{
  "best_metric": 0.5517624616622925,
  "best_model_checkpoint": "./output_v2/34bCodellama_CodeLlama-34B-hf_codellama_blob_1/checkpoint-1200",
  "epoch": 1.9656019656019657,
  "global_step": 1400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.5798,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.6599,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.6807,
      "step": 3
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.4016,
      "step": 4
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.2944,
      "step": 5
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.5477,
      "step": 6
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.1555,
      "step": 7
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8678,
      "step": 8
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.1943,
      "step": 9
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8046,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0481,
      "step": 11
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7954,
      "step": 12
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.3301,
      "step": 13
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9544,
      "step": 14
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8664,
      "step": 15
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.1004,
      "step": 16
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6603,
      "step": 17
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5278,
      "step": 18
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5971,
      "step": 19
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6106,
      "step": 20
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9927,
      "step": 21
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7429,
      "step": 22
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5403,
      "step": 23
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8739,
      "step": 24
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7539,
      "step": 25
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7598,
      "step": 26
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9449,
      "step": 27
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7586,
      "step": 28
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5068,
      "step": 29
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6072,
      "step": 30
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6695,
      "step": 31
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9185,
      "step": 32
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6406,
      "step": 33
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.2219,
      "step": 34
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8394,
      "step": 35
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6016,
      "step": 36
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.3244,
      "step": 37
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5451,
      "step": 38
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7036,
      "step": 39
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6548,
      "step": 40
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9478,
      "step": 41
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9163,
      "step": 42
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9001,
      "step": 43
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7748,
      "step": 44
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7414,
      "step": 45
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.4948,
      "step": 46
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7041,
      "step": 47
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6931,
      "step": 48
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7196,
      "step": 49
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6884,
      "step": 50
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5539,
      "step": 51
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6652,
      "step": 52
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9422,
      "step": 53
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.447,
      "step": 54
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7642,
      "step": 55
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.4434,
      "step": 56
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7617,
      "step": 57
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.5312,
      "step": 58
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7388,
      "step": 59
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9614,
      "step": 60
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 1.2599,
      "step": 61
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.595,
      "step": 62
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7218,
      "step": 63
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5194,
      "step": 64
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5519,
      "step": 65
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5712,
      "step": 66
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6011,
      "step": 67
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.6643,
      "step": 68
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7581,
      "step": 69
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 1.2109,
      "step": 70
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.8814,
      "step": 71
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.507,
      "step": 72
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7569,
      "step": 73
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7431,
      "step": 74
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.7837,
      "step": 75
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.5242,
      "step": 76
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.3062,
      "step": 77
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.5109,
      "step": 78
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.8998,
      "step": 79
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.875,
      "step": 80
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.7363,
      "step": 81
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.6465,
      "step": 82
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.8307,
