7042,
      "step": 2628
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.729,
      "step": 2629
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.72,
      "step": 2630
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.6966,
      "step": 2631
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.6889,
      "step": 2632
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.7385,
      "step": 2633
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.7363,
      "step": 2634
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.7177,
      "step": 2635
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.7202,
      "step": 2636
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.6795,
      "step": 2637
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.7212,
      "step": 2638
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.7281,
      "step": 2639
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.6861,
      "step": 2640
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.7712,
      "step": 2641
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.7596,
      "step": 2642
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.7521,
      "step": 2643
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.7072,
      "step": 2644
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0006,
      "loss": 3.7204,
      "step": 2645
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.6854,
      "step": 2646
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.6943,
      "step": 2647
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.7634,
      "step": 2648
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.712,
      "step": 2649
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.6769,
      "step": 2650
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.6848,
      "step": 2651
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.7787,
      "step": 2652
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.7453,
      "step": 2653
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.742,
      "step": 2654
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.7405,
      "step": 2655
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.6991,
      "step": 2656
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.7331,
      "step": 2657
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.7386,
      "step": 2658
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.7262,
      "step": 2659
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.7625,
      "step": 2660
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.7372,
      "step": 2661
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.7038,
      "step": 2662
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.7138,
      "step": 2663
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.7967,
      "step": 2664
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0006,
      "loss": 3.7343,
      "step": 2665
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.711,
      "step": 2666
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.6982,
      "step": 2667
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.625,
      "step": 2668
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.7056,
      "step": 2669
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.7325,
      "step": 2670
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.7485,
      "step": 2671
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.6786,
      "step": 2672
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.7083,
      "step": 2673
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.7135,
      "step": 2674
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.7266,
      "step": 2675
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.7418,
      "step": 2676
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.7241,
      "step": 2677
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.7226,
      "step": 2678
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.7558,
      "step": 2679
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.7234,
      "step": 2680
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.7186,
      "step": 2681
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.6891,
      "step": 2682
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.7105,
      "step": 2683
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.683,
      "step": 2684
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0006,
      "loss": 3.6452,
      "step": 2685
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.6545,
      "step": 2686
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.7321,
      "step": 2687
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.6961,
      "step": 2688
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.7454,
      "step": 2689
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.7431,
      "step": 2690
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.7105,
      "step": 2691
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.6851,
      "step": 2692
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.7455,
      "step": 2693
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.7144,
      "step": 2694
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.6557,
      "step": 2695
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.7304,
      "step": 2696
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.7391,
      "step": 2697
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.7514,
      "step": 2698
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.7479,
      "step": 2699
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.7572,
      "step": 2700
    },
    {
      "epoch": 1.55,
      "eval_loss": 3.6021976470947266,
      "eval_perplexity": 38.29977035522461,
      "eval_runtime": 47.031,
      "eval_samples_per_second": 2126.259,
      "eval_steps_per_second": 66.446,
      "step": 2700
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.7213,
      "step": 2701
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.6982,
      "step": 2702
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.7372,
      "step": 2703
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.756,
      "step": 2704
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.7158,
      "step": 2705
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0006,
      "loss": 3.7025,
      "step": 2706
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.7497,
      "step": 2707
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.7297,
      "step": 2708
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.7145,
      "step": 2709
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.7016,
      "step": 2710
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.6579,
      "step": 2711
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.687,
      "step": 2712
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.7339,
      "step": 2713
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.7288,
      "step": 2714
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.6699,
      "step": 2715
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.6992,
      "step": 2716
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.6756,
      "step": 2717
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.6988,
      "step": 2718
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.7239,
      "step": 2719
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.6907,
      "step": 2720
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.7129,
      "step": 2721
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.7303,
      "step": 2722
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.7358,
      "step": 2723
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.7121,
      "step": 2724
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.7289,
      "step": 2725
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0006,
      "loss": 3.7629,
      "step": 2726
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.7702,
      "step": 2727
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.7381,
      "step": 2728
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.7461,
      "step": 2729
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.7625,
      "step": 2730
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.7105,
      "step": 2731
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.7032,
      "step": 2732
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.7138,
      "step": 2733
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.7241,
      "step": 2734
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.7175,
      "step": 2735
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.6521,
      "step": 2736
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.6449,
      "step": 2737
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.6778,
      "step": 2738
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.7188,
      "step": 2739
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.6964,
      "step": 2740
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.6922,
      "step": 2741
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.711,
      "step": 2742
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.7317,
      "step": 2743
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.7419,
      "step": 2744
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.6992,
      "step": 2745
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0006,
      "loss": 3.6804,
      "step": 2746
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.7163,
      "step": 2747
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.7195,
      "step": 2748
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.6601,
      "step": 2749
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.6388,
      "step": 2750
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.6534,
      "step": 2751
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.726,
      "step": 2752
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.7001,
      "step": 2753
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.6718,
      "step": 2754
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.7369,
      "step": 2755
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.6924,
      "step": 2756
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.702,
      "step": 2757
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.7258,
      "step": 2758
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.7535,
      "step": 2759
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.673,
      "step": 2760
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.6877,
      "step": 2761
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.7277,
      "step": 2762
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.6677,
      "step": 2763
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.6794,
      "step": 2764
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.7045,
      "step": 2765
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.7212,
      "step": 2766
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.7342,
      "step": 2767
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.7051,
      "step": 2768
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.6949,
      "step": 2769
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.7131,
      "step": 2770
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.7598,
      "step": 2771
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.7113,
      "step": 2772
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.7109,
      "step": 2773
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.7049,
      "step": 2774
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.7149,
      "step": 2775
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.68,
      "step": 2776
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.7082,
      "step": 2777
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.709,
      "step": 2778
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.6463,
      "step": 2779
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.6987,
      "step": 2780
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.7,
      "step": 2781
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.7032,
      "step": 2782
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.7064,
      "step": 2783
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.7182,
      "step": 2784
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.6588,
      "step": 2785
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.6639,
      "step": 2786
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.7041,
      "step": 2787
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.7186,
      "step": 2788
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.7229,
      "step": 2789
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.7289,
      "step": 2790
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6862,
      "step": 2791
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6627,
      "step": 2792
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.676,
      "step": 2793
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6674,
      "step": 2794
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6537,
      "step": 2795
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6719,
      "step": 2796
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6857,
      "step": 2797
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6831,
      "step": 2798
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6337,
      "step": 2799
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6807,
      "step": 2800
    },
    {
      "epoch": 1.6,
      "eval_loss": 3.579197645187378,
      "eval_perplexity": 37.456111907958984,
      "eval_runtime": 47.1639,
      "eval_samples_per_second": 2120.267,
      "eval_steps_per_second": 66.258,
      "step": 2800
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6804,
      "step": 2801
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6859,
      "step": 2802
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6635,
      "step": 2803
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6384,
      "step": 2804
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6293,
      "step": 2805
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6972,
      "step": 2806
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.7337,
      "step": 2807
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.6922,
      "step": 2808
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.681,
      "step": 2809
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.7118,
      "step": 2810
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.7038,
      "step": 2811
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.7291,
      "step": 2812
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.6866,
      "step": 2813
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5952,
      "step": 2814
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.685,
      "step": 2815
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.6691,
      "step": 2816
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.6614,
      "step": 2817
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.7254,
      "step": 2818
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.7163,
      "step": 2819
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.7031,
      "step": 2820
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.657,
      "step": 2821
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.6183,
      "step": 2822
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.6885,
      "step": 2823
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.6626,
      "step": 2824
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.6539,
      "step": 2825
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.6728,
      "step": 2826
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.663,
      "step": 2827
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.6296,
      "step": 2828
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.6617,
      "step": 2829
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.7163,
      "step": 2830
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.65,
      "step": 2831
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.6738,
      "step": 2832
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.6884,
      "step": 2833
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.661,
      "step": 2834
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.6587,
      "step": 2835
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.6836,
      "step": 2836
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.677,
      "step": 2837
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.6823,
      "step": 2838
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.6891,
      "step": 2839
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.7088,
      "step": 2840
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.7223,
      "step": 2841
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.6277,
      "step": 2842
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.6117,
      "step": 2843
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.6273,
      "step": 2844
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.6822,
      "step": 2845
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.6749,
      "step": 2846
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.7021,
      "step": 2847
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6911,
      "step": 2848
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.684,
      "step": 2849
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6378,
      "step": 2850
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6955,
      "step": 2851
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6917,
      "step": 2852
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6781,
      "step": 2853
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.7406,
      "step": 2854
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.629,
      "step": 2855
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6673,
      "step": 2856
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6702,
      "step": 2857
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6137,
      "step": 2858
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6336,
      "step": 2859
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6525,
      "step": 2860
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6991,
      "step": 2861
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.639,
      "step": 2862
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6557,
      "step": 2863
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6912,
      "step": 2864
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.7364,
      "step": 2865
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6612,
      "step": 2866
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6476,
      "step": 2867
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6615,
      "step": 2868
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6739,
      "step": 2869
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6408,
      "step": 2870
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6961,
      "step": 2871
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.695,
      "step": 2872
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6639,
      "step": 2873
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6386,
      "step": 2874
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6815,
      "step": 2875
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6706,
      "step": 2876
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6689,
      "step": 2877
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6653,
      "step": 2878
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6648,
      "step": 2879
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6679,
      "step": 2880
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6871,
      "step": 2881
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.7284,
      "step": 2882
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6975,
      "step": 2883
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6918,
      "step": 2884
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6706,
      "step": 2885
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6838,
      "step": 2886
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6875,
      "step": 2887
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6932,
      "step": 2888
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6984,
      "step": 2889
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6873,
      "step": 2890
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6621,
      "step": 2891
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6683,
      "step": 2892
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6656,
      "step": 2893
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.7335,
      "step": 2894
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.7029,
      "step": 2895
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6557,
      "step": 2896
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6558,
      "step": 2897
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6862,
      "step": 2898
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.679,
      "step": 2899
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.684,
      "step": 2900
    },
    {
      "epoch": 1.65,
      "eval_loss": 3.553069591522217,
      "eval_perplexity": 36.49708557128906,
      "eval_runtime": 47.1156,
      "eval_samples_per_second": 2122.439,
      "eval_steps_per_second": 66.326,
      "step": 2900
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6518,
      "step": 2901
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.653,
      "step": 2902
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6699,
      "step": 2903
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6962,
      "step": 2904
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6618,
      "step": 2905
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5956,
      "step": 2906
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6323,
      "step": 2907
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6526,
      "step": 2908
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6441,
      "step": 2909
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6961,
      "step": 2910
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.676,
      "step": 2911
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6515,
      "step": 2912
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6467,
      "step": 2913
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6881,
      "step": 2914
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6329,
      "step": 2915
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6211,
      "step": 2916
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6903,
      "step": 2917
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6981,
      "step": 2918
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6629,
      "step": 2919
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.643,
      "step": 2920
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.7377,
      "step": 2921
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.7224,
      "step": 2922
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6856,
      "step": 2923
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6653,
      "step": 2924
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6965,
      "step": 2925
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6567,
      "step": 2926
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6349,
      "step": 2927
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6492,
      "step": 2928
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6886,
      "step": 2929
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6875,
      "step": 2930
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6902,
      "step": 2931
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6744,
      "step": 2932
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.681,
      "step": 2933
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6697,
      "step": 2934
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6656,
      "step": 2935
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6432,
      "step": 2936
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.7179,
      "step": 2937
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6845,
      "step": 2938
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6622,
      "step": 2939
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6763,
      "step": 2940
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6691,
      "step": 2941
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6282,
      "step": 2942
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6238,
      "step": 2943
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6619,
      "step": 2944
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6296,
      "step": 2945
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6253,
      "step": 2946
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6404,
      "step": 2947
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6373,
      "step": 2948
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6461,
      "step": 2949
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6306,
      "step": 2950
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.65,
      "step": 2951
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6783,
      "step": 2952
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6599,
      "step": 2953
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.5975,
      "step": 2954
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6093,
      "step": 2955
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6153,
      "step": 2956
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.614,
      "step": 2957
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6601,
      "step": 2958
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6935,
      "step": 2959
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.7084,
      "step": 2960
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6271,
      "step": 2961
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6308,
      "step": 2962
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6342,
      "step": 2963
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6524,
      "step": 2964
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6606,
      "step": 2965
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.696,
      "step": 2966
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.7061,
      "step": 2967
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6956,
      "step": 2968
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6553,
      "step": 2969
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.7101,
      "step": 2970
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.7072,
      "step": 2971
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.635,
      "step": 2972
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6419,
      "step": 2973
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6083,
      "step": 2974
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.5968,
      "step": 2975
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.626,
      "step": 2976
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6266,
      "step": 2977
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6304,
      "step": 2978
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6662,
      "step": 2979
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6987,
      "step": 2980
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6109,
      "step": 2981
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6535,
      "step": 2982
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6905,
      "step": 2983
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.65,
      "step": 2984
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6518,
      "step": 2985
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6476,
      "step": 2986
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6334,
      "step": 2987
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6536,
      "step": 2988
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.692,
      "step": 2989
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6944,
      "step": 2990
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.7074,
      "step": 2991
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.7244,
      "step": 2992
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6382,
      "step": 2993
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6432,
      "step": 2994
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6468,
      "step": 2995
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6356,
      "step": 2996
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6261,
      "step": 2997
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6286,
      "step": 2998
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6597,
      "step": 2999
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.67,
      "step": 3000
    },
    {
      "epoch": 1.7,
      "eval_loss": 3.5297343730926514,
      "eval_perplexity": 35.67425537109375,
      "eval_runtime": 47.6317,
      "eval_samples_per_second": 2099.444,
      "eval_steps_per_second": 65.608,
      "step": 3000
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6565,
      "step": 3001
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6441,
      "step": 3002
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6605,
      "step": 3003
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6585,
      "step": 3004
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6591,
      "step": 3005
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6567,
      "step": 3006
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6401,
      "step": 3007
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5617,
      "step": 3008
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5967,
      "step": 3009
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5977,
      "step": 3010
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.632,
      "step": 3011
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.6613,
      "step": 3012
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.682,
      "step": 3013
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.6656,
      "step": 3014
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.6115,
      "step": 3015
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.6526,
      "step": 3016
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.6432,
      "step": 3017
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.6288,
      "step": 3018
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.6537,
      "step": 3019
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.6298,
      "step": 3020
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5926,
      "step": 3021
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5972,
      "step": 3022
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5867,
      "step": 3023
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.6267,
      "step": 3024
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.6397,
      "step": 3025
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.643,
      "step": 3026
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.6037,
      "step": 3027
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.6313,
      "step": 3028
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.6411,
      "step": 3029
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.6699,
      "step": 3030
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.6817,
      "step": 3031
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.6291,
      "step": 3032
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.623,
      "step": 3033
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.6601,
      "step": 3034
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.6691,
      "step": 3035
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5971,
      "step": 3036
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.628,
      "step": 3037
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.6574,
      "step": 3038
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.641,
      "step": 3039
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.629,
      "step": 3040
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.6478,
      "step": 3041
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.6806,
      "step": 3042
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.6891,
      "step": 3043
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.6365,
      "step": 3044
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.6261,
      "step": 3045
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.6403,
      "step": 3046
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.6666,
      "step": 3047
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.677,
      "step": 3048
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.6858,
      "step": 3049
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.6474,
      "step": 3050
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6516,
      "step": 3051
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6542,
      "step": 3052
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6893,
      "step": 3053
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.687,
      "step": 3054
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6647,
      "step": 3055
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6063,
      "step": 3056
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6482,
      "step": 3057
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6247,
      "step": 3058
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6554,
      "step": 3059
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6525,
      "step": 3060
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6276,
      "step": 3061
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.631,
      "step": 3062
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6739,
      "step": 3063
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6176,
      "step": 3064
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6104,
      "step": 3065
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6259,
      "step": 3066
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6556,
      "step": 3067
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6456,
      "step": 3068
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6219,
      "step": 3069
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6343,
      "step": 3070
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6372,
      "step": 3071
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6481,
      "step": 3072
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6552,
      "step": 3073
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6401,
      "step": 3074
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6679,
      "step": 3075
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5949,
      "step": 3076
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6362,
      "step": 3077
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6481,
      "step": 3078
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6276,
      "step": 3079
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6463,
      "step": 3080
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6598,
      "step": 3081
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6692,
      "step": 3082
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6657,
      "step": 3083
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6141,
      "step": 3084
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6141,
      "step": 3085
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6381,
      "step": 3086
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6691,
      "step": 3087
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6551,
      "step": 3088
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.632,
      "step": 3089
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.6132,
      "step": 3090
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5686,
      "step": 3091
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5621,
      "step": 3092
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6331,
      "step": 3093
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6907,
      "step": 3094