      "step": 83
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.6284,
      "step": 84
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.5862,
      "step": 85
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.7502,
      "step": 86
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.6251,
      "step": 87
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.7613,
      "step": 88
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.6527,
      "step": 89
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002,
      "loss": 0.5793,
      "step": 90
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002,
      "loss": 0.459,
      "step": 91
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002,
      "loss": 0.7156,
      "step": 92
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002,
      "loss": 0.6078,
      "step": 93
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002,
      "loss": 0.2572,
      "step": 94
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002,
      "loss": 0.5447,
      "step": 95
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002,
      "loss": 0.6109,
      "step": 96
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002,
      "loss": 1.0691,
      "step": 97
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002,
      "loss": 0.3149,
      "step": 98
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002,
      "loss": 0.5433,
      "step": 99
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002,
      "loss": 0.7371,
      "step": 100
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002,
      "loss": 0.8157,
      "step": 101
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002,
      "loss": 0.5899,
      "step": 102
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002,
      "loss": 0.6743,
      "step": 103
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 0.9741,
      "step": 104
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 0.5697,
      "step": 105
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 0.6596,
      "step": 106
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 0.7502,
      "step": 107
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 0.6138,
      "step": 108
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 0.6065,
      "step": 109
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 0.6661,
      "step": 110
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002,
      "loss": 0.987,
      "step": 111
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002,
      "loss": 0.773,
      "step": 112
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002,
      "loss": 0.4769,
      "step": 113
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002,
      "loss": 0.8634,
      "step": 114
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002,
      "loss": 0.9495,
      "step": 115
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002,
      "loss": 0.6708,
      "step": 116
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002,
      "loss": 0.7206,
      "step": 117
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002,
      "loss": 0.628,
      "step": 118
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002,
      "loss": 0.7703,
      "step": 119
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002,
      "loss": 0.6506,
      "step": 120
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002,
      "loss": 0.8484,
      "step": 121
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002,
      "loss": 0.8516,
      "step": 122
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002,
      "loss": 0.4612,
      "step": 123
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002,
      "loss": 0.5995,
      "step": 124
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002,
      "loss": 0.6679,
      "step": 125
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002,
      "loss": 0.8091,
      "step": 126
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002,
      "loss": 0.6702,
      "step": 127
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002,
      "loss": 0.7588,
      "step": 128
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002,
      "loss": 0.5301,
      "step": 129
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002,
      "loss": 0.4867,
      "step": 130
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002,
      "loss": 1.0696,
      "step": 131
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002,
      "loss": 0.3769,
      "step": 132
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002,
      "loss": 0.558,
      "step": 133
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002,
      "loss": 0.7511,
      "step": 134
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002,
      "loss": 0.3991,
      "step": 135
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002,
      "loss": 0.6713,
      "step": 136
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002,
      "loss": 0.4538,
      "step": 137
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002,
      "loss": 0.5503,
      "step": 138
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002,
      "loss": 0.6534,
      "step": 139
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002,
      "loss": 0.4622,
      "step": 140
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002,
      "loss": 0.5731,
      "step": 141
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002,
      "loss": 0.7238,
      "step": 142
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002,
      "loss": 0.8143,
      "step": 143
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002,
      "loss": 0.539,
      "step": 144
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002,
      "loss": 0.447,
      "step": 145
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002,
      "loss": 0.4711,
      "step": 146
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002,
      "loss": 0.5584,
      "step": 147
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002,
      "loss": 0.5887,
      "step": 148
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002,
      "loss": 0.3808,
      "step": 149
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002,
      "loss": 0.5771,
      "step": 150
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002,
      "loss": 0.5276,
      "step": 151