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6055,
      "step": 3095
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6317,
      "step": 3096
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6582,
      "step": 3097
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6508,
      "step": 3098
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6393,
      "step": 3099
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.587,
      "step": 3100
    },
    {
      "epoch": 1.75,
      "eval_loss": 3.513960599899292,
      "eval_perplexity": 35.14211654663086,
      "eval_runtime": 47.6003,
      "eval_samples_per_second": 2100.826,
      "eval_steps_per_second": 65.651,
      "step": 3100
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6037,
      "step": 3101
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6565,
      "step": 3102
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6508,
      "step": 3103
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6626,
      "step": 3104
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.641,
      "step": 3105
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.637,
      "step": 3106
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6655,
      "step": 3107
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6414,
      "step": 3108
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6116,
      "step": 3109
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6697,
      "step": 3110
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5908,
      "step": 3111
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.6141,
      "step": 3112
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.6279,
      "step": 3113
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.6248,
      "step": 3114
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.618,
      "step": 3115
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.6142,
      "step": 3116
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5858,
      "step": 3117
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.6058,
      "step": 3118
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.6305,
      "step": 3119
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5861,
      "step": 3120
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5601,
      "step": 3121
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5931,
      "step": 3122
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.6298,
      "step": 3123
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.635,
      "step": 3124
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.6439,
      "step": 3125
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.6737,
      "step": 3126
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.6892,
      "step": 3127
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.6569,
      "step": 3128
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.6479,
      "step": 3129
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5954,
      "step": 3130
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.6731,
      "step": 3131
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.6516,
      "step": 3132
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.6175,
      "step": 3133
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.6259,
      "step": 3134
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5975,
      "step": 3135
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5942,
      "step": 3136
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.6092,
      "step": 3137
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5857,
      "step": 3138
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5873,
      "step": 3139
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.569,
      "step": 3140
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5684,
      "step": 3141
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.6808,
      "step": 3142
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.659,
      "step": 3143
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.6206,
      "step": 3144
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.6322,
      "step": 3145
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.6152,
      "step": 3146
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5906,
      "step": 3147
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.6073,
      "step": 3148
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.6285,
      "step": 3149
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.6099,
      "step": 3150
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.6085,
      "step": 3151
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.5918,
      "step": 3152
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.6161,
      "step": 3153
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.5704,
      "step": 3154
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.6435,
      "step": 3155
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.5921,
      "step": 3156
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.546,
      "step": 3157
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.6887,
      "step": 3158
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.6473,
      "step": 3159
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.6313,
      "step": 3160
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.5804,
      "step": 3161
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.6154,
      "step": 3162
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.6073,
      "step": 3163
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.613,
      "step": 3164
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.6384,
      "step": 3165
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.6581,
      "step": 3166
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.6288,
      "step": 3167
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.6684,
      "step": 3168
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.69,
      "step": 3169
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.6605,
      "step": 3170
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.5902,
      "step": 3171
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.6174,
      "step": 3172
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.6131,
      "step": 3173
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6315,
      "step": 3174
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6082,
      "step": 3175
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6003,
      "step": 3176
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6279,
      "step": 3177
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6944,
      "step": 3178
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.62,
      "step": 3179
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.625,
      "step": 3180
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6624,
      "step": 3181
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6867,
      "step": 3182
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6378,
      "step": 3183
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6095,
      "step": 3184
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6003,
      "step": 3185
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.622,
      "step": 3186
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6343,
      "step": 3187
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6215,
      "step": 3188
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6131,
      "step": 3189
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6237,
      "step": 3190
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6343,
      "step": 3191
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6348,
      "step": 3192
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.6451,
      "step": 3193
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.6398,
      "step": 3194
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.6196,
      "step": 3195
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5873,
      "step": 3196
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5747,
      "step": 3197
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.6281,
      "step": 3198
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.6465,
      "step": 3199
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.6259,
      "step": 3200
    },
    {
      "epoch": 2.02,
      "eval_loss": 3.4913344383239746,
      "eval_perplexity": 34.36025619506836,
      "eval_runtime": 47.5424,
      "eval_samples_per_second": 2103.386,
      "eval_steps_per_second": 65.731,
      "step": 3200
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.6266,
      "step": 3201
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.6323,
      "step": 3202
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.5594,
      "step": 3203
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.5552,
      "step": 3204
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.6171,
      "step": 3205
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.5534,
      "step": 3206
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.5648,
      "step": 3207
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.5708,
      "step": 3208
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.6172,
      "step": 3209
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0006,
      "loss": 3.631,
      "step": 3210
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5906,
      "step": 3211
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5801,
      "step": 3212
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5839,
      "step": 3213
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.6426,
      "step": 3214
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.6067,
      "step": 3215
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5673,
      "step": 3216
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5918,
      "step": 3217
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5932,
      "step": 3218
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5761,
      "step": 3219
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.575,
      "step": 3220
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5993,
      "step": 3221
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5387,
      "step": 3222
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5689,
      "step": 3223
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5878,
      "step": 3224
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5972,
      "step": 3225
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5977,
      "step": 3226
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5711,
      "step": 3227
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5655,
      "step": 3228
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.543,
      "step": 3229
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0006,
      "loss": 3.5894,
      "step": 3230
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.5868,
      "step": 3231
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6102,
      "step": 3232
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6281,
      "step": 3233
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.5814,
      "step": 3234
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.5358,
      "step": 3235
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.5441,
      "step": 3236
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.559,
      "step": 3237
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.5362,
      "step": 3238
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.543,
      "step": 3239
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.5719,
      "step": 3240
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.5742,
      "step": 3241
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.5416,
      "step": 3242
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.533,
      "step": 3243
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.5439,
      "step": 3244
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.5761,
      "step": 3245
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.5452,
      "step": 3246
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.5454,
      "step": 3247
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.5076,
      "step": 3248
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.5733,
      "step": 3249
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0006,
      "loss": 3.6032,
      "step": 3250
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.581,
      "step": 3251
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5654,
      "step": 3252
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.6024,
      "step": 3253
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5834,
      "step": 3254
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5998,
      "step": 3255
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.6129,
      "step": 3256
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.4999,
      "step": 3257
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5425,
      "step": 3258
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5626,
      "step": 3259
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5269,
      "step": 3260
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.6116,
      "step": 3261
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5906,
      "step": 3262
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5773,
      "step": 3263
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5632,
      "step": 3264
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5182,
      "step": 3265
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5625,
      "step": 3266
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5515,
      "step": 3267
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5451,
      "step": 3268
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5613,
      "step": 3269
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0006,
      "loss": 3.5495,
      "step": 3270
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5254,
      "step": 3271
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5322,
      "step": 3272
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5965,
      "step": 3273
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5485,
      "step": 3274
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5247,
      "step": 3275
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5872,
      "step": 3276
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5561,
      "step": 3277
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5469,
      "step": 3278
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5521,
      "step": 3279
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5747,
      "step": 3280
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5838,
      "step": 3281
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5733,
      "step": 3282
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5878,
      "step": 3283
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.6264,
      "step": 3284
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5223,
      "step": 3285
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5136,
      "step": 3286
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5027,
      "step": 3287
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5689,
      "step": 3288
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5564,
      "step": 3289
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5768,
      "step": 3290
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0006,
      "loss": 3.5759,
      "step": 3291
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.585,
      "step": 3292
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.5277,
      "step": 3293
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.571,
      "step": 3294
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.5787,
      "step": 3295
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.5704,
      "step": 3296
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6005,
      "step": 3297
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.555,
      "step": 3298
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.5376,
      "step": 3299
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.5743,
      "step": 3300
    },
    {
      "epoch": 1.63,
      "eval_loss": 3.484829902648926,
      "eval_perplexity": 34.17951583862305,
      "eval_runtime": 93.0994,
      "eval_samples_per_second": 1074.121,
      "eval_steps_per_second": 33.566,
      "step": 3300
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.5181,
      "step": 3301
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.4948,
      "step": 3302
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.5361,
      "step": 3303
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.5841,
      "step": 3304
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.5484,
      "step": 3305
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.5212,
      "step": 3306
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.5628,
      "step": 3307
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.6298,
      "step": 3308
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.5631,
      "step": 3309
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.5448,
      "step": 3310
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0006,
      "loss": 3.5331,
      "step": 3311
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5765,
      "step": 3312
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5372,
      "step": 3313
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5686,
      "step": 3314
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.616,
      "step": 3315
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5622,
      "step": 3316
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5487,
      "step": 3317
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5481,
      "step": 3318
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5791,
      "step": 3319
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5575,
      "step": 3320
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5553,
      "step": 3321
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5446,
      "step": 3322
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5611,
      "step": 3323
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5891,
      "step": 3324
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.6029,
      "step": 3325
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5909,
      "step": 3326
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5544,
      "step": 3327
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5768,
      "step": 3328
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5549,
      "step": 3329
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5693,
      "step": 3330
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0006,
      "loss": 3.5804,
      "step": 3331
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5985,
      "step": 3332
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5639,
      "step": 3333
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5538,
      "step": 3334
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5564,
      "step": 3335
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5748,
      "step": 3336
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6005,
      "step": 3337
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.6028,
      "step": 3338
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5562,
      "step": 3339
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5502,
      "step": 3340
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5518,
      "step": 3341
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5789,
      "step": 3342
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5965,
      "step": 3343
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5478,
      "step": 3344
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5525,
      "step": 3345
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5618,
      "step": 3346
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5963,
      "step": 3347
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5713,
      "step": 3348
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.491,
      "step": 3349
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.5122,
      "step": 3350
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0006,
      "loss": 3.564,
      "step": 3351
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.537,
      "step": 3352
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5578,
      "step": 3353
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5766,
      "step": 3354
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5569,
      "step": 3355
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5532,
      "step": 3356
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5505,
      "step": 3357
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5528,
      "step": 3358
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5006,
      "step": 3359
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5676,
      "step": 3360
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5892,
      "step": 3361
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5809,
      "step": 3362
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5213,
      "step": 3363
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6138,
      "step": 3364
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.6081,
      "step": 3365
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5886,
      "step": 3366
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5651,
      "step": 3367
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5884,
      "step": 3368
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5632,
      "step": 3369
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5283,
      "step": 3370
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.537,
      "step": 3371
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0006,
      "loss": 3.5884,
      "step": 3372
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5935,
      "step": 3373
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5757,
      "step": 3374
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5698,
      "step": 3375
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5672,
      "step": 3376
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5714,
      "step": 3377
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5667,
      "step": 3378
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5285,
      "step": 3379
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.6017,
      "step": 3380
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5886,
      "step": 3381
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.563,
      "step": 3382
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5769,
      "step": 3383
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5676,
      "step": 3384
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5324,
      "step": 3385
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5054,
      "step": 3386
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5551,
      "step": 3387
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5349,
      "step": 3388
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5219,
      "step": 3389
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5199,
      "step": 3390
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5348,
      "step": 3391
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0006,
      "loss": 3.5422,
      "step": 3392
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.522,
      "step": 3393
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.539,
      "step": 3394
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.5716,
      "step": 3395
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.5541,
      "step": 3396
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.5167,
      "step": 3397
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.4907,
      "step": 3398
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.5208,
      "step": 3399
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.4975,
      "step": 3400
    },
    {
      "epoch": 1.68,
      "eval_loss": 3.4686694145202637,
      "eval_perplexity": 33.63052749633789,
      "eval_runtime": 47.1317,
      "eval_samples_per_second": 2121.716,
      "eval_steps_per_second": 66.304,
      "step": 3400
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.5361,
      "step": 3401
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.5877,
      "step": 3402
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6021,
      "step": 3403
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.5421,
      "step": 3404
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.5181,
      "step": 3405
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.5263,
      "step": 3406
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.5525,
      "step": 3407
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.5572,
      "step": 3408
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.5726,
      "step": 3409
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6225,
      "step": 3410
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.6042,
      "step": 3411
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0006,
      "loss": 3.5654,
      "step": 3412
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.5677,
      "step": 3413
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6324,
      "step": 3414
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.5462,
      "step": 3415
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.5204,
      "step": 3416
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.513,
      "step": 3417
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.4872,
      "step": 3418
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.5271,
      "step": 3419
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.5206,
      "step": 3420
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.5319,
      "step": 3421
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.5306,
      "step": 3422
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6076,
      "step": 3423
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.5316,
      "step": 3424
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.5146,
      "step": 3425
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.6092,
      "step": 3426
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.5723,
      "step": 3427
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.5532,
      "step": 3428
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.556,
      "step": 3429
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.5484,
      "step": 3430
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.5586,
      "step": 3431
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0006,
      "loss": 3.5895,
      "step": 3432
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5898,
      "step": 3433
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6024,
      "step": 3434
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.6277,
      "step": 3435
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.552,
      "step": 3436
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5168,
      "step": 3437
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5623,
      "step": 3438
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5348,
      "step": 3439
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5212,
      "step": 3440
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5373,
      "step": 3441
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5413,
      "step": 3442
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.578,
      "step": 3443
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5579,
      "step": 3444
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5364,
      "step": 3445
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5629,
      "step": 3446
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5571,
      "step": 3447
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5594,
      "step": 3448
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.536,
      "step": 3449
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5614,
      "step": 3450
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.4722,
      "step": 3451
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.4768,
      "step": 3452
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0006,
      "loss": 3.5157,
      "step": 3453
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5186,
      "step": 3454
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5537,
      "step": 3455
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5695,
      "step": 3456
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5693,
      "step": 3457
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5119,
      "step": 3458
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5333,
      "step": 3459
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5334,
      "step": 3460
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5311,
      "step": 3461
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5593,
      "step": 3462
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5154,
      "step": 3463
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5194,
      "step": 3464
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5006,
      "step": 3465
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5069,
      "step": 3466
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.511,
      "step": 3467
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5296,
      "step": 3468
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5654,
      "step": 3469
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5279,
      "step": 3470
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5166,
      "step": 3471
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5455,
      "step": 3472
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0006,
      "loss": 3.5637,
      "step": 3473
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5914,
      "step": 3474
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5389,
      "step": 3475
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5211,
      "step": 3476
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5729,
      "step": 3477
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5678,
      "step": 3478
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5094,
      "step": 3479
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5097,
      "step": 3480
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5787,
      "step": 3481
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5461,
      "step": 3482
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5484,
      "step": 3483
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5351,
      "step": 3484
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5657,
      "step": 3485
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5949,
      "step": 3486
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5559,
      "step": 3487
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5184,
      "step": 3488
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5497,
      "step": 3489
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5695,
      "step": 3490
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.572,
      "step": 3491
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5884,
      "step": 3492
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0006,
      "loss": 3.5531,
      "step": 3493
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5518,
      "step": 3494
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5537,
      "step": 3495
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5707,
      "step": 3496
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.6049,
      "step": 3497
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5785,
      "step": 3498
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5293,
      "step": 3499
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5147,
      "step": 3500
    },
    {
      "epoch": 1.73,
      "eval_loss": 3.4488327503204346,
      "eval_perplexity": 32.96355438232422,
      "eval_runtime": 47.1345,
      "eval_samples_per_second": 2121.589,
      "eval_steps_per_second": 66.3,
      "step": 3500
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5485,
      "step": 3501
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5567,
      "step": 3502
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5458,
      "step": 3503
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5321,
      "step": 3504
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5383,
      "step": 3505
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5799,
      "step": 3506
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5265,
      "step": 3507
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5193,
      "step": 3508
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.539,
      "step": 3509
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5458,
      "step": 3510
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5439,
      "step": 3511
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.5258,
      "step": 3512
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0006,
      "loss": 3.529,
      "step": 3513
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5516,
      "step": 3514
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5462,
      "step": 3515
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5394,
      "step": 3516
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5382,
      "step": 3517
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5758,
      "step": 3518
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5172,
      "step": 3519
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.51,
      "step": 3520
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5713,
      "step": 3521
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5365,
      "step": 3522
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5472,
      "step": 3523
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.565,
      "step": 3524
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5744,
      "step": 3525
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5631,
      "step": 3526
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5329,
      "step": 3527
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5101,
      "step": 3528
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5658,
      "step": 3529
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5704,
      "step": 3530
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5583,
      "step": 3531
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5359,
      "step": 3532
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.5356,
      "step": 3533
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0006,
      "loss": 3.4795,
      "step": 3534
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.4674,
      "step": 3535
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.4882,
      "step": 3536
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.6081,
      "step": 3537
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.527,
      "step": 3538
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5273,
      "step": 3539
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5541,
      "step": 3540
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5616,
      "step": 3541
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.545,
      "step": 3542
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5081,
      "step": 3543
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.4744,
      "step": 3544
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5635,
      "step": 3545
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5627,
      "step": 3546
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5548,
      "step": 3547
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5625,
      "step": 3548
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5405,
      "step": 3549
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5721,
      "step": 3550
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5434,
      "step": 3551
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5135,
      "step": 3552
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.5776,
      "step": 3553
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0006,
      "loss": 3.515,
      "step": 3554
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.506,
      "step": 3555