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002,
      "loss": 0.4132,
      "step": 152
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002,
      "loss": 0.9056,
      "step": 153
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 0.5737,
      "step": 154
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 0.6992,
      "step": 155
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 0.4023,
      "step": 156
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 0.6238,
      "step": 157
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 0.5521,
      "step": 158
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 0.6641,
      "step": 159
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 0.4478,
      "step": 160
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002,
      "loss": 0.865,
      "step": 161
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002,
      "loss": 0.4332,
      "step": 162
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002,
      "loss": 0.8205,
      "step": 163
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002,
      "loss": 0.6539,
      "step": 164
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002,
      "loss": 0.6957,
      "step": 165
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002,
      "loss": 0.6279,
      "step": 166
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002,
      "loss": 0.6073,
      "step": 167
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002,
      "loss": 0.5831,
      "step": 168
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002,
      "loss": 0.5383,
      "step": 169
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002,
      "loss": 1.0065,
      "step": 170
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002,
      "loss": 0.5239,
      "step": 171
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002,
      "loss": 0.4246,
      "step": 172
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002,
      "loss": 0.633,
      "step": 173
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002,
      "loss": 0.4318,
      "step": 174
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002,
      "loss": 0.5228,
      "step": 175
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002,
      "loss": 0.791,
      "step": 176
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002,
      "loss": 0.6141,
      "step": 177
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002,
      "loss": 0.4894,
      "step": 178
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002,
      "loss": 1.0344,
      "step": 179
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002,
      "loss": 0.7332,
      "step": 180
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002,
      "loss": 0.6845,
      "step": 181
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002,
      "loss": 0.8027,
      "step": 182
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002,
      "loss": 0.6537,
      "step": 183
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002,
      "loss": 0.7333,
      "step": 184
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002,
      "loss": 0.6149,
      "step": 185
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002,
      "loss": 1.0586,
      "step": 186
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002,
      "loss": 0.6911,
      "step": 187
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002,
      "loss": 0.75,
      "step": 188
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002,
      "loss": 0.4598,
      "step": 189
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002,
      "loss": 0.5292,
      "step": 190
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002,
      "loss": 0.6164,
      "step": 191
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002,
      "loss": 0.3216,
      "step": 192
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002,
      "loss": 0.6229,
      "step": 193
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002,
      "loss": 0.6054,
      "step": 194
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002,
      "loss": 0.4821,
      "step": 195
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002,
      "loss": 0.8481,
      "step": 196
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002,
      "loss": 0.633,
      "step": 197
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002,
      "loss": 0.4537,
      "step": 198
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002,
      "loss": 0.6689,
      "step": 199
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002,
      "loss": 0.8891,
      "step": 200
    },
    {
      "epoch": 0.28,
      "eval_loss": 0.6127411127090454,
      "eval_runtime": 397.1016,
      "eval_samples_per_second": 2.518,
      "eval_steps_per_second": 1.259,
      "step": 200
    },
    {
      "epoch": 0.28,
      "mmlu_eval_accuracy": 0.49968036766977914,
      "mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
      "mmlu_eval_accuracy_anatomy": 0.5,
      "mmlu_eval_accuracy_astronomy": 0.375,
      "mmlu_eval_accuracy_business_ethics": 0.2727272727272727,
      "mmlu_eval_accuracy_clinical_knowledge": 0.4482758620689655,
      "mmlu_eval_accuracy_college_biology": 0.5,
      "mmlu_eval_accuracy_college_chemistry": 0.25,
      "mmlu_eval_accuracy_college_computer_science": 0.5454545454545454,
      "mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
      "mmlu_eval_accuracy_college_medicine": 0.5454545454545454,
      "mmlu_eval_accuracy_college_physics": 0.5454545454545454,
      "mmlu_eval_accuracy_computer_security": 0.45454545454545453,
      "mmlu_eval_accuracy_conceptual_physics": 0.2692307692307692,
      "mmlu_eval_accuracy_econometrics": 0.3333333333333333,
      "mmlu_eval_accuracy_electrical_engineering": 0.25,
      "mmlu_eval_accuracy_elementary_mathematics": 0.3902439024390244,
      "mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
      "mmlu_eval_accuracy_global_facts": 0.3,
      "mmlu_eval_accuracy_high_school_biology": 0.46875,
      "mmlu_eval_accuracy_high_school_chemistry": 0.2727272727272727,
      "mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
      "mmlu_eval_accuracy_high_school_european_history": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.3488372093023256,
      "mmlu_eval_accuracy_high_school_mathematics": 0.1724137931034483,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.5384615384615384,
      "mmlu_eval_accuracy_high_school_physics": 0.35294117647058826,
      "mmlu_eval_accuracy_high_school_psychology": 0.8333333333333334,
      "mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
      "mmlu_eval_accuracy_high_school_us_history": 0.8181818181818182,
      "mmlu_eval_accuracy_high_school_world_history": 0.6153846153846154,
      "mmlu_eval_accuracy_human_aging": 0.4782608695652174,
      "mmlu_eval_accuracy_human_sexuality": 0.5,
      "mmlu_eval_accuracy_international_law": 0.8461538461538461,
      "mmlu_eval_accuracy_jurisprudence": 0.5454545454545454,
      "mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
      "mmlu_eval_accuracy_machine_learning": 0.6363636363636364,
      "mmlu_eval_accuracy_management": 0.7272727272727273,
      "mmlu_eval_accuracy_marketing": 0.68,
      "mmlu_eval_accuracy_medical_genetics": 0.7272727272727273,
      "mmlu_eval_accuracy_miscellaneous": 0.6511627906976745,
      "mmlu_eval_accuracy_moral_disputes": 0.5,
      "mmlu_eval_accuracy_moral_scenarios": 0.21,
      "mmlu_eval_accuracy_nutrition": 0.696969696969697,
      "mmlu_eval_accuracy_philosophy": 0.5882352941176471,
      "mmlu_eval_accuracy_prehistory": 0.5428571428571428,
      "mmlu_eval_accuracy_professional_accounting": 0.3225806451612903,
      "mmlu_eval_accuracy_professional_law": 0.37058823529411766,
      "mmlu_eval_accuracy_professional_medicine": 0.4838709677419355,
      "mmlu_eval_accuracy_professional_psychology": 0.4492753623188406,
      "mmlu_eval_accuracy_public_relations": 0.4166666666666667,
      "mmlu_eval_accuracy_security_studies": 0.3333333333333333,
      "mmlu_eval_accuracy_sociology": 0.7272727272727273,
      "mmlu_eval_accuracy_us_foreign_policy": 0.5454545454545454,
      "mmlu_eval_accuracy_virology": 0.4444444444444444,
      "mmlu_eval_accuracy_world_religions": 0.7368421052631579,
      "mmlu_loss": 0.9665496518188607,
      "step": 200
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002,
      "loss": 0.8622,
      "step": 201
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002,
      "loss": 0.7277,
      "step": 202
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 0.7358,
      "step": 203
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 0.6171,
      "step": 204
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 0.7573,
      "step": 205
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 0.7947,
      "step": 206
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 0.438,
      "step": 207
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 0.4314,
      "step": 208
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 0.7751,
      "step": 209
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 0.7193,
      "step": 210
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002,
      "loss": 0.5671,
      "step": 211
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002,
      "loss": 0.6712,
      "step": 212
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002,
      "loss": 0.5761,
      "step": 213
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002,
      "loss": 0.6589,
      "step": 214
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002,
      "loss": 0.4178,
      "step": 215
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002,
      "loss": 0.4622,
      "step": 216
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002,
      "loss": 0.5189,
      "step": 217
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002,
      "loss": 0.8789,
      "step": 218
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002,
      "loss": 0.2529,
      "step": 219
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002,
      "loss": 0.5861,
      "step": 220
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002,
      "loss": 0.6578,
      "step": 221
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002,
      "loss": 0.5289,
      "step": 222
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002,
      "loss": 0.5529,
      "step": 223
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002,
      "loss": 0.6496,
      "step": 224
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002,
      "loss": 0.7043,
      "step": 225
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002,
      "loss": 0.5955,
      "step": 226
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002,
      "loss": 0.6942,
      "step": 227
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002,
      "loss": 0.3691,
      "step": 228
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002,
      "loss": 0.6348,
      "step": 229
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002,
      "loss": 0.5973,
      "step": 230
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002,
      "loss": 0.2706,
      "step": 231
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002,
      "loss": 0.6844,
      "step": 232
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002,
      "loss": 1.2182,
      "step": 233
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002,
      "loss": 0.6845,
      "step": 234
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002,
      "loss": 0.6271,
      "step": 235
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002,
      "loss": 0.7151,
      "step": 236
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002,
      "loss": 0.5568,
      "step": 237
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002,
      "loss": 0.9098,
      "step": 238
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002,
      "loss": 1.0819,
      "step": 239
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002,
      "loss": 0.5894,
      "step": 240
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002,
      "loss": 0.833,
      "step": 241
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002,
      "loss": 0.4099,
      "step": 242
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002,
      "loss": 0.8302,
      "step": 243
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002,
      "loss": 0.4284,
      "step": 244
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002,
      "loss": 0.5485,
      "step": 245
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002,
      "loss": 0.5654,
      "step": 246
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002,
      "loss": 0.5599,
      "step": 247
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002,