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5338,
      "step": 3556
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5391,
      "step": 3557
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5265,
      "step": 3558
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.501,
      "step": 3559
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5016,
      "step": 3560
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5098,
      "step": 3561
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5215,
      "step": 3562
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5171,
      "step": 3563
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.4729,
      "step": 3564
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5085,
      "step": 3565
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5299,
      "step": 3566
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5442,
      "step": 3567
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5328,
      "step": 3568
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5865,
      "step": 3569
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.583,
      "step": 3570
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5731,
      "step": 3571
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5855,
      "step": 3572
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5132,
      "step": 3573
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0006,
      "loss": 3.5591,
      "step": 3574
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5673,
      "step": 3575
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5342,
      "step": 3576
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5436,
      "step": 3577
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.4906,
      "step": 3578
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5087,
      "step": 3579
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5318,
      "step": 3580
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5036,
      "step": 3581
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.4959,
      "step": 3582
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.486,
      "step": 3583
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.4626,
      "step": 3584
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5778,
      "step": 3585
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5623,
      "step": 3586
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5583,
      "step": 3587
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5311,
      "step": 3588
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5238,
      "step": 3589
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.4884,
      "step": 3590
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5006,
      "step": 3591
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5344,
      "step": 3592
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5104,
      "step": 3593
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0006,
      "loss": 3.5082,
      "step": 3594
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.5178,
      "step": 3595
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.5188,
      "step": 3596
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.4992,
      "step": 3597
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.5112,
      "step": 3598
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.531,
      "step": 3599
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.4495,
      "step": 3600
    },
    {
      "epoch": 1.78,
      "eval_loss": 3.436147451400757,
      "eval_perplexity": 32.56142044067383,
      "eval_runtime": 47.2417,
      "eval_samples_per_second": 2116.773,
      "eval_steps_per_second": 66.149,
      "step": 3600
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.5772,
      "step": 3601
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.5574,
      "step": 3602
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.5488,
      "step": 3603
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.5008,
      "step": 3604
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.5225,
      "step": 3605
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0006,
      "loss": 3.5175,
      "step": 3606
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.5,
      "step": 3607
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.5503,
      "step": 3608
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.5552,
      "step": 3609
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.5444,
      "step": 3610
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.5528,
      "step": 3611
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.6061,
      "step": 3612
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.5697,
      "step": 3613
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.5198,
      "step": 3614
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.4992,
      "step": 3615
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006,
      "loss": 3.5383,
      "step": 3616
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5319,
      "step": 3617
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5257,
      "step": 3618
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.514,
      "step": 3619
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5362,
      "step": 3620
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.597,
      "step": 3621
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5554,
      "step": 3622
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5244,
      "step": 3623
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5721,
      "step": 3624
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.581,
      "step": 3625
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5497,
      "step": 3626
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5233,
      "step": 3627
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5181,
      "step": 3628
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.528,
      "step": 3629
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5305,
      "step": 3630
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5434,
      "step": 3631
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5208,
      "step": 3632
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5315,
      "step": 3633
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5393,
      "step": 3634
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5474,
      "step": 3635
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5555,
      "step": 3636
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006,
      "loss": 3.5592,
      "step": 3637
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5275,
      "step": 3638
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5094,
      "step": 3639
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5038,
      "step": 3640
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5279,
      "step": 3641
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5307,
      "step": 3642
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5629,
      "step": 3643
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5795,
      "step": 3644
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5501,
      "step": 3645
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5504,
      "step": 3646
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5445,
      "step": 3647
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.6219,
      "step": 3648
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5913,
      "step": 3649
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5789,
      "step": 3650
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5756,
      "step": 3651
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.6186,
      "step": 3652
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.6368,
      "step": 3653
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.6085,
      "step": 3654
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5417,
      "step": 3655
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5344,
      "step": 3656
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006,
      "loss": 3.5475,
      "step": 3657
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5565,
      "step": 3658
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5614,
      "step": 3659
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5122,
      "step": 3660
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5403,
      "step": 3661
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5226,
      "step": 3662
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5305,
      "step": 3663
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5296,
      "step": 3664
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5948,
      "step": 3665
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5877,
      "step": 3666
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.578,
      "step": 3667
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5916,
      "step": 3668
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5856,
      "step": 3669
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5905,
      "step": 3670
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5911,
      "step": 3671
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5664,
      "step": 3672
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5814,
      "step": 3673
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5664,
      "step": 3674
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5399,
      "step": 3675
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.588,
      "step": 3676
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0006,
      "loss": 3.5366,
      "step": 3677
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5407,
      "step": 3678
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5569,
      "step": 3679
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5515,
      "step": 3680
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5634,
      "step": 3681
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5835,
      "step": 3682
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5814,
      "step": 3683
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5502,
      "step": 3684
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5674,
      "step": 3685
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5415,
      "step": 3686
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5496,
      "step": 3687
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5875,
      "step": 3688
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5628,
      "step": 3689
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5747,
      "step": 3690
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.587,
      "step": 3691
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5782,
      "step": 3692
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5712,
      "step": 3693
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5365,
      "step": 3694
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5496,
      "step": 3695
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5807,
      "step": 3696
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0006,
      "loss": 3.5734,
      "step": 3697
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5463,
      "step": 3698
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5645,
      "step": 3699
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5819,
      "step": 3700
    },
    {
      "epoch": 2.05,
      "eval_loss": 3.424142599105835,
      "eval_perplexity": 32.15679931640625,
      "eval_runtime": 47.4465,
      "eval_samples_per_second": 2107.635,
      "eval_steps_per_second": 65.864,
      "step": 3700
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.576,
      "step": 3701
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5448,
      "step": 3702
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5187,
      "step": 3703
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5503,
      "step": 3704
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5391,
      "step": 3705
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5341,
      "step": 3706
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5622,
      "step": 3707
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5319,
      "step": 3708
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5331,
      "step": 3709
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5134,
      "step": 3710
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5176,
      "step": 3711
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5408,
      "step": 3712
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5426,
      "step": 3713
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5537,
      "step": 3714
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.544,
      "step": 3715
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5535,
      "step": 3716
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5327,
      "step": 3717
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0006,
      "loss": 3.5206,
      "step": 3718
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5447,
      "step": 3719
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5659,
      "step": 3720
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5615,
      "step": 3721
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5883,
      "step": 3722
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.6107,
      "step": 3723
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5841,
      "step": 3724
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.6098,
      "step": 3725
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5832,
      "step": 3726
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5563,
      "step": 3727
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5813,
      "step": 3728
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5516,
      "step": 3729
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5697,
      "step": 3730
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5699,
      "step": 3731
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5323,
      "step": 3732
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5359,
      "step": 3733
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5425,
      "step": 3734
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5096,
      "step": 3735
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5626,
      "step": 3736
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.5504,
      "step": 3737
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0006,
      "loss": 3.547,
      "step": 3738
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5129,
      "step": 3739
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.4606,
      "step": 3740
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5125,
      "step": 3741
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5357,
      "step": 3742
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5549,
      "step": 3743
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5734,
      "step": 3744
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5786,
      "step": 3745
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5733,
      "step": 3746
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5077,
      "step": 3747
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5434,
      "step": 3748
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5118,
      "step": 3749
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.516,
      "step": 3750
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5283,
      "step": 3751
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5859,
      "step": 3752
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5461,
      "step": 3753
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5326,
      "step": 3754
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5317,
      "step": 3755
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.477,
      "step": 3756
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.4986,
      "step": 3757
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0006,
      "loss": 3.5196,
      "step": 3758
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.5237,
      "step": 3759
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.576,
      "step": 3760
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.5547,
      "step": 3761
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.5312,
      "step": 3762
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.5557,
      "step": 3763
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.568,
      "step": 3764
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.5355,
      "step": 3765
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.4839,
      "step": 3766
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.57,
      "step": 3767
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.5411,
      "step": 3768
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.5605,
      "step": 3769
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.6014,
      "step": 3770
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.5605,
      "step": 3771
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.53,
      "step": 3772
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.5066,
      "step": 3773
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.5075,
      "step": 3774
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.5421,
      "step": 3775
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.5768,
      "step": 3776
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.5463,
      "step": 3777
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0006,
      "loss": 3.5077,
      "step": 3778
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5404,
      "step": 3779
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5606,
      "step": 3780
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5515,
      "step": 3781
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5049,
      "step": 3782
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.4806,
      "step": 3783
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.594,
      "step": 3784
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5963,
      "step": 3785
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5478,
      "step": 3786
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5488,
      "step": 3787
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5343,
      "step": 3788
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5664,
      "step": 3789
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.6035,
      "step": 3790
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5607,
      "step": 3791
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5665,
      "step": 3792
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5322,
      "step": 3793
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5231,
      "step": 3794
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5587,
      "step": 3795
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5609,
      "step": 3796
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5611,
      "step": 3797
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5253,
      "step": 3798
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0006,
      "loss": 3.5094,
      "step": 3799
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5135,
      "step": 3800
    },
    {
      "epoch": 2.1,
      "eval_loss": 3.4110772609710693,
      "eval_perplexity": 31.747404098510742,
      "eval_runtime": 46.9559,
      "eval_samples_per_second": 2129.659,
      "eval_steps_per_second": 66.552,
      "step": 3800
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5414,
      "step": 3801
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5457,
      "step": 3802
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5683,
      "step": 3803
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5515,
      "step": 3804
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5579,
      "step": 3805
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5413,
      "step": 3806
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5579,
      "step": 3807
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5188,
      "step": 3808
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5294,
      "step": 3809
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5675,
      "step": 3810
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5396,
      "step": 3811
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5047,
      "step": 3812
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5167,
      "step": 3813
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5562,
      "step": 3814
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5644,
      "step": 3815
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.5082,
      "step": 3816
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.4948,
      "step": 3817
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.47,
      "step": 3818
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0006,
      "loss": 3.4731,
      "step": 3819
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5592,
      "step": 3820
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5711,
      "step": 3821
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5245,
      "step": 3822
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5251,
      "step": 3823
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5094,
      "step": 3824
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5349,
      "step": 3825
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5793,
      "step": 3826
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5739,
      "step": 3827
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5119,
      "step": 3828
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5141,
      "step": 3829
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5631,
      "step": 3830
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5444,
      "step": 3831
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5373,
      "step": 3832
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5325,
      "step": 3833
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.494,
      "step": 3834
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5106,
      "step": 3835
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.4894,
      "step": 3836
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5036,
      "step": 3837
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.5424,
      "step": 3838
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0006,
      "loss": 3.4929,
      "step": 3839
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5079,
      "step": 3840
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.4951,
      "step": 3841
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5594,
      "step": 3842
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5469,
      "step": 3843
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5554,
      "step": 3844
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5637,
      "step": 3845
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5477,
      "step": 3846
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5163,
      "step": 3847
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5096,
      "step": 3848
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5233,
      "step": 3849
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.4924,
      "step": 3850
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5331,
      "step": 3851
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5402,
      "step": 3852
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5448,
      "step": 3853
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5519,
      "step": 3854
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5588,
      "step": 3855
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5588,
      "step": 3856
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5667,
      "step": 3857
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5557,
      "step": 3858
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5439,
      "step": 3859
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0006,
      "loss": 3.5352,
      "step": 3860
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5571,
      "step": 3861
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5597,
      "step": 3862
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5052,
      "step": 3863
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5128,
      "step": 3864
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5185,
      "step": 3865
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5601,
      "step": 3866
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.517,
      "step": 3867
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5043,
      "step": 3868
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5225,
      "step": 3869
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5039,
      "step": 3870
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5588,
      "step": 3871
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5569,
      "step": 3872
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5181,
      "step": 3873
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5151,
      "step": 3874
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5015,
      "step": 3875
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5278,
      "step": 3876
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5837,
      "step": 3877
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5435,
      "step": 3878
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.5184,
      "step": 3879
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0006,
      "loss": 3.4828,
      "step": 3880
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.4724,
      "step": 3881
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.5392,
      "step": 3882
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.5602,
      "step": 3883
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.5582,
      "step": 3884
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.5341,
      "step": 3885
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.5295,
      "step": 3886
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.5448,
      "step": 3887
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.4878,
      "step": 3888
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.4913,
      "step": 3889
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.5918,
      "step": 3890
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.5275,
      "step": 3891
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.5152,
      "step": 3892
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.553,
      "step": 3893
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.498,
      "step": 3894
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.521,
      "step": 3895
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.5333,
      "step": 3896
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.5092,
      "step": 3897
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.5326,
      "step": 3898
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.5006,
      "step": 3899
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0006,
      "loss": 3.512,
      "step": 3900
    },
    {
      "epoch": 2.14,
      "eval_loss": 3.3959755897521973,
      "eval_perplexity": 31.270023345947266,
      "eval_runtime": 47.2721,
      "eval_samples_per_second": 2115.413,
      "eval_steps_per_second": 66.107,
      "step": 3900
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5673,
      "step": 3901
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5654,
      "step": 3902
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5501,
      "step": 3903
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.556,
      "step": 3904
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5048,
      "step": 3905
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5893,
      "step": 3906
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5116,
      "step": 3907
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5156,
      "step": 3908
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5511,
      "step": 3909
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.582,
      "step": 3910
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5559,
      "step": 3911
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5388,
      "step": 3912
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.565,
      "step": 3913
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5287,
      "step": 3914
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5656,
      "step": 3915
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5678,
      "step": 3916
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.523,
      "step": 3917
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5125,
      "step": 3918
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5155,
      "step": 3919
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0006,
      "loss": 3.5033,
      "step": 3920
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5044,
      "step": 3921
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5322,
      "step": 3922
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5123,
      "step": 3923
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5152,
      "step": 3924
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5283,
      "step": 3925
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5472,
      "step": 3926
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5167,
      "step": 3927
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5111,
      "step": 3928
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5444,
      "step": 3929
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5149,
      "step": 3930
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5128,
      "step": 3931
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5005,
      "step": 3932
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5536,
      "step": 3933
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5856,
      "step": 3934
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5469,
      "step": 3935
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.523,
      "step": 3936
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5222,
      "step": 3937
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5205,
      "step": 3938
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.4998,
      "step": 3939
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.504,
      "step": 3940
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0006,
      "loss": 3.5283,
      "step": 3941
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.5247,
      "step": 3942
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.5112,
      "step": 3943
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.5005,
      "step": 3944
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.4996,
      "step": 3945
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.5276,
      "step": 3946
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.5189,
      "step": 3947
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.5136,
      "step": 3948
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.5313,
      "step": 3949
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.5181,
      "step": 3950
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.5485,
      "step": 3951
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.5322,
      "step": 3952
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.4959,
      "step": 3953
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.4624,
      "step": 3954
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.4924,
      "step": 3955
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.5417,
      "step": 3956
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.508,
      "step": 3957
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.5121,
      "step": 3958
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.5173,
      "step": 3959
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.496,
      "step": 3960
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0006,
      "loss": 3.452,
      "step": 3961
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.4873,
      "step": 3962
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.5191,
      "step": 3963
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.5486,
      "step": 3964
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.5309,
      "step": 3965
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.5051,
      "step": 3966
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.4667,
      "step": 3967
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.4412,
      "step": 3968
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.5353,
      "step": 3969
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.5342,
      "step": 3970
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.5032,
      "step": 3971
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.4936,
      "step": 3972
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.5351,
      "step": 3973
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.536,
      "step": 3974
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.503,
      "step": 3975
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.5124,
      "step": 3976
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.4652,
      "step": 3977
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.5325,
      "step": 3978
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.553,
      "step": 3979
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.5138,
      "step": 3980
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0006,
      "loss": 3.4702,
      "step": 3981
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.5512,
      "step": 3982
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.5235,
      "step": 3983
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.4742,
      "step": 3984
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.5196,
      "step": 3985
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.4705,
      "step": 3986
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.4594,
      "step": 3987
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.4877,
      "step": 3988
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.5143,
      "step": 3989
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.5395,
      "step": 3990
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.5198,
      "step": 3991
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.5016,
      "step": 3992
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.4975,
      "step": 3993
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.5229,
      "step": 3994
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.5417,
      "step": 3995
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.5957,
      "step": 3996
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.5287,
      "step": 3997
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.5661,
      "step": 3998
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.5747,
      "step": 3999
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.5456,
      "step": 4000
    },
    {
      "epoch": 2.19,
      "eval_loss": 3.387756824493408,
      "eval_perplexity": 31.041810989379883,
      "eval_runtime": 46.8609,
      "eval_samples_per_second": 2133.976,
      "eval_steps_per_second": 66.687,
      "step": 4000
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0006,
      "loss": 3.525,
      "step": 4001
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.5464,
      "step": 4002
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.5261,
      "step": 4003
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.5006,
      "step": 4004
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.5226,
      "step": 4005
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.4878,
      "step": 4006
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.498,
      "step": 4007
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.522,
      "step": 4008
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.5156,
      "step": 4009
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.5259,
      "step": 4010
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.5367,
      "step": 4011
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.5229,
      "step": 4012
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.4908,
      "step": 4013
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.5228,
      "step": 4014
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.4803,