      "loss": 0.5756,
      "step": 248
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002,
      "loss": 0.4219,
      "step": 249
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002,
      "loss": 0.7499,
      "step": 250
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002,
      "loss": 0.6233,
      "step": 251
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002,
      "loss": 0.6872,
      "step": 252
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 0.8836,
      "step": 253
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 0.8242,
      "step": 254
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 0.9424,
      "step": 255
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 0.6088,
      "step": 256
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 0.6363,
      "step": 257
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 0.6582,
      "step": 258
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 1.0579,
      "step": 259
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002,
      "loss": 0.5946,
      "step": 260
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002,
      "loss": 0.9046,
      "step": 261
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002,
      "loss": 0.6693,
      "step": 262
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002,
      "loss": 0.6784,
      "step": 263
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002,
      "loss": 0.4804,
      "step": 264
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002,
      "loss": 0.5627,
      "step": 265
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002,
      "loss": 0.4451,
      "step": 266
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002,
      "loss": 0.3804,
      "step": 267
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002,
      "loss": 0.4614,
      "step": 268
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002,
      "loss": 0.4386,
      "step": 269
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002,
      "loss": 0.5054,
      "step": 270
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002,
      "loss": 0.3299,
      "step": 271
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002,
      "loss": 0.3767,
      "step": 272
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002,
      "loss": 0.6474,
      "step": 273
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002,
      "loss": 0.5278,
      "step": 274
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002,
      "loss": 0.6876,
      "step": 275
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002,
      "loss": 0.6743,
      "step": 276
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002,
      "loss": 0.6905,
      "step": 277
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002,
      "loss": 0.5167,
      "step": 278
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002,
      "loss": 0.8943,
      "step": 279
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002,
      "loss": 0.5534,
      "step": 280
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002,
      "loss": 0.6501,
      "step": 281
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002,
      "loss": 0.4966,
      "step": 282
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002,
      "loss": 0.5887,
      "step": 283
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002,
      "loss": 0.5823,
      "step": 284
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002,
      "loss": 0.4327,
      "step": 285
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002,
      "loss": 0.6986,
      "step": 286
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002,
      "loss": 0.3851,
      "step": 287
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002,
      "loss": 0.9356,
      "step": 288
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002,
      "loss": 0.4721,
      "step": 289
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002,
      "loss": 0.5978,
      "step": 290
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002,
      "loss": 0.5783,
      "step": 291
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002,
      "loss": 0.5125,
      "step": 292
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002,
      "loss": 0.5305,
      "step": 293
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002,
      "loss": 0.5606,
      "step": 294
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002,
      "loss": 1.4431,
      "step": 295
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002,
      "loss": 0.7331,
      "step": 296
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002,
      "loss": 0.6763,
      "step": 297
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002,
      "loss": 0.4489,
      "step": 298
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002,
      "loss": 0.5757,
      "step": 299
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002,
      "loss": 0.6899,
      "step": 300
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002,
      "loss": 0.5681,
      "step": 301
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002,
      "loss": 0.8638,
      "step": 302
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002,
      "loss": 0.5007,
      "step": 303
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002,
      "loss": 0.4922,
      "step": 304
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002,
      "loss": 0.6042,
      "step": 305
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002,
      "loss": 0.6398,
      "step": 306
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002,
      "loss": 0.6903,
      "step": 307
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002,
      "loss": 0.8877,
      "step": 308
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002,
      "loss": 0.7088,
      "step": 309
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 0.6013,
      "step": 310
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 0.8319,
      "step": 311
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 0.4064,
      "step": 312
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 0.6466,
      "step": 313
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 0.6873,
      "step": 314
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 0.9556,
      "step": 315
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 0.4119,