      "step": 4015
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.4727,
      "step": 4016
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.5452,
      "step": 4017
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.5297,
      "step": 4018
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.5102,
      "step": 4019
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.5422,
      "step": 4020
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.4953,
      "step": 4021
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0006,
      "loss": 3.5124,
      "step": 4022
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5011,
      "step": 4023
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.4827,
      "step": 4024
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5005,
      "step": 4025
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5164,
      "step": 4026
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5008,
      "step": 4027
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5291,
      "step": 4028
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5986,
      "step": 4029
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5159,
      "step": 4030
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.4969,
      "step": 4031
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.489,
      "step": 4032
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5138,
      "step": 4033
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5436,
      "step": 4034
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5369,
      "step": 4035
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5113,
      "step": 4036
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5175,
      "step": 4037
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5532,
      "step": 4038
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5563,
      "step": 4039
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5129,
      "step": 4040
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5305,
      "step": 4041
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0006,
      "loss": 3.5047,
      "step": 4042
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.5036,
      "step": 4043
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.506,
      "step": 4044
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.4927,
      "step": 4045
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.5051,
      "step": 4046
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.517,
      "step": 4047
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.4638,
      "step": 4048
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.5314,
      "step": 4049
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.4868,
      "step": 4050
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.4587,
      "step": 4051
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.5359,
      "step": 4052
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.5016,
      "step": 4053
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.4724,
      "step": 4054
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.4669,
      "step": 4055
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.5544,
      "step": 4056
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.5064,
      "step": 4057
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.5069,
      "step": 4058
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.5246,
      "step": 4059
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.4636,
      "step": 4060
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.4962,
      "step": 4061
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0006,
      "loss": 3.5319,
      "step": 4062
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.5335,
      "step": 4063
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.5033,
      "step": 4064
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.495,
      "step": 4065
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.5173,
      "step": 4066
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.5069,
      "step": 4067
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.5225,
      "step": 4068
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.5224,
      "step": 4069
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.4888,
      "step": 4070
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.4912,
      "step": 4071
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.509,
      "step": 4072
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.5371,
      "step": 4073
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.5172,
      "step": 4074
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.49,
      "step": 4075
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.5158,
      "step": 4076
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.4908,
      "step": 4077
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.4709,
      "step": 4078
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.4827,
      "step": 4079
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.5244,
      "step": 4080
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.4622,
      "step": 4081
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0006,
      "loss": 3.5124,
      "step": 4082
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.5692,
      "step": 4083
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.523,
      "step": 4084
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.5087,
      "step": 4085
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.5123,
      "step": 4086
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.4868,
      "step": 4087
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.5351,
      "step": 4088
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.5192,
      "step": 4089
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.5063,
      "step": 4090
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.4838,
      "step": 4091
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.558,
      "step": 4092
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.5112,
      "step": 4093
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.4951,
      "step": 4094
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.4713,
      "step": 4095
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.4799,
      "step": 4096
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.4689,
      "step": 4097
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.4949,
      "step": 4098
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.5616,
      "step": 4099
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.4541,
      "step": 4100
    },
    {
      "epoch": 2.24,
      "eval_loss": 3.373004198074341,
      "eval_perplexity": 30.587841033935547,
      "eval_runtime": 47.4604,
      "eval_samples_per_second": 2107.021,
      "eval_steps_per_second": 65.844,
      "step": 4100
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.479,
      "step": 4101
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.4848,
      "step": 4102
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0006,
      "loss": 3.503,
      "step": 4103
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.5117,
      "step": 4104
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.4843,
      "step": 4105
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.4665,
      "step": 4106
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.447,
      "step": 4107
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.5518,
      "step": 4108
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.5204,
      "step": 4109
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.4843,
      "step": 4110
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.4447,
      "step": 4111
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.5224,
      "step": 4112
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.5153,
      "step": 4113
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.5146,
      "step": 4114
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.4792,
      "step": 4115
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.4956,
      "step": 4116
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.4608,
      "step": 4117
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.4945,
      "step": 4118
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.5546,
      "step": 4119
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.4881,
      "step": 4120
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.5111,
      "step": 4121
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.5457,
      "step": 4122
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0006,
      "loss": 3.5522,
      "step": 4123
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.4993,
      "step": 4124
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.5019,
      "step": 4125
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.4876,
      "step": 4126
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.5502,
      "step": 4127
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.4792,
      "step": 4128
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.4752,
      "step": 4129
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.4632,
      "step": 4130
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.5241,
      "step": 4131
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.484,
      "step": 4132
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.491,
      "step": 4133
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.4946,
      "step": 4134
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.5018,
      "step": 4135
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.4971,
      "step": 4136
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.4678,
      "step": 4137
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.4878,
      "step": 4138
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.5284,
      "step": 4139
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.526,
      "step": 4140
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.5106,
      "step": 4141
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.4877,
      "step": 4142
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0006,
      "loss": 3.4833,
      "step": 4143
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.5193,
      "step": 4144
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.5326,
      "step": 4145
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.5173,
      "step": 4146
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.5045,
      "step": 4147
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.4801,
      "step": 4148
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.5136,
      "step": 4149
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.5505,
      "step": 4150
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.5247,
      "step": 4151
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.5122,
      "step": 4152
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.4875,
      "step": 4153
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.4197,
      "step": 4154
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.5336,
      "step": 4155
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.547,
      "step": 4156
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.5175,
      "step": 4157
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.4808,
      "step": 4158
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.5315,
      "step": 4159
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.5123,
      "step": 4160
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.5164,
      "step": 4161
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.5085,
      "step": 4162
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.0006,
      "loss": 3.498,
      "step": 4163
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4744,
      "step": 4164
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4609,
      "step": 4165
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4376,
      "step": 4166
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.5115,
      "step": 4167
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.5069,
      "step": 4168
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.5068,
      "step": 4169
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4997,
      "step": 4170
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4925,
      "step": 4171
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4992,
      "step": 4172
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4839,
      "step": 4173
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4733,
      "step": 4174
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4869,
      "step": 4175
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4899,
      "step": 4176
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4869,
      "step": 4177
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.5128,
      "step": 4178
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4866,
      "step": 4179
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4525,
      "step": 4180
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4641,
      "step": 4181
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4921,
      "step": 4182
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4643,
      "step": 4183
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0006,
      "loss": 3.4431,
      "step": 4184
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.4704,
      "step": 4185
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.5121,
      "step": 4186
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.5861,
      "step": 4187
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.5323,
      "step": 4188
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.5214,
      "step": 4189
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.5073,
      "step": 4190
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.4749,
      "step": 4191
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.4857,
      "step": 4192
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.4746,
      "step": 4193
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.4949,
      "step": 4194
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.4996,
      "step": 4195
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.483,
      "step": 4196
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.4767,
      "step": 4197
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.4542,
      "step": 4198
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.5298,
      "step": 4199
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.5251,
      "step": 4200
    },
    {
      "epoch": 2.29,
      "eval_loss": 3.3648183345794678,
      "eval_perplexity": 30.342681884765625,
      "eval_runtime": 47.1939,
      "eval_samples_per_second": 2118.916,
      "eval_steps_per_second": 66.216,
      "step": 4200
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.5115,
      "step": 4201
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.4686,
      "step": 4202
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.5206,
      "step": 4203
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0006,
      "loss": 3.495,
      "step": 4204
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.4984,
      "step": 4205
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.4694,
      "step": 4206
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.5006,
      "step": 4207
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.4844,
      "step": 4208
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.497,
      "step": 4209
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.4744,
      "step": 4210
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.5195,
      "step": 4211
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.5184,
      "step": 4212
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.5377,
      "step": 4213
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.4777,
      "step": 4214
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.4911,
      "step": 4215
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.4727,
      "step": 4216
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.4572,
      "step": 4217
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.491,
      "step": 4218
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.5055,
      "step": 4219
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.4742,
      "step": 4220
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.4462,
      "step": 4221
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.5077,
      "step": 4222
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.4989,
      "step": 4223
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0006,
      "loss": 3.5118,
      "step": 4224
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.5128,
      "step": 4225
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.495,
      "step": 4226
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.5015,
      "step": 4227
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.4897,
      "step": 4228
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.4766,
      "step": 4229
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.5012,
      "step": 4230
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.4994,
      "step": 4231
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.4892,
      "step": 4232
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.4982,
      "step": 4233
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.5142,
      "step": 4234
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.5056,
      "step": 4235
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.5014,
      "step": 4236
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.5171,
      "step": 4237
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.4396,
      "step": 4238
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.473,
      "step": 4239
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.4789,
      "step": 4240
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.4929,
      "step": 4241
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.515,
      "step": 4242
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.5266,
      "step": 4243
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0006,
      "loss": 3.5226,
      "step": 4244
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.4809,
      "step": 4245
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.404,
      "step": 4246
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.4459,
      "step": 4247
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.4592,
      "step": 4248
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.4699,
      "step": 4249
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.5055,
      "step": 4250
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.454,
      "step": 4251
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.4698,
      "step": 4252
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.4713,
      "step": 4253
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.46,
      "step": 4254
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.4878,
      "step": 4255
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.5103,
      "step": 4256
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.5416,
      "step": 4257
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.5422,
      "step": 4258
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.5173,
      "step": 4259
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.4876,
      "step": 4260
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.4349,
      "step": 4261
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.5395,
      "step": 4262
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.54,
      "step": 4263
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.5024,
      "step": 4264
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0006,
      "loss": 3.4698,
      "step": 4265
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.4754,
      "step": 4266
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.4777,
      "step": 4267
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.4709,
      "step": 4268
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.4536,
      "step": 4269
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.4654,
      "step": 4270
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.4934,
      "step": 4271
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.4712,
      "step": 4272
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.4821,
      "step": 4273
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.5448,
      "step": 4274
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.4613,
      "step": 4275
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.4387,
      "step": 4276
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.4755,
      "step": 4277
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.5158,
      "step": 4278
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.4729,
      "step": 4279
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.4744,
      "step": 4280
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.475,
      "step": 4281
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.4968,
      "step": 4282
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.478,
      "step": 4283
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.447,
      "step": 4284
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0006,
      "loss": 3.4348,
      "step": 4285
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.5137,
      "step": 4286
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.4794,
      "step": 4287
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.4489,
      "step": 4288
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.4489,
      "step": 4289
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.4925,
      "step": 4290
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.503,
      "step": 4291
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.5003,
      "step": 4292
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.4757,
      "step": 4293
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.5074,
      "step": 4294
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.4998,
      "step": 4295
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.4777,
      "step": 4296
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.4832,
      "step": 4297
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.4804,
      "step": 4298
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.4711,
      "step": 4299
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.4708,
      "step": 4300
    },
    {
      "epoch": 2.34,
      "eval_loss": 3.3548226356506348,
      "eval_perplexity": 30.058399200439453,
      "eval_runtime": 48.1571,
      "eval_samples_per_second": 2076.536,
      "eval_steps_per_second": 64.892,
      "step": 4300
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.5431,
      "step": 4301
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.5186,
      "step": 4302
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.4678,
      "step": 4303
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.4536,
      "step": 4304
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0006,
      "loss": 3.497,
      "step": 4305
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.5217,
      "step": 4306
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.5034,
      "step": 4307
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.4943,
      "step": 4308
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.457,
      "step": 4309
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.5004,
      "step": 4310
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.4634,
      "step": 4311
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.4542,
      "step": 4312
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.5406,
      "step": 4313
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.5178,
      "step": 4314
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.5059,
      "step": 4315
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.5269,
      "step": 4316
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.4948,
      "step": 4317
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.5039,
      "step": 4318
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.5076,
      "step": 4319
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.5367,
      "step": 4320
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.5595,
      "step": 4321
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.5086,
      "step": 4322
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.5225,
      "step": 4323
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.5302,
      "step": 4324
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0006,
      "loss": 3.4925,
      "step": 4325
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4614,
      "step": 4326
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4909,
      "step": 4327
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4961,
      "step": 4328
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4847,
      "step": 4329
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4732,
      "step": 4330
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4502,
      "step": 4331
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.463,
      "step": 4332
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4644,
      "step": 4333
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4808,
      "step": 4334
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.5136,
      "step": 4335
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.5328,
      "step": 4336
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4804,
      "step": 4337
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4807,
      "step": 4338
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4518,
      "step": 4339
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4413,
      "step": 4340
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4414,
      "step": 4341
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4895,
      "step": 4342
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4644,
      "step": 4343
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.475,
      "step": 4344
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4887,
      "step": 4345
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0006,
      "loss": 3.4219,
      "step": 4346
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.4166,
      "step": 4347
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.4254,
      "step": 4348
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.504,
      "step": 4349
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.5207,
      "step": 4350
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.5161,
      "step": 4351
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.436,
      "step": 4352
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.474,
      "step": 4353
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.4774,
      "step": 4354
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.4956,
      "step": 4355
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.5017,
      "step": 4356
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.4646,
      "step": 4357
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.4942,
      "step": 4358
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.4884,
      "step": 4359
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.4729,
      "step": 4360
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.4752,
      "step": 4361
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.4542,
      "step": 4362
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.4431,
      "step": 4363
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.4363,
      "step": 4364
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.5679,
      "step": 4365
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0006,
      "loss": 3.5038,
      "step": 4366
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.4828,
      "step": 4367
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.4453,
      "step": 4368
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.4831,
      "step": 4369
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.4772,
      "step": 4370
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.4828,
      "step": 4371
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.4836,
      "step": 4372
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.4574,
      "step": 4373
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.5031,
      "step": 4374
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.5332,
      "step": 4375
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.5322,
      "step": 4376
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.5064,
      "step": 4377
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.4305,
      "step": 4378
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.4294,
      "step": 4379
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.517,
      "step": 4380
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.499,
      "step": 4381
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.4838,
      "step": 4382
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.4784,
      "step": 4383
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.4689,
      "step": 4384
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.4774,
      "step": 4385
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0006,
      "loss": 3.4835,
      "step": 4386
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4876,
      "step": 4387
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.5013,
      "step": 4388
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4654,
      "step": 4389
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4349,
      "step": 4390
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4641,
      "step": 4391
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.5142,
      "step": 4392
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4676,
      "step": 4393
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4311,
      "step": 4394
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4178,
      "step": 4395
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4375,
      "step": 4396
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4574,
      "step": 4397
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4496,
      "step": 4398
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.441,
      "step": 4399
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4586,
      "step": 4400
    },
    {
      "epoch": 2.39,
      "eval_loss": 3.3473732471466064,
      "eval_perplexity": 29.815526962280273,
      "eval_runtime": 47.4905,
      "eval_samples_per_second": 2105.683,
      "eval_steps_per_second": 65.803,
      "step": 4400
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4925,
      "step": 4401
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4914,
      "step": 4402
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4714,
      "step": 4403
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4226,
      "step": 4404
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4799,
      "step": 4405
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0006,
      "loss": 3.4633,
      "step": 4406
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4628,
      "step": 4407
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4702,
      "step": 4408
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.462,
      "step": 4409
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.5006,
      "step": 4410
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4854,
      "step": 4411
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4158,
      "step": 4412
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4204,
      "step": 4413
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.439,
      "step": 4414
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4876,
      "step": 4415
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.5408,
      "step": 4416
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4901,
      "step": 4417
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4527,
      "step": 4418
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4409,
      "step": 4419
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4527,
      "step": 4420
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4377,
      "step": 4421
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4513,
      "step": 4422
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.5134,
      "step": 4423
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4477,
      "step": 4424
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4504,
      "step": 4425
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4962,
      "step": 4426
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0006,
      "loss": 3.4887,
      "step": 4427
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.4872,
      "step": 4428
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.5094,
      "step": 4429
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.4949,
      "step": 4430
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.4713,
      "step": 4431
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.4975,
      "step": 4432
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.4242,
      "step": 4433
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.4374,
      "step": 4434
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.4799,
      "step": 4435
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.5393,
      "step": 4436
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.5261,
      "step": 4437
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.5112,
      "step": 4438
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.4516,
      "step": 4439
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.5033,
      "step": 4440
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.4957,
      "step": 4441
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.4696,
      "step": 4442
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.4505,
      "step": 4443
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.473,
      "step": 4444
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.4882,
      "step": 4445
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.4982,
      "step": 4446
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0006,
      "loss": 3.4819,
      "step": 4447
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.4562,
      "step": 4448
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.4521,
      "step": 4449
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.4509,
      "step": 4450
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.4666,
      "step": 4451
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.5079,
      "step": 4452
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.497,
      "step": 4453
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.512,
      "step": 4454
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.5307,
      "step": 4455
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.4691,
      "step": 4456
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.4915,
      "step": 4457
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.5128,
      "step": 4458
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.5058,
      "step": 4459
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.5635,
      "step": 4460
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.4785,
      "step": 4461
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.435,
      "step": 4462
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.4284,
      "step": 4463
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.489,
      "step": 4464
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.4729,
      "step": 4465
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.4675,
      "step": 4466
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.0006,
      "loss": 3.4649,
      "step": 4467
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4584,
      "step": 4468
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4949,
      "step": 4469
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4735,
      "step": 4470
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4659,
      "step": 4471
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4717,
      "step": 4472
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4611,
      "step": 4473
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4747,
      "step": 4474
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.505,
      "step": 4475
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.472,
      "step": 4476
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4628,
      "step": 4477
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4646,
      "step": 4478
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.5035,
      "step": 4479
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4443,
      "step": 4480
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4602,
      "step": 4481
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4778,
      "step": 4482
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4731,
      "step": 4483
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4791,
      "step": 4484
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4492,
      "step": 4485
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4293,
      "step": 4486
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0006,
      "loss": 3.4447,
      "step": 4487
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.5204,
      "step": 4488
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4451,
      "step": 4489
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4423,
      "step": 4490
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4855,
      "step": 4491
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.5008,
      "step": 4492
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4774,
      "step": 4493
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4441,
      "step": 4494
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4484,
      "step": 4495
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4732,
      "step": 4496
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4462,
      "step": 4497
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4387,
      "step": 4498
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4566,
      "step": 4499
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.457,
      "step": 4500
    },
    {
      "epoch": 2.44,
      "eval_loss": 3.3354873657226562,
      "eval_perplexity": 29.464481353759766,
      "eval_runtime": 47.8213,
      "eval_samples_per_second": 2091.119,
      "eval_steps_per_second": 65.347,
      "step": 4500
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.432,
      "step": 4501
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4234,
      "step": 4502
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4447,
      "step": 4503
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.5012,
      "step": 4504
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4825,
      "step": 4505
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4387,
      "step": 4506
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4613,
      "step": 4507
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0006,
      "loss": 3.4729,
      "step": 4508
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4652,
      "step": 4509
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4277,
      "step": 4510
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.5018,
      "step": 4511
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4469,
      "step": 4512
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4376,
      "step": 4513
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4393,
      "step": 4514
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4876,
      "step": 4515
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4923,
      "step": 4516
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4819,
      "step": 4517
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.5078,
      "step": 4518
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4837,
      "step": 4519
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.5195,
      "step": 4520
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4699,
      "step": 4521
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4401,
      "step": 4522
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4803,
      "step": 4523
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.5027,
      "step": 4524
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4845,
      "step": 4525
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4637,
      "step": 4526
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4379,
      "step": 4527
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0006,
      "loss": 3.4744,
      "step": 4528
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.46,
      "step": 4529
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.4637,
      "step": 4530
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.4331,
      "step": 4531
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.4824,
      "step": 4532
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.4663,
      "step": 4533
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.428,
      "step": 4534
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.4416,
      "step": 4535
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.4709,
      "step": 4536
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.461,
      "step": 4537
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.4579,
      "step": 4538
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.4601,
      "step": 4539
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.487,
      "step": 4540
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.4662,
      "step": 4541
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.4374,
      "step": 4542
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.4723,
      "step": 4543
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.4665,
      "step": 4544
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.4484,
      "step": 4545
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.4375,
      "step": 4546
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.5027,
      "step": 4547
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0006,
      "loss": 3.4822,
      "step": 4548
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4697,
      "step": 4549
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4443,
      "step": 4550
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4484,
      "step": 4551
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4408,
      "step": 4552
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4665,
      "step": 4553
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4688,
      "step": 4554
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.3719,
      "step": 4555
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4469,
      "step": 4556
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4463,
      "step": 4557
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4525,
      "step": 4558
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.5295,
      "step": 4559
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4955,
      "step": 4560
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4806,
      "step": 4561
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4456,
      "step": 4562
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4433,
      "step": 4563
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4328,
      "step": 4564
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4399,
      "step": 4565
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4559,
      "step": 4566
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4409,
      "step": 4567
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4569,
      "step": 4568
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0006,
      "loss": 3.4761,
      "step": 4569
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.482,
      "step": 4570
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4869,
      "step": 4571
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4656,
      "step": 4572
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4421,
      "step": 4573
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4,
      "step": 4574
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4679,
      "step": 4575
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4622,
      "step": 4576
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4459,
      "step": 4577
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4441,
      "step": 4578
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4683,
      "step": 4579
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.495,
      "step": 4580
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4926,
      "step": 4581
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4704,
      "step": 4582
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4666,
      "step": 4583
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4763,
      "step": 4584
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4641,
      "step": 4585
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4409,
      "step": 4586
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.5043,
      "step": 4587
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4742,
      "step": 4588
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0006,
      "loss": 3.4585,
      "step": 4589
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.4421,
      "step": 4590
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.4401,
      "step": 4591
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.4452,
      "step": 4592
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.4196,
      "step": 4593
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.4032,
      "step": 4594
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.429,
      "step": 4595
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.4498,
      "step": 4596
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.4581,
      "step": 4597
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.4307,
      "step": 4598
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.422,
      "step": 4599
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.4526,
      "step": 4600
    },
    {
      "epoch": 2.49,
      "eval_loss": 3.3241026401519775,
      "eval_perplexity": 29.14130401611328,
      "eval_runtime": 151.342,
      "eval_samples_per_second": 660.755,
      "eval_steps_per_second": 20.649,
      "step": 4600
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.4415,
      "step": 4601
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.446,
      "step": 4602
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.4871,
      "step": 4603
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.4574,
      "step": 4604
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.4405,
      "step": 4605
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.4747,
      "step": 4606
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.4549,
      "step": 4607
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.3979,
      "step": 4608
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0006,
      "loss": 3.3951,
      "step": 4609
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4591,
      "step": 4610
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4824,
      "step": 4611
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4954,
      "step": 4612
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4938,
      "step": 4613
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4858,
      "step": 4614
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.5042,
      "step": 4615
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4558,
      "step": 4616
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4508,
      "step": 4617
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4822,
      "step": 4618
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4536,
      "step": 4619
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4211,
      "step": 4620
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4296,
      "step": 4621
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4199,
      "step": 4622
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4433,
      "step": 4623
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4425,
      "step": 4624
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4573,
      "step": 4625
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4751,
      "step": 4626
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4089,
      "step": 4627
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4427,
      "step": 4628
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0006,
      "loss": 3.4629,
      "step": 4629
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4421,
      "step": 4630
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4242,
      "step": 4631
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4052,
      "step": 4632
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4109,
      "step": 4633
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.5157,
      "step": 4634
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4805,
      "step": 4635
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4655,
      "step": 4636
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4935,
      "step": 4637
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4774,
      "step": 4638
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4389,
      "step": 4639
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4023,
      "step": 4640
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4306,
      "step": 4641
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4993,
      "step": 4642
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4406,
      "step": 4643
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4531,
      "step": 4644
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4836,
      "step": 4645
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4287,
      "step": 4646
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4789,
      "step": 4647
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4656,
      "step": 4648
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.4362,
      "step": 4649
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0006,
      "loss": 3.497,
      "step": 4650
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4352,
      "step": 4651
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4204,
      "step": 4652
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4367,
      "step": 4653
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4566,
      "step": 4654
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4335,
      "step": 4655
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4382,
      "step": 4656
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4083,
      "step": 4657
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4761,
      "step": 4658
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.464,
      "step": 4659
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4628,
      "step": 4660
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4471,
      "step": 4661
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4237,
      "step": 4662
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4491,
      "step": 4663
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4455,
      "step": 4664
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4164,
      "step": 4665
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.475,
      "step": 4666
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.5006,
      "step": 4667
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4685,
      "step": 4668
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.434,
      "step": 4669
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0006,
      "loss": 3.4647,
      "step": 4670
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.3907,
      "step": 4671
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4233,
      "step": 4672
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4924,
      "step": 4673
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4482,
      "step": 4674
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.416,
      "step": 4675
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4133,
      "step": 4676
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.5098,
      "step": 4677
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4762,
      "step": 4678
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4612,
      "step": 4679
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4464,
      "step": 4680
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4362,
      "step": 4681
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4433,
      "step": 4682
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4529,
      "step": 4683
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4619,
      "step": 4684
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4933,
      "step": 4685
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4695,
      "step": 4686
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4333,
      "step": 4687
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4504,
      "step": 4688
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.5304,
      "step": 4689
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0006,
      "loss": 3.4518,
      "step": 4690
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.4494,
      "step": 4691
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.4442,
      "step": 4692
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.3467,
      "step": 4693
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.4291,
      "step": 4694
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.4701,
      "step": 4695
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.475,
      "step": 4696
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.4079,
      "step": 4697
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.4298,
      "step": 4698
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.4409,
      "step": 4699
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.4551,
      "step": 4700
    },
    {
      "epoch": 2.54,
      "eval_loss": 3.3163704872131348,
      "eval_perplexity": 28.924863815307617,
      "eval_runtime": 47.7298,
      "eval_samples_per_second": 2095.129,
      "eval_steps_per_second": 65.473,
      "step": 4700
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.4704,
      "step": 4701
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.444,
      "step": 4702
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.4613,
      "step": 4703
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.4777,
      "step": 4704
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.462,
      "step": 4705
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.4413,
      "step": 4706
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.4352,
      "step": 4707
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.4499,
      "step": 4708
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.4167,
      "step": 4709
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.0006,
      "loss": 3.3839,
      "step": 4710
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.3876,
      "step": 4711
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4562,
      "step": 4712
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4261,
      "step": 4713
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4765,
      "step": 4714
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4668,
      "step": 4715
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.433,
      "step": 4716
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4282,
      "step": 4717
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4652,
      "step": 4718
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4446,
      "step": 4719
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.387,
      "step": 4720
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4702,
      "step": 4721
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4548,
      "step": 4722
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4893,
      "step": 4723
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4848,
      "step": 4724
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.5001,
      "step": 4725
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4619,
      "step": 4726
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4341,
      "step": 4727
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4707,
      "step": 4728
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4951,
      "step": 4729
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4536,
      "step": 4730
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0006,
      "loss": 3.4409,
      "step": 4731
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4755,
      "step": 4732
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4448,
      "step": 4733
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4658,
      "step": 4734
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4458,
      "step": 4735
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.393,
      "step": 4736
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4232,
      "step": 4737
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4792,
      "step": 4738
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4579,
      "step": 4739
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4109,
      "step": 4740
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4419,
      "step": 4741
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4142,
      "step": 4742
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4355,
      "step": 4743
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4637,
      "step": 4744
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4228,
      "step": 4745
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4581,
      "step": 4746
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4782,
      "step": 4747
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4753,
      "step": 4748
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4521,
      "step": 4749
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4721,
      "step": 4750
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0006,
      "loss": 3.4895,
      "step": 4751
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.5044,
      "step": 4752
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.4748,
      "step": 4753
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.4897,
      "step": 4754
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.4967,
      "step": 4755
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.4443,
      "step": 4756
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.448,
      "step": 4757
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.4443,
      "step": 4758
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.4659,
      "step": 4759
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.4597,
      "step": 4760
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.3956,
      "step": 4761
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.3943,
      "step": 4762
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.4132,
      "step": 4763
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.4736,
      "step": 4764
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.4284,
      "step": 4765
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.4441,
      "step": 4766
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.4538,
      "step": 4767
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.4699,
      "step": 4768
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.4934,
      "step": 4769
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.42,
      "step": 4770
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0006,
      "loss": 3.4314,
      "step": 4771
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4548,
      "step": 4772
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4532,
      "step": 4773
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.3889,
      "step": 4774
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.3915,
      "step": 4775
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4053,
      "step": 4776
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4673,
      "step": 4777
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4371,
      "step": 4778
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.424,
      "step": 4779
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.489,
      "step": 4780
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4389,
      "step": 4781
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4444,
      "step": 4782
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4831,
      "step": 4783
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4775,
      "step": 4784
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4002,
      "step": 4785
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4378,
      "step": 4786
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4498,
      "step": 4787
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4067,
      "step": 4788
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4278,
      "step": 4789
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4423,
      "step": 4790
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0006,
      "loss": 3.4527,
      "step": 4791
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4649,
      "step": 4792
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4551,
      "step": 4793
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4305,
      "step": 4794
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4358,
      "step": 4795
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4866,
      "step": 4796
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4563,
      "step": 4797
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4296,
      "step": 4798
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4476,
      "step": 4799
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4453,
      "step": 4800
    },
    {
      "epoch": 2.59,
      "eval_loss": 3.3122336864471436,
      "eval_perplexity": 28.803497314453125,
      "eval_runtime": 131.1156,
      "eval_samples_per_second": 762.686,
      "eval_steps_per_second": 23.834,
      "step": 4800
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4313,
      "step": 4801
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4254,
      "step": 4802
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4473,
      "step": 4803
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.3917,
      "step": 4804
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4415,
      "step": 4805
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4411,
      "step": 4806
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4331,
      "step": 4807
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4515,
      "step": 4808
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4396,
      "step": 4809
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4148,
      "step": 4810
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.3967,
      "step": 4811
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0006,
      "loss": 3.4429,
      "step": 4812
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.4474,
      "step": 4813
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.4545,
      "step": 4814
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.4762,
      "step": 4815
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.4172,
      "step": 4816
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.4034,
      "step": 4817
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.4059,
      "step": 4818
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.3948,
      "step": 4819
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.3996,
      "step": 4820
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.4196,
      "step": 4821
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.424,
      "step": 4822
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.4267,
      "step": 4823
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.3853,
      "step": 4824
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.4073,
      "step": 4825
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.4184,
      "step": 4826
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.4288,
      "step": 4827
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.4112,
      "step": 4828
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.3735,
      "step": 4829
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.3931,
      "step": 4830
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.4401,
      "step": 4831
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0006,
      "loss": 3.4782,
      "step": 4832
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.445,
      "step": 4833
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.4212,
      "step": 4834
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.4458,
      "step": 4835
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.4501,
      "step": 4836
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.471,
      "step": 4837
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.4353,
      "step": 4838
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.3418,
      "step": 4839
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.4438,
      "step": 4840
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.4002,
      "step": 4841
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.4109,
      "step": 4842
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.4621,
      "step": 4843
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.46,
      "step": 4844
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.4301,
      "step": 4845
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.3953,
      "step": 4846
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.3764,
      "step": 4847
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.4217,
      "step": 4848
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.4202,
      "step": 4849
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.3985,
      "step": 4850
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.4336,
      "step": 4851
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0006,
      "loss": 3.4063,
      "step": 4852
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.3756,
      "step": 4853
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.3923,
      "step": 4854
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.4585,
      "step": 4855
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.3892,
      "step": 4856
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.407,
      "step": 4857
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.4289,
      "step": 4858
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.4237,
      "step": 4859
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.4025,
      "step": 4860
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.4272,
      "step": 4861
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.4254,
      "step": 4862
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.4279,
      "step": 4863
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.4411,
      "step": 4864
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.4442,
      "step": 4865
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.4646,
      "step": 4866
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.3727,
      "step": 4867
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.3614,
      "step": 4868
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.3916,
      "step": 4869
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.4241,
      "step": 4870
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.4258,
      "step": 4871
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0006,
      "loss": 3.4363,
      "step": 4872
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.4443,
      "step": 4873
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.439,
      "step": 4874
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.3953,
      "step": 4875
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.4383,
      "step": 4876
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.4388,
      "step": 4877
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.4403,
      "step": 4878
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.4725,
      "step": 4879
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.4019,
      "step": 4880
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.3987,
      "step": 4881
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.4222,
      "step": 4882
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.3545,
      "step": 4883
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.371,
      "step": 4884
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.3998,
      "step": 4885
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.4399,
      "step": 4886
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.3848,
      "step": 4887
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.3921,
      "step": 4888
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.4376,
      "step": 4889
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.4958,
      "step": 4890
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.4033,
      "step": 4891
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.4084,
      "step": 4892
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0006,
      "loss": 3.4037,
      "step": 4893
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4254,
      "step": 4894
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4005,
      "step": 4895
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4466,
      "step": 4896
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.457,
      "step": 4897
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4105,
      "step": 4898
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.3923,
      "step": 4899
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4325,
      "step": 4900
    },
    {
      "epoch": 2.64,
      "eval_loss": 3.302237033843994,
      "eval_perplexity": 28.53045654296875,
      "eval_runtime": 47.669,
      "eval_samples_per_second": 2097.8,
      "eval_steps_per_second": 65.556,
      "step": 4900
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4151,
      "step": 4901
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4165,
      "step": 4902
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4244,
      "step": 4903
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4129,
      "step": 4904
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4262,
      "step": 4905
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4478,
      "step": 4906
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4737,
      "step": 4907
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4374,
      "step": 4908
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4234,
      "step": 4909
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4332,
      "step": 4910
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4294,
      "step": 4911
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4463,
      "step": 4912
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0006,
      "loss": 3.4464,
      "step": 4913
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.4509,
      "step": 4914
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.4325,
      "step": 4915
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.4359,
      "step": 4916
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.41,
      "step": 4917
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.4279,
      "step": 4918
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.4874,
      "step": 4919
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.4513,
      "step": 4920
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.4205,
      "step": 4921
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.4018,
      "step": 4922
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.4264,
      "step": 4923
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.4525,
      "step": 4924
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.4341,
      "step": 4925
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.3972,
      "step": 4926
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.4156,
      "step": 4927
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.431,
      "step": 4928
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.4301,
      "step": 4929
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.4186,
      "step": 4930
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.355,
      "step": 4931
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.3834,
      "step": 4932
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0006,
      "loss": 3.4234,
      "step": 4933
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.3837,
      "step": 4934
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4493,
      "step": 4935
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4432,
      "step": 4936
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4143,
      "step": 4937
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.411,
      "step": 4938
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4352,
      "step": 4939
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4018,
      "step": 4940
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.3725,
      "step": 4941
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4441,
      "step": 4942
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4694,
      "step": 4943
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4178,
      "step": 4944
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4065,
      "step": 4945
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4905,
      "step": 4946
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4698,
      "step": 4947
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4371,
      "step": 4948
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4274,
      "step": 4949
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4533,
      "step": 4950
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4078,
      "step": 4951
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.3939,
      "step": 4952
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.0006,
      "loss": 3.4119,
      "step": 4953
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.4629,
      "step": 4954
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.4633,
      "step": 4955
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.4386,
      "step": 4956
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.422,
      "step": 4957
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.4338,
      "step": 4958
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.45,
      "step": 4959
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.4189,
      "step": 4960
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.3927,
      "step": 4961
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.4906,
      "step": 4962
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.4534,
      "step": 4963
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.4268,
      "step": 4964
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.4301,
      "step": 4965
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.433,
      "step": 4966
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.3913,
      "step": 4967
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.3844,
      "step": 4968
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.4169,
      "step": 4969
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.3983,
      "step": 4970
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.3747,
      "step": 4971
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.3999,
      "step": 4972
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.3999,
      "step": 4973
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0006,
      "loss": 3.4064,
      "step": 4974
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.3959,
      "step": 4975
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.4144,
      "step": 4976
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.4351,
      "step": 4977
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.4102,
      "step": 4978
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.3617,
      "step": 4979
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.374,
      "step": 4980
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.3806,
      "step": 4981
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.3738,
      "step": 4982
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.4227,
      "step": 4983
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.4554,
      "step": 4984
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.4507,
      "step": 4985
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.4045,
      "step": 4986
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.3783,
      "step": 4987
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.4173,
      "step": 4988
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.4135,
      "step": 4989
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.4206,
      "step": 4990
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.4601,
      "step": 4991
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.478,
      "step": 4992
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.4621,
      "step": 4993
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0006,
      "loss": 3.4209,
      "step": 4994
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.4735,
      "step": 4995
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.4636,
      "step": 4996
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.4055,
      "step": 4997
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.4092,
      "step": 4998
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.3832,
      "step": 4999
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.3515,
      "step": 5000
    },
    {
      "epoch": 2.69,
      "eval_loss": 3.296806812286377,
      "eval_perplexity": 28.37639617919922,
      "eval_runtime": 46.9844,
      "eval_samples_per_second": 2128.366,
      "eval_steps_per_second": 66.511,
      "step": 5000
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.3998,
      "step": 5001
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.4008,
      "step": 5002
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.3995,
      "step": 5003
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.4062,
      "step": 5004
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.4611,
      "step": 5005
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.3748,
      "step": 5006
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.4176,
      "step": 5007
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.4642,
      "step": 5008
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.4147,
      "step": 5009
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.426,
      "step": 5010
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.41,
      "step": 5011
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.4111,
      "step": 5012
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.4209,
      "step": 5013
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.0006,
      "loss": 3.4617,
      "step": 5014
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.4576,
      "step": 5015
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.4739,
      "step": 5016
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.4927,
      "step": 5017
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.3995,
      "step": 5018
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.3989,
      "step": 5019
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.4215,
      "step": 5020
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.4067,
      "step": 5021
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.3963,
      "step": 5022
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.4056,
      "step": 5023
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.4317,
      "step": 5024
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.4478,
      "step": 5025
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.4195,
      "step": 5026
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.4146,
      "step": 5027
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.4246,
      "step": 5028
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.4361,
      "step": 5029
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.4237,
      "step": 5030
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.4258,
      "step": 5031
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.4026,
      "step": 5032
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.3304,
      "step": 5033
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0006,
      "loss": 3.375,
      "step": 5034
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.3823,
      "step": 5035
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.3829,
      "step": 5036
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.4291,
      "step": 5037
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.4446,
      "step": 5038
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.4208,
      "step": 5039
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.3901,
      "step": 5040
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.422,
      "step": 5041
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.3982,
      "step": 5042
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.4136,
      "step": 5043
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.4184,
      "step": 5044
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.4018,
      "step": 5045
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.371,
      "step": 5046
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.3612,
      "step": 5047
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.3617,
      "step": 5048
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.4069,
      "step": 5049
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.4183,
      "step": 5050
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.4138,
      "step": 5051
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.3896,
      "step": 5052
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.4001,
      "step": 5053
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.42,
      "step": 5054
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0006,
      "loss": 3.4413,
      "step": 5055
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.4506,
      "step": 5056
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.3827,
      "step": 5057
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.4201,
      "step": 5058
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.4366,
      "step": 5059
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.4355,
      "step": 5060
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.3619,
      "step": 5061
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.4216,
      "step": 5062
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.4226,
      "step": 5063
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.407,
      "step": 5064
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.4164,
      "step": 5065
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.4178,
      "step": 5066
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.4537,
      "step": 5067
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.4584,
      "step": 5068
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.4212,
      "step": 5069
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.3977,
      "step": 5070
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.4134,
      "step": 5071
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.4361,
      "step": 5072
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.4622,
      "step": 5073
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.447,
      "step": 5074
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0006,
      "loss": 3.4255,
      "step": 5075
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.418,
      "step": 5076
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.4304,
      "step": 5077
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.4562,
      "step": 5078
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.464,
      "step": 5079
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.4455,
      "step": 5080
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.3981,
      "step": 5081
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.3946,
      "step": 5082
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.4154,
      "step": 5083
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.4233,
      "step": 5084
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.4247,
      "step": 5085
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.4107,
      "step": 5086
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.4046,
      "step": 5087
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.4464,
      "step": 5088
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.4029,
      "step": 5089
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.39,
      "step": 5090
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.4125,
      "step": 5091
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.4135,
      "step": 5092
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.4208,
      "step": 5093
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.4151,
      "step": 5094
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0006,
      "loss": 3.3963,
      "step": 5095
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.4225,
      "step": 5096
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.4292,
      "step": 5097
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.4413,
      "step": 5098
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.4037,
      "step": 5099
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.4439,
      "step": 5100
    },
    {
      "epoch": 2.74,
      "eval_loss": 3.2911128997802734,
      "eval_perplexity": 28.225265502929688,
      "eval_runtime": 47.5761,
      "eval_samples_per_second": 2101.896,
      "eval_steps_per_second": 65.684,
      "step": 5100
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.3949,
      "step": 5101
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.3968,
      "step": 5102
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.4338,
      "step": 5103
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.4002,
      "step": 5104
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.4474,
      "step": 5105
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.4303,
      "step": 5106
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.4354,
      "step": 5107
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.4486,
      "step": 5108
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.3955,
      "step": 5109
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.3951,
      "step": 5110
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.4257,
      "step": 5111
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.4501,
      "step": 5112
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.4288,
      "step": 5113
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.4102,
      "step": 5114
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0006,
      "loss": 3.3958,
      "step": 5115
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.3466,
      "step": 5116
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.3675,
      "step": 5117
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.3747,
      "step": 5118
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.4778,
      "step": 5119
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.385,
      "step": 5120
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.4173,
      "step": 5121
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.435,
      "step": 5122
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.4229,
      "step": 5123
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.4134,
      "step": 5124
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.3716,
      "step": 5125
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.3813,
      "step": 5126
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.4351,
      "step": 5127
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.4407,
      "step": 5128
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.4345,
      "step": 5129
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.435,
      "step": 5130
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.4111,
      "step": 5131
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.4498,
      "step": 5132
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.4135,
      "step": 5133
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.4159,
      "step": 5134
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.4418,
      "step": 5135
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0006,
      "loss": 3.3715,
      "step": 5136
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.3951,
      "step": 5137
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.4173,
      "step": 5138
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.3986,
      "step": 5139
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.4046,
      "step": 5140
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.385,
      "step": 5141
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.376,
      "step": 5142
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.3898,
      "step": 5143
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.4035,
      "step": 5144
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.3734,
      "step": 5145
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.3599,
      "step": 5146
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.3821,
      "step": 5147
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.3982,
      "step": 5148
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.4134,
      "step": 5149
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.4263,
      "step": 5150
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.4458,
      "step": 5151
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.4666,
      "step": 5152
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.4613,
      "step": 5153
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.4283,
      "step": 5154
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.3809,
      "step": 5155
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0006,
      "loss": 3.4702,
      "step": 5156
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.4297,
      "step": 5157
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.3967,
      "step": 5158
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.4166,
      "step": 5159
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.3832,
      "step": 5160
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.3892,
      "step": 5161
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.4059,
      "step": 5162
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.3683,
      "step": 5163
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.3791,
      "step": 5164
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.3617,
      "step": 5165
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.3613,
      "step": 5166
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.4557,
      "step": 5167
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.4392,
      "step": 5168
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.4205,
      "step": 5169
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.4039,
      "step": 5170
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.3856,
      "step": 5171
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.3678,
      "step": 5172
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.381,
      "step": 5173
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.3986,
      "step": 5174
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.3947,
      "step": 5175
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0006,
      "loss": 3.4,
      "step": 5176
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0006,
      "loss": 3.3767,
      "step": 5177
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0006,
      "loss": 3.3912,
      "step": 5178
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0006,
      "loss": 3.345,
      "step": 5179
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0006,
      "loss": 3.4359,
      "step": 5180
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0006,
      "loss": 3.3909,
      "step": 5181
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0006,
      "loss": 3.3401,
      "step": 5182
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0006,
      "loss": 3.4638,
      "step": 5183
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0006,
      "loss": 3.4579,
      "step": 5184
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0006,
      "loss": 3.4024,
      "step": 5185
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0006,
      "loss": 3.3838,
      "step": 5186
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0006,
      "loss": 3.4022,
      "step": 5187
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0006,
      "loss": 3.3931,
      "step": 5188
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0006,
      "loss": 3.4,
      "step": 5189
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0006,
      "loss": 3.4139,
      "step": 5190
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0006,
      "loss": 3.4382,
      "step": 5191
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0006,
      "loss": 3.4187,
      "step": 5192
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0006,
      "loss": 3.445,
      "step": 5193
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0006,
      "loss": 3.475,
      "step": 5194
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0006,
      "loss": 3.4393,
      "step": 5195
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0006,
      "loss": 3.3979,
      "step": 5196
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0006,
      "loss": 3.3841,
      "step": 5197
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0006,
      "loss": 3.4086,
      "step": 5198
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.416,
      "step": 5199
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.3965,
      "step": 5200
    },
    {
      "epoch": 3.01,
      "eval_loss": 3.284156084060669,
      "eval_perplexity": 28.02582550048828,
      "eval_runtime": 47.2087,
      "eval_samples_per_second": 2118.256,
      "eval_steps_per_second": 66.195,
      "step": 5200
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4013,
      "step": 5201
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4182,
      "step": 5202
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4851,
      "step": 5203
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4129,
      "step": 5204
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4197,
      "step": 5205
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4501,
      "step": 5206
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4584,
      "step": 5207
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4124,
      "step": 5208
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4101,
      "step": 5209
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4004,
      "step": 5210
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4013,
      "step": 5211
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4309,
      "step": 5212
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4223,
      "step": 5213
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.3902,
      "step": 5214
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4108,
      "step": 5215
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4186,
      "step": 5216
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4316,
      "step": 5217
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0006,
      "loss": 3.4398,
      "step": 5218
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4342,
      "step": 5219
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4077,
      "step": 5220
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.3827,
      "step": 5221
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.3799,
      "step": 5222
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4135,
      "step": 5223
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4341,
      "step": 5224
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4391,
      "step": 5225
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.43,
      "step": 5226
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4045,
      "step": 5227
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.398,
      "step": 5228
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4214,
      "step": 5229
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4609,
      "step": 5230
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4505,
      "step": 5231
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4177,
      "step": 5232
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4549,
      "step": 5233
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4657,
      "step": 5234
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4999,
      "step": 5235
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4316,
      "step": 5236
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4014,
      "step": 5237
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.3814,
      "step": 5238
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0006,
      "loss": 3.4034,
      "step": 5239
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.4277,
      "step": 5240
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.4038,
      "step": 5241
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.3692,
      "step": 5242
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.3893,
      "step": 5243
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.3919,
      "step": 5244
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.3799,
      "step": 5245
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.3918,
      "step": 5246
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.4525,
      "step": 5247
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.43,
      "step": 5248
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.4485,
      "step": 5249
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.4304,
      "step": 5250
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.4395,
      "step": 5251
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.4514,
      "step": 5252
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.4321,
      "step": 5253
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.417,
      "step": 5254
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.4461,
      "step": 5255
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.3988,
      "step": 5256
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.4134,
      "step": 5257
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.4376,
      "step": 5258
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0006,
      "loss": 3.3817,
      "step": 5259
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.41,
      "step": 5260
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4091,
      "step": 5261
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4041,
      "step": 5262
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4317,
      "step": 5263
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4336,
      "step": 5264
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4381,
      "step": 5265
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.3901,
      "step": 5266
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4195,
      "step": 5267
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4114,
      "step": 5268
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4143,
      "step": 5269
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4367,
      "step": 5270
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4202,
      "step": 5271
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4509,
      "step": 5272
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4327,
      "step": 5273
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4202,
      "step": 5274
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4162,
      "step": 5275
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4157,
      "step": 5276
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4151,
      "step": 5277
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4266,
      "step": 5278
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0006,
      "loss": 3.4219,
      "step": 5279
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.4097,
      "step": 5280
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.421,
      "step": 5281
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.4405,
      "step": 5282
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.4385,
      "step": 5283
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.381,
      "step": 5284
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.3836,
      "step": 5285
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.4091,
      "step": 5286
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.3851,
      "step": 5287
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.4296,
      "step": 5288
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.3968,
      "step": 5289
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.3783,
      "step": 5290
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.396,
      "step": 5291
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.3795,
      "step": 5292
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.3805,
      "step": 5293
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.3982,
      "step": 5294
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.4182,
      "step": 5295
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.3973,
      "step": 5296
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.4009,
      "step": 5297
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.4031,
      "step": 5298
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0006,
      "loss": 3.4003,
      "step": 5299
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.3849,
      "step": 5300
    },
    {
      "epoch": 3.06,
      "eval_loss": 3.2769124507904053,
      "eval_perplexity": 27.812728881835938,
      "eval_runtime": 47.5004,
      "eval_samples_per_second": 2105.247,
      "eval_steps_per_second": 65.789,
      "step": 5300
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.3965,
      "step": 5301
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.4161,
      "step": 5302
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.4425,
      "step": 5303
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.4433,
      "step": 5304
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.4686,
      "step": 5305
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.4298,
      "step": 5306
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.4799,
      "step": 5307
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.4445,
      "step": 5308
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.4122,
      "step": 5309
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.4336,
      "step": 5310
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.4113,
      "step": 5311
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.4288,
      "step": 5312
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.4111,
      "step": 5313
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.3835,
      "step": 5314
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.4259,
      "step": 5315
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.381,
      "step": 5316
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.3751,
      "step": 5317
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.4149,
      "step": 5318
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.4166,
      "step": 5319
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0006,
      "loss": 3.3958,
      "step": 5320
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.3868,
      "step": 5321
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.319,
      "step": 5322
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.3891,
      "step": 5323
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.4058,
      "step": 5324
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.4241,
      "step": 5325
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.4157,
      "step": 5326
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.4703,
      "step": 5327
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.4117,
      "step": 5328
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.3688,
      "step": 5329
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.405,
      "step": 5330
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.3764,
      "step": 5331
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.3788,
      "step": 5332
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.4015,
      "step": 5333
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.4492,
      "step": 5334
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.4032,
      "step": 5335
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.3985,
      "step": 5336
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.3744,
      "step": 5337
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.3376,
      "step": 5338
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.3711,
      "step": 5339
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0006,
      "loss": 3.3792,
      "step": 5340
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.3926,
      "step": 5341
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.453,
      "step": 5342
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.4087,
      "step": 5343
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.397,
      "step": 5344
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.4115,
      "step": 5345
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.441,
      "step": 5346
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.3625,
      "step": 5347
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.3742,
      "step": 5348
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.4282,
      "step": 5349
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.4039,
      "step": 5350
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.4488,
      "step": 5351
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.4483,
      "step": 5352
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.4145,
      "step": 5353
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.3867,
      "step": 5354
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.3791,
      "step": 5355
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.3845,
      "step": 5356
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.395,
      "step": 5357
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.4458,
      "step": 5358
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.4013,
      "step": 5359
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0006,
      "loss": 3.3924,
      "step": 5360
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.3992,
      "step": 5361
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.4418,
      "step": 5362
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.4025,
      "step": 5363
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.3706,
      "step": 5364
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.3381,
      "step": 5365
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.5021,
      "step": 5366
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.4441,
      "step": 5367
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.407,
      "step": 5368
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.4101,
      "step": 5369
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.4098,
      "step": 5370
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.4546,
      "step": 5371
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.4417,
      "step": 5372
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.4042,
      "step": 5373
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.4348,
      "step": 5374
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.4043,
      "step": 5375
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.3789,
      "step": 5376
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.426,
      "step": 5377
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.4323,
      "step": 5378
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.4232,
      "step": 5379
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.3655,
      "step": 5380
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0006,
      "loss": 3.3775,
      "step": 5381
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.4055,
      "step": 5382
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.4141,
      "step": 5383
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.4166,
      "step": 5384
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.4379,
      "step": 5385
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.432,
      "step": 5386
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.4088,
      "step": 5387
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.4012,
      "step": 5388
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.4178,
      "step": 5389
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.3958,
      "step": 5390
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.4026,
      "step": 5391
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.4296,
      "step": 5392
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.4021,
      "step": 5393
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.3753,
      "step": 5394
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.3963,
      "step": 5395
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.4163,
      "step": 5396
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.4244,
      "step": 5397
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.3881,
      "step": 5398
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.3508,
      "step": 5399
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.3294,
      "step": 5400
    },
    {
      "epoch": 3.1,
      "eval_loss": 3.2752230167388916,
      "eval_perplexity": 27.781827926635742,
      "eval_runtime": 98.0218,
      "eval_samples_per_second": 1020.181,
      "eval_steps_per_second": 31.881,
      "step": 5400
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0006,
      "loss": 3.3484,
      "step": 5401
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.4458,
      "step": 5402
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.427,
      "step": 5403
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.3769,
      "step": 5404
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.379,
      "step": 5405
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.4001,
      "step": 5406
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.4144,
      "step": 5407
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.4354,
      "step": 5408
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.4349,
      "step": 5409
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.378,
      "step": 5410
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.3855,
      "step": 5411
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.4223,
      "step": 5412
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.4031,
      "step": 5413
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.4194,
      "step": 5414
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.3765,
      "step": 5415
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.3719,
      "step": 5416
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.3769,
      "step": 5417
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.3635,
      "step": 5418
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.3786,
      "step": 5419
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.4026,
      "step": 5420
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0006,
      "loss": 3.3706,
      "step": 5421
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.3679,
      "step": 5422
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.3829,
      "step": 5423
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.4273,
      "step": 5424
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.4126,
      "step": 5425
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.4333,
      "step": 5426
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.4255,
      "step": 5427
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.4068,
      "step": 5428
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.3762,
      "step": 5429
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.4058,
      "step": 5430
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.3739,
      "step": 5431
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.3643,
      "step": 5432
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.4142,
      "step": 5433
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.4214,
      "step": 5434
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.4148,
      "step": 5435
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.4086,
      "step": 5436
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.4329,
      "step": 5437
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.4427,
      "step": 5438
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.4395,
      "step": 5439
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.4094,
      "step": 5440
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0006,
      "loss": 3.4112,
      "step": 5441
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.4233,
      "step": 5442
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.4174,
      "step": 5443
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.4174,
      "step": 5444
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.3811,
      "step": 5445
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.396,
      "step": 5446
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.3915,
      "step": 5447
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.4276,
      "step": 5448
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.3777,
      "step": 5449
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.3906,
      "step": 5450
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.3883,
      "step": 5451
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.3632,
      "step": 5452
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.4571,
      "step": 5453
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.4196,
      "step": 5454
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.3829,
      "step": 5455
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.3791,
      "step": 5456
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.3819,
      "step": 5457
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.4277,
      "step": 5458
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.4421,
      "step": 5459
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.3982,
      "step": 5460
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.3988,
      "step": 5461
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0006,
      "loss": 3.3382,
      "step": 5462
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.3565,
      "step": 5463
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.4056,
      "step": 5464
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.4414,
      "step": 5465
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.414,
      "step": 5466
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.4076,
      "step": 5467
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.404,
      "step": 5468
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.4032,
      "step": 5469
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.3841,
      "step": 5470
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.3634,
      "step": 5471
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.4543,
      "step": 5472
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.3927,
      "step": 5473
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.3963,
      "step": 5474
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.4142,
      "step": 5475
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.358,
      "step": 5476
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.422,
      "step": 5477
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.4052,
      "step": 5478
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.3676,
      "step": 5479
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.3955,
      "step": 5480
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.3727,
      "step": 5481
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0006,
      "loss": 3.4121,
      "step": 5482
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.4285,
      "step": 5483
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.4176,
      "step": 5484
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.4442,
      "step": 5485
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.3977,
      "step": 5486
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.3863,
      "step": 5487
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.4482,
      "step": 5488
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.3864,
      "step": 5489
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.3943,
      "step": 5490
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.4233,
      "step": 5491
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.4478,
      "step": 5492
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.4302,
      "step": 5493
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.4308,
      "step": 5494
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.4311,
      "step": 5495
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.3913,
      "step": 5496
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.4656,
      "step": 5497
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.4228,
      "step": 5498
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.3979,
      "step": 5499
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.3771,
      "step": 5500
    },
    {
      "epoch": 3.15,
      "eval_loss": 3.2662177085876465,
      "eval_perplexity": 27.52883529663086,
      "eval_runtime": 47.3822,
      "eval_samples_per_second": 2110.499,
      "eval_steps_per_second": 65.953,
      "step": 5500
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.3902,
      "step": 5501
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0006,
      "loss": 3.3847,
      "step": 5502
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.3815,
      "step": 5503
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.4002,
      "step": 5504
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.3972,
      "step": 5505
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.3924,
      "step": 5506
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.4127,
      "step": 5507
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.432,
      "step": 5508
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.3944,
      "step": 5509
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.4004,
      "step": 5510
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.3897,
      "step": 5511
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.3854,
      "step": 5512
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.3899,
      "step": 5513
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.3864,
      "step": 5514
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.4341,
      "step": 5515
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.4453,
      "step": 5516
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.4152,
      "step": 5517
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.401,
      "step": 5518
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.3969,
      "step": 5519
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.3853,
      "step": 5520
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.3806,
      "step": 5521
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.0006,
      "loss": 3.3838,
      "step": 5522
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.4053,
      "step": 5523
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.3746,
      "step": 5524
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.4064,
      "step": 5525
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.384,
      "step": 5526
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.3706,
      "step": 5527
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.4067,
      "step": 5528
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.3903,
      "step": 5529
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.4056,
      "step": 5530
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.4079,
      "step": 5531
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.3953,
      "step": 5532
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.4313,
      "step": 5533
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.397,
      "step": 5534
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.357,
      "step": 5535
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.3296,
      "step": 5536
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.3858,
      "step": 5537
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.4113,
      "step": 5538
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.3751,
      "step": 5539
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.3934,
      "step": 5540
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.4037,
      "step": 5541
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.3608,
      "step": 5542
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0006,
      "loss": 3.3162,
      "step": 5543
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.3805,
      "step": 5544
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.4155,
      "step": 5545
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.4185,
      "step": 5546
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.3999,
      "step": 5547
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.385,
      "step": 5548
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.3266,
      "step": 5549
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.3467,
      "step": 5550
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.4087,
      "step": 5551
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.4137,
      "step": 5552
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.3811,
      "step": 5553
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.3671,
      "step": 5554
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.3998,
      "step": 5555
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.4078,
      "step": 5556
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.4022,
      "step": 5557
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.3762,
      "step": 5558
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.3354,
      "step": 5559
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.4479,
      "step": 5560
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.409,
      "step": 5561
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.3897,
      "step": 5562
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0006,
      "loss": 3.346,
      "step": 5563
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.4724,
      "step": 5564
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.3791,
      "step": 5565
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.3596,
      "step": 5566
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.3948,
      "step": 5567
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.3486,
      "step": 5568
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.3595,
      "step": 5569
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.362,
      "step": 5570
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.3779,
      "step": 5571
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.4239,
      "step": 5572
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.3893,
      "step": 5573
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.3793,
      "step": 5574
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.3661,
      "step": 5575
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.4195,
      "step": 5576
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.4376,
      "step": 5577
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.4598,
      "step": 5578
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.4057,
      "step": 5579
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.4652,
      "step": 5580
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.4382,
      "step": 5581
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.4164,
      "step": 5582
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0006,
      "loss": 3.4006,
      "step": 5583
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.4382,
      "step": 5584
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.3992,
      "step": 5585
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.3844,
      "step": 5586
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.3906,
      "step": 5587
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.371,
      "step": 5588
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.395,
      "step": 5589
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.4014,
      "step": 5590
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.3827,
      "step": 5591
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.416,
      "step": 5592
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.4219,
      "step": 5593
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.4073,
      "step": 5594
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.362,
      "step": 5595
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.4163,
      "step": 5596
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.3643,
      "step": 5597
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.3515,
      "step": 5598
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.4126,
      "step": 5599
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.4069,
      "step": 5600
    },
    {
      "epoch": 3.2,
      "eval_loss": 3.260463237762451,
      "eval_perplexity": 27.377153396606445,
      "eval_runtime": 46.686,
      "eval_samples_per_second": 2141.97,
      "eval_steps_per_second": 66.937,
      "step": 5600
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.407,
      "step": 5601
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.3971,
      "step": 5602
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0006,
      "loss": 3.3759,
      "step": 5603
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.4007,
      "step": 5604
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.3647,
      "step": 5605
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.3762,
      "step": 5606
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.3762,
      "step": 5607
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.3988,
      "step": 5608
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.392,
      "step": 5609
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.426,
      "step": 5610
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.4636,
      "step": 5611
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.3888,
      "step": 5612
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.3557,
      "step": 5613
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.3844,
      "step": 5614
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.3879,
      "step": 5615
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.4302,
      "step": 5616
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.4294,
      "step": 5617
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.3858,
      "step": 5618
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.4088,
      "step": 5619
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.446,
      "step": 5620
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.4367,
      "step": 5621
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.3843,
      "step": 5622
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.4126,
      "step": 5623
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0006,
      "loss": 3.3971,
      "step": 5624
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.3948,
      "step": 5625
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.367,
      "step": 5626
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.3813,
      "step": 5627
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.3992,
      "step": 5628
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.3675,
      "step": 5629
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.3544,
      "step": 5630
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.415,
      "step": 5631
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.3653,
      "step": 5632
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.3582,
      "step": 5633
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.4045,
      "step": 5634
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.382,
      "step": 5635
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.3496,
      "step": 5636
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.3646,
      "step": 5637
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.4241,
      "step": 5638
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.379,
      "step": 5639
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.4149,
      "step": 5640
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.3898,
      "step": 5641
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.335,
      "step": 5642
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.3987,
      "step": 5643
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0006,
      "loss": 3.4167,
      "step": 5644
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.4044,
      "step": 5645
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.3868,
      "step": 5646
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.3855,
      "step": 5647
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.408,
      "step": 5648
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.3923,
      "step": 5649
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.3963,
      "step": 5650
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.3988,
      "step": 5651
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.3817,
      "step": 5652
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.3787,
      "step": 5653
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.3816,
      "step": 5654
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.4268,
      "step": 5655
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.3949,
      "step": 5656
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.3802,
      "step": 5657
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.395,
      "step": 5658
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.3759,
      "step": 5659
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.3616,
      "step": 5660
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.3595,
      "step": 5661
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.4009,
      "step": 5662
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.3401,
      "step": 5663
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0006,
      "loss": 3.4214,
      "step": 5664
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.4394,
      "step": 5665
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.4049,
      "step": 5666
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.3949,
      "step": 5667
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.3752,
      "step": 5668
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.3792,
      "step": 5669
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.4009,
      "step": 5670
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.4136,
      "step": 5671
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.3846,
      "step": 5672
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.3809,
      "step": 5673
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.4275,
      "step": 5674
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.4031,
      "step": 5675
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.367,
      "step": 5676
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.3536,
      "step": 5677
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.3697,
      "step": 5678
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.3437,
      "step": 5679
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.405,
      "step": 5680
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.425,
      "step": 5681
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.3412,
      "step": 5682
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.3849,
      "step": 5683
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0006,
      "loss": 3.3723,
      "step": 5684
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.4054,
      "step": 5685
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.3796,
      "step": 5686
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.3947,
      "step": 5687
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.3421,
      "step": 5688
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.3518,
      "step": 5689
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.434,
      "step": 5690
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.3778,
      "step": 5691
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.3593,
      "step": 5692
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.3547,
      "step": 5693
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.396,
      "step": 5694
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.4117,
      "step": 5695
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.3888,
      "step": 5696
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.3557,
      "step": 5697
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.375,
      "step": 5698
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.3628,
      "step": 5699
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.3911,
      "step": 5700
    },
    {
      "epoch": 3.25,
      "eval_loss": 3.2595653533935547,
      "eval_perplexity": 27.36156463623047,
      "eval_runtime": 47.4026,
      "eval_samples_per_second": 2109.591,
      "eval_steps_per_second": 65.925,
      "step": 5700
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.4315,
      "step": 5701
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.3567,
      "step": 5702
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.4195,
      "step": 5703
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.4283,
      "step": 5704
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0006,
      "loss": 3.4293,
      "step": 5705
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.3877,
      "step": 5706
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.3744,
      "step": 5707
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.4046,
      "step": 5708
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.4082,
      "step": 5709
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.378,
      "step": 5710
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.3507,
      "step": 5711
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.3621,
      "step": 5712
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.4076,
      "step": 5713
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.3688,
      "step": 5714
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.3931,
      "step": 5715
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.3787,
      "step": 5716
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.3816,
      "step": 5717
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.387,
      "step": 5718
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.3632,
      "step": 5719
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.3799,
      "step": 5720
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.4067,
      "step": 5721
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.4207,
      "step": 5722
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.369,
      "step": 5723
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.3859,
      "step": 5724
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0006,
      "loss": 3.3655,
      "step": 5725
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.4173,
      "step": 5726
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.4207,
      "step": 5727
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.401,
      "step": 5728
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.3707,
      "step": 5729
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.3736,
      "step": 5730
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.4267,
      "step": 5731
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.4331,
      "step": 5732
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.407,
      "step": 5733
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.409,
      "step": 5734
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.3458,
      "step": 5735
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.3378,
      "step": 5736
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.4192,
      "step": 5737
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.4492,
      "step": 5738
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.3808,
      "step": 5739
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.388,
      "step": 5740
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.4062,
      "step": 5741
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.4055,
      "step": 5742
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.4031,
      "step": 5743
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.3959,
      "step": 5744
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0006,
      "loss": 3.3777,
      "step": 5745
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3638,
      "step": 5746
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3428,
      "step": 5747
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3428,
      "step": 5748
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3911,
      "step": 5749
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.406,
      "step": 5750
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3827,
      "step": 5751
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3795,
      "step": 5752
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3817,
      "step": 5753
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3832,
      "step": 5754
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3636,
      "step": 5755
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3658,
      "step": 5756
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3757,
      "step": 5757
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3739,
      "step": 5758
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3926,
      "step": 5759
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3967,
      "step": 5760
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3761,
      "step": 5761
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3424,
      "step": 5762
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.368,
      "step": 5763
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3808,
      "step": 5764
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0006,
      "loss": 3.3538,
      "step": 5765
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.3301,
      "step": 5766
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.3749,
      "step": 5767
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.4114,
      "step": 5768
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.4709,
      "step": 5769
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.4284,
      "step": 5770
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.3992,
      "step": 5771
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.4036,
      "step": 5772
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.3555,
      "step": 5773
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.3811,
      "step": 5774
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.3674,
      "step": 5775
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.3765,
      "step": 5776
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.3737,
      "step": 5777
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.3811,
      "step": 5778
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.3537,
      "step": 5779
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.3563,
      "step": 5780
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.4202,
      "step": 5781
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.4244,
      "step": 5782
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.3652,
      "step": 5783
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.3742,
      "step": 5784
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.4062,
      "step": 5785
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0006,
      "loss": 3.3874,
      "step": 5786
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.3733,
      "step": 5787
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.3619,
      "step": 5788
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.3766,
      "step": 5789
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.3872,
      "step": 5790
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.394,
      "step": 5791
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.3621,
      "step": 5792
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.413,
      "step": 5793
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.4358,
      "step": 5794
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.397,
      "step": 5795
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.3839,
      "step": 5796
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.3863,
      "step": 5797
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.3446,
      "step": 5798
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.3518,
      "step": 5799
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.3876,
      "step": 5800
    },
    {
      "epoch": 3.3,
      "eval_loss": 3.249647378921509,
      "eval_perplexity": 27.08446502685547,
      "eval_runtime": 47.1341,
      "eval_samples_per_second": 2121.608,
      "eval_steps_per_second": 66.3,
      "step": 5800
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.405,
      "step": 5801
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.3454,
      "step": 5802
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.3679,
      "step": 5803
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.3825,
      "step": 5804
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.3847,
      "step": 5805
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0006,
      "loss": 3.4097,
      "step": 5806
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.3974,
      "step": 5807
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.3858,
      "step": 5808
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.392,
      "step": 5809
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.3904,
      "step": 5810
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.3733,
      "step": 5811
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.3911,
      "step": 5812
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.3893,
      "step": 5813
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.3862,
      "step": 5814
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.3796,
      "step": 5815
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.4016,
      "step": 5816
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.3921,
      "step": 5817
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.4155,
      "step": 5818
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.3782,
      "step": 5819
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.3352,
      "step": 5820
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.3745,
      "step": 5821
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.3808,
      "step": 5822
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.3877,
      "step": 5823
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.4013,
      "step": 5824
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.4331,
      "step": 5825
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.0006,
      "loss": 3.4111,
      "step": 5826
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.3538,
      "step": 5827
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.2959,
      "step": 5828
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.3617,
      "step": 5829
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.3707,
      "step": 5830
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.3635,
      "step": 5831
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.3769,
      "step": 5832
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.3495,
      "step": 5833
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.3559,
      "step": 5834
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.3606,
      "step": 5835
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.3537,
      "step": 5836
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.4011,
      "step": 5837
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.4137,
      "step": 5838
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.4409,
      "step": 5839
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.4261,
      "step": 5840
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.4234,
      "step": 5841
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.3466,
      "step": 5842
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.3587,
      "step": 5843
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.4394,
      "step": 5844
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.4306,
      "step": 5845
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0006,
      "loss": 3.3859,
      "step": 5846
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3547,
      "step": 5847
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.37,
      "step": 5848
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3828,
      "step": 5849
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3649,
      "step": 5850
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3406,
      "step": 5851
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3602,
      "step": 5852
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.383,
      "step": 5853
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3821,
      "step": 5854
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3772,
      "step": 5855
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.4325,
      "step": 5856
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3236,
      "step": 5857
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3591,
      "step": 5858
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.377,
      "step": 5859
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3833,
      "step": 5860
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3628,
      "step": 5861
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3501,
      "step": 5862
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3877,
      "step": 5863
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3776,
      "step": 5864
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3647,
      "step": 5865
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3521,
      "step": 5866
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0006,
      "loss": 3.3362,
      "step": 5867
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.4057,
      "step": 5868
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.358,
      "step": 5869
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.3478,
      "step": 5870
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.3634,
      "step": 5871
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.3898,
      "step": 5872
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.4159,
      "step": 5873
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.3855,
      "step": 5874
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.3658,
      "step": 5875
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.4082,
      "step": 5876
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.3959,
      "step": 5877
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.372,
      "step": 5878
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.3812,
      "step": 5879
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.372,
      "step": 5880
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.3638,
      "step": 5881
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.3972,
      "step": 5882
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.4164,
      "step": 5883
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.4029,
      "step": 5884
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.34,
      "step": 5885
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.3551,
      "step": 5886
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0006,
      "loss": 3.3803,
      "step": 5887
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.4211,
      "step": 5888
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.4103,
      "step": 5889
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.3635,
      "step": 5890
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.3543,
      "step": 5891
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.401,
      "step": 5892
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.3517,
      "step": 5893
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.3712,
      "step": 5894
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.4402,
      "step": 5895
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.4026,
      "step": 5896
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.4246,
      "step": 5897
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.4094,
      "step": 5898
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.3986,
      "step": 5899
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.412,
      "step": 5900
    },
    {
      "epoch": 3.35,
      "eval_loss": 3.247203826904297,
      "eval_perplexity": 27.024383544921875,
      "eval_runtime": 47.2397,
      "eval_samples_per_second": 2116.865,
      "eval_steps_per_second": 66.152,
      "step": 5900
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.4128,
      "step": 5901
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.4324,
      "step": 5902
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.4312,
      "step": 5903
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.3967,
      "step": 5904
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.4332,
      "step": 5905
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.4107,
      "step": 5906
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0006,
      "loss": 3.3748,
      "step": 5907
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3558,
      "step": 5908
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3999,
      "step": 5909
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3904,
      "step": 5910
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.37,
      "step": 5911
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3671,
      "step": 5912
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3537,
      "step": 5913
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3571,
      "step": 5914
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3637,
      "step": 5915
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3889,
      "step": 5916
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.4166,
      "step": 5917
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.422,
      "step": 5918
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3714,
      "step": 5919
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3788,
      "step": 5920
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3507,
      "step": 5921
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3545,
      "step": 5922
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3292,
      "step": 5923
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3844,
      "step": 5924
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3791,
      "step": 5925
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3684,
      "step": 5926
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0006,
      "loss": 3.3751,
      "step": 5927
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3278,
      "step": 5928
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3104,
      "step": 5929
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.33,
      "step": 5930
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3962,
      "step": 5931
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.4348,
      "step": 5932
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3799,
      "step": 5933
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3546,
      "step": 5934
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3637,
      "step": 5935
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3853,
      "step": 5936
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3993,
      "step": 5937
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3884,
      "step": 5938
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3563,
      "step": 5939
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.4098,
      "step": 5940
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3811,
      "step": 5941
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3699,
      "step": 5942
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3772,
      "step": 5943
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3585,
      "step": 5944
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3397,
      "step": 5945
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3661,
      "step": 5946
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.4557,
      "step": 5947
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0006,
      "loss": 3.3979,
      "step": 5948
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.3641,
      "step": 5949
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.3629,
      "step": 5950
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.355,
      "step": 5951
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.4018,
      "step": 5952
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.3753,
      "step": 5953
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.3689,
      "step": 5954
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.3426,
      "step": 5955
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.4264,
      "step": 5956
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.4283,
      "step": 5957
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.4219,
      "step": 5958
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.3929,
      "step": 5959
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.3063,
      "step": 5960
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.3753,
      "step": 5961
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.3886,
      "step": 5962
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.3901,
      "step": 5963
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.3905,
      "step": 5964
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.3739,
      "step": 5965
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.3633,
      "step": 5966
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.3694,
      "step": 5967
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0006,
      "loss": 3.3912,
      "step": 5968
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.4007,
      "step": 5969
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3804,
      "step": 5970
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3575,
      "step": 5971
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3425,
      "step": 5972
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3722,
      "step": 5973
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.4048,
      "step": 5974
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3461,
      "step": 5975
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3398,
      "step": 5976
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3256,
      "step": 5977
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3238,
      "step": 5978
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3633,
      "step": 5979
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3346,
      "step": 5980
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3527,
      "step": 5981
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3509,
      "step": 5982
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3818,
      "step": 5983
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.4129,
      "step": 5984
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3555,
      "step": 5985
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3312,
      "step": 5986
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3825,
      "step": 5987
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0006,
      "loss": 3.3734,
      "step": 5988
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.3751,
      "step": 5989
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.358,
      "step": 5990
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.3781,
      "step": 5991
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.3996,
      "step": 5992
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.3782,
      "step": 5993
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.3106,
      "step": 5994
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.2996,
      "step": 5995
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.3737,
      "step": 5996
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.3866,
      "step": 5997
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.4372,
      "step": 5998
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.3861,
      "step": 5999
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.3521,
      "step": 6000
    },
    {
      "epoch": 3.4,
      "eval_loss": 3.242154836654663,
      "eval_perplexity": 26.887659072875977,
      "eval_runtime": 47.372,
      "eval_samples_per_second": 2110.954,
      "eval_steps_per_second": 65.967,
      "step": 6000
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.332,
      "step": 6001
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.3448,
      "step": 6002
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.341,
      "step": 6003
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.3773,
      "step": 6004
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.3853,
      "step": 6005
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.3396,
      "step": 6006
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.3875,
      "step": 6007
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0006,
      "loss": 3.3923,
      "step": 6008
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.3767,
      "step": 6009
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.3847,
      "step": 6010
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.4163,
      "step": 6011
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.3926,
      "step": 6012
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.3799,
      "step": 6013
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.3855,
      "step": 6014
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.3189,
      "step": 6015
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.3443,
      "step": 6016
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.4055,
      "step": 6017
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.4385,
      "step": 6018
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.4353,
      "step": 6019
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.3924,
      "step": 6020
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.3539,
      "step": 6021
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.4024,
      "step": 6022
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.3879,
      "step": 6023
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.3632,
      "step": 6024
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.3683,
      "step": 6025
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.365,
      "step": 6026
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.4093,
      "step": 6027
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.4084,
      "step": 6028
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0006,
      "loss": 3.3921,
      "step": 6029
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.3546,
      "step": 6030
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.3703,
      "step": 6031
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.3543,
      "step": 6032
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.38,
      "step": 6033
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.4007,
      "step": 6034
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.4029,
      "step": 6035
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.4113,
      "step": 6036
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.4128,
      "step": 6037
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.3566,
      "step": 6038
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.4194,
      "step": 6039
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.4164,
      "step": 6040
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.4016,
      "step": 6041
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.4472,
      "step": 6042
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.3763,
      "step": 6043
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.3294,
      "step": 6044
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.3385,
      "step": 6045
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.3752,
      "step": 6046
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.3849,
      "step": 6047
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.3594,
      "step": 6048
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0006,
      "loss": 3.3695,
      "step": 6049
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3482,
      "step": 6050
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3968,
      "step": 6051
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3931,
      "step": 6052
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3652,
      "step": 6053
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3681,
      "step": 6054
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3731,
      "step": 6055
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3793,
      "step": 6056
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.4017,
      "step": 6057
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3712,
      "step": 6058
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3585,
      "step": 6059
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3812,
      "step": 6060
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.389,
      "step": 6061
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3314,
      "step": 6062
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3801,
      "step": 6063
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3727,
      "step": 6064
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3798,
      "step": 6065
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3857,
      "step": 6066
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3383,
      "step": 6067
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3521,
      "step": 6068
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0006,
      "loss": 3.3445,
      "step": 6069
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.4323,
      "step": 6070
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3339,
      "step": 6071
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3595,
      "step": 6072
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3896,
      "step": 6073
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.4107,
      "step": 6074
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3823,
      "step": 6075
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.342,
      "step": 6076
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3393,
      "step": 6077
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3811,
      "step": 6078
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3504,
      "step": 6079
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3441,
      "step": 6080
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3511,
      "step": 6081
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3501,
      "step": 6082
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.336,
      "step": 6083
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3418,
      "step": 6084
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3402,
      "step": 6085
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.4246,
      "step": 6086
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3728,
      "step": 6087
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3463,
      "step": 6088
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3606,
      "step": 6089
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0006,
      "loss": 3.3956,
      "step": 6090
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.338,
      "step": 6091
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.3508,
      "step": 6092
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.3985,
      "step": 6093
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.3451,
      "step": 6094
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.3502,
      "step": 6095
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.3535,
      "step": 6096
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.3866,
      "step": 6097
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.4114,
      "step": 6098
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.391,
      "step": 6099
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.4029,
      "step": 6100
    },
    {
      "epoch": 3.45,
      "eval_loss": 3.236910343170166,
      "eval_perplexity": 26.72882652282715,
      "eval_runtime": 47.5142,
      "eval_samples_per_second": 2104.633,
      "eval_steps_per_second": 65.77,
      "step": 6100
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.3849,
      "step": 6101
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.4388,
      "step": 6102
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.3611,
      "step": 6103
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.3536,
      "step": 6104
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.3719,
      "step": 6105
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.4353,
      "step": 6106
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.3915,
      "step": 6107
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.3511,
      "step": 6108
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.3397,
      "step": 6109
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0006,
      "loss": 3.3927,
      "step": 6110
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.3703,
      "step": 6111
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.3518,
      "step": 6112
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.341,
      "step": 6113
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.4042,
      "step": 6114
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.3441,
      "step": 6115
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.3398,
      "step": 6116
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.3419,
      "step": 6117
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.3823,
      "step": 6118
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.3658,
      "step": 6119
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.347,
      "step": 6120
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.368,
      "step": 6121
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.4027,
      "step": 6122
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.3706,
      "step": 6123
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.3389,
      "step": 6124
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.3771,
      "step": 6125
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.3836,
      "step": 6126
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.3511,
      "step": 6127
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.3611,
      "step": 6128
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.3876,
      "step": 6129
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0006,
      "loss": 3.4155,
      "step": 6130
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3562,
      "step": 6131
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3454,
      "step": 6132
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3473,
      "step": 6133
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3623,
      "step": 6134
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3734,
      "step": 6135
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3464,
      "step": 6136
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.2828,
      "step": 6137
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3668,
      "step": 6138
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3547,
      "step": 6139
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3703,
      "step": 6140
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.4278,
      "step": 6141
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.4009,
      "step": 6142
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3612,
      "step": 6143
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3593,
      "step": 6144
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3288,
      "step": 6145
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3637,
      "step": 6146
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3507,
      "step": 6147
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3502,
      "step": 6148
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3454,
      "step": 6149
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0006,
      "loss": 3.3732,
      "step": 6150
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.3876,
      "step": 6151
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.3845,
      "step": 6152
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.3831,
      "step": 6153
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.3823,
      "step": 6154
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.3244,
      "step": 6155
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.3233,
      "step": 6156
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.3799,
      "step": 6157
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.37,
      "step": 6158
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.3484,
      "step": 6159
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.3643,
      "step": 6160
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.377,
      "step": 6161
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.4076,
      "step": 6162
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.3871,
      "step": 6163
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.3842,
      "step": 6164
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.3664,
      "step": 6165
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.394,
      "step": 6166
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.3552,
      "step": 6167
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.3571,
      "step": 6168
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.4137,
      "step": 6169
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.38,
      "step": 6170
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0006,
      "loss": 3.3406,
      "step": 6171
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3508,
      "step": 6172
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3545,
      "step": 6173
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3461,
      "step": 6174
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.312,
      "step": 6175
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3063,
      "step": 6176
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3481,
      "step": 6177
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3651,
      "step": 6178
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3516,
      "step": 6179
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3345,
      "step": 6180
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3446,
      "step": 6181
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3518,
      "step": 6182
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3634,
      "step": 6183
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3523,
      "step": 6184
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.4038,
      "step": 6185
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3745,
      "step": 6186
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3554,
      "step": 6187
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3709,
      "step": 6188
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.357,
      "step": 6189
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3064,
      "step": 6190
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0006,
      "loss": 3.3138,
      "step": 6191
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3492,
      "step": 6192
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.416,
      "step": 6193
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3978,
      "step": 6194
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3933,
      "step": 6195
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3947,
      "step": 6196
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.401,
      "step": 6197
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3637,
      "step": 6198
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3753,
      "step": 6199
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.38,
      "step": 6200
    },
    {
      "epoch": 3.5,
      "eval_loss": 3.234070301055908,
      "eval_perplexity": 26.67009735107422,
      "eval_runtime": 164.579,
      "eval_samples_per_second": 607.611,
      "eval_steps_per_second": 18.988,
      "step": 6200
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3513,
      "step": 6201
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3394,
      "step": 6202
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3337,
      "step": 6203
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.321,
      "step": 6204
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3622,
      "step": 6205
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3552,
      "step": 6206
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3638,
      "step": 6207
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3712,
      "step": 6208
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3336,
      "step": 6209
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3557,
      "step": 6210
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0006,
      "loss": 3.3656,
      "step": 6211
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3378,
      "step": 6212
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3489,
      "step": 6213
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3135,
      "step": 6214
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3425,
      "step": 6215
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.4187,
      "step": 6216
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3763,
      "step": 6217
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3836,
      "step": 6218
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3953,
      "step": 6219
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3685,
      "step": 6220
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3254,
      "step": 6221
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3264,
      "step": 6222
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3478,
      "step": 6223
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.394,
      "step": 6224
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3536,
      "step": 6225
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3681,
      "step": 6226
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3902,
      "step": 6227
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3335,
      "step": 6228
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.4005,
      "step": 6229
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3669,
      "step": 6230
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.0006,
      "loss": 3.3677,
      "step": 6231
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3909,
      "step": 6232
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3334,
      "step": 6233
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3463,
      "step": 6234
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3429,
      "step": 6235
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3632,
      "step": 6236
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3507,
      "step": 6237
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3307,
      "step": 6238
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3097,
      "step": 6239
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3828,
      "step": 6240
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3718,
      "step": 6241
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3639,
      "step": 6242
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.338,
      "step": 6243
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3363,
      "step": 6244
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3663,
      "step": 6245
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3414,
      "step": 6246
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3306,
      "step": 6247
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3954,
      "step": 6248
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.4131,
      "step": 6249
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3757,
      "step": 6250
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3332,
      "step": 6251
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0006,
      "loss": 3.3636,
      "step": 6252
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.323,
      "step": 6253
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.3475,
      "step": 6254
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.3917,
      "step": 6255
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.3504,
      "step": 6256
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.3293,
      "step": 6257
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.3574,
      "step": 6258
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.4007,
      "step": 6259
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.3714,
      "step": 6260
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.3726,
      "step": 6261
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.3544,
      "step": 6262
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.3325,
      "step": 6263
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.3575,
      "step": 6264
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.3781,
      "step": 6265
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.3717,
      "step": 6266
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.4019,
      "step": 6267
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.3703,
      "step": 6268
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.35,
      "step": 6269
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.3842,
      "step": 6270
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.4273,
      "step": 6271
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0006,
      "loss": 3.358,
      "step": 6272
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3712,
      "step": 6273
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3223,
      "step": 6274
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.2616,
      "step": 6275
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3821,
      "step": 6276
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3808,
      "step": 6277
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3669,
      "step": 6278
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3199,
      "step": 6279
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3624,
      "step": 6280
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3695,
      "step": 6281
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3514,
      "step": 6282
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3745,
      "step": 6283
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3658,
      "step": 6284
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3591,
      "step": 6285
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.4006,
      "step": 6286
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.373,
      "step": 6287
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3445,
      "step": 6288
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3498,
      "step": 6289
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3459,
      "step": 6290
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.3262,
      "step": 6291
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0006,
      "loss": 3.2732,
      "step": 6292
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.0006,
      "loss": 3.3329,
      "step": 6293
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.0006,
      "loss": 3.3596,
      "step": 6294
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.0006,
      "loss": 3.3301,
      "step": 6295
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.0006,
      "loss": 3.3994,
      "step": 6296
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.0006,
      "loss": 3.3596,
      "step": 6297
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.0006,
      "loss": 3.3599,
      "step": 6298
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.0006,
      "loss": 3.3293,
      "step": 6299
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.0006,
      "loss": 3.3989,
      "step": 6300
    },
    {
      "epoch": 3.55,
      "eval_loss": 3.2291667461395264,
      "eval_perplexity": 26.528430938720703,
      "eval_runtime": 47.6241,
      "eval_samples_per_second": 2099.779,
      "eval_steps_per_second": 65.618,
      "step": 6300
    }
  ],
  "logging_steps": 1,
  "max_steps": 16200,
  "num_train_epochs": 8,
  "save_steps": 100,
  "total_flos": 1.685622072833527e+19,
  "trial_name": null,
  "trial_params": null
}