      "step": 316
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002,
      "loss": 0.5019,
      "step": 317
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002,
      "loss": 0.6108,
      "step": 318
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002,
      "loss": 0.5208,
      "step": 319
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002,
      "loss": 0.6555,
      "step": 320
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002,
      "loss": 0.3786,
      "step": 321
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002,
      "loss": 0.7625,
      "step": 322
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002,
      "loss": 0.6542,
      "step": 323
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002,
      "loss": 0.7737,
      "step": 324
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002,
      "loss": 0.6379,
      "step": 325
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002,
      "loss": 0.5464,
      "step": 326
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002,
      "loss": 0.6817,
      "step": 327
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002,
      "loss": 0.4032,
      "step": 328
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002,
      "loss": 1.1049,
      "step": 329
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002,
      "loss": 0.5744,
      "step": 330
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002,
      "loss": 0.3844,
      "step": 331
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002,
      "loss": 0.6441,
      "step": 332
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002,
      "loss": 0.3441,
      "step": 333
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002,
      "loss": 0.6981,
      "step": 334
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002,
      "loss": 0.437,
      "step": 335
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002,
      "loss": 0.9664,
      "step": 336
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002,
      "loss": 0.5144,
      "step": 337
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002,
      "loss": 0.6048,
      "step": 338
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002,
      "loss": 0.4976,
      "step": 339
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002,
      "loss": 0.5221,
      "step": 340
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002,
      "loss": 0.4715,
      "step": 341
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002,
      "loss": 0.7036,
      "step": 342
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002,
      "loss": 0.5815,
      "step": 343
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002,
      "loss": 0.4843,
      "step": 344
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002,
      "loss": 0.5816,
      "step": 345
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002,
      "loss": 0.2778,
      "step": 346
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002,
      "loss": 0.5073,
      "step": 347
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002,
      "loss": 0.3484,
      "step": 348
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002,
      "loss": 0.4671,
      "step": 349
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002,
      "loss": 0.5155,
      "step": 350
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002,
      "loss": 0.5333,
      "step": 351
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002,
      "loss": 0.7035,
      "step": 352
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0002,
      "loss": 0.6029,
      "step": 353
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0002,
      "loss": 0.9084,
      "step": 354
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0002,
      "loss": 0.6098,
      "step": 355
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0002,
      "loss": 0.3604,
      "step": 356
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0002,
      "loss": 0.5221,
      "step": 357
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0002,
      "loss": 0.4751,
      "step": 358
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0002,
      "loss": 0.9367,
      "step": 359
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 0.418,
      "step": 360
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 0.3642,
      "step": 361
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 0.5618,
      "step": 362
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 0.5638,
      "step": 363
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 0.9389,
      "step": 364
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 0.8087,
      "step": 365
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 0.5299,
      "step": 366
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0002,
      "loss": 0.8063,
      "step": 367
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0002,
      "loss": 0.4529,
      "step": 368
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0002,
      "loss": 0.6113,
      "step": 369
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0002,
      "loss": 0.813,
      "step": 370
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0002,
      "loss": 0.7885,
      "step": 371
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0002,
      "loss": 0.4745,
      "step": 372
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0002,
      "loss": 0.7722,
      "step": 373
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002,
      "loss": 0.7242,
      "step": 374
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002,
      "loss": 0.4531,
      "step": 375
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002,
      "loss": 0.9054,
      "step": 376
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002,
      "loss": 0.9152,
      "step": 377
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002,
      "loss": 0.4414,
      "step": 378
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002,
      "loss": 0.5136,
      "step": 379
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002,
      "loss": 0.3315,
      "step": 380
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002,
      "loss": 0.5251,
      "step": 381
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0002,
      "loss": 0.7332,
      "step": 382
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0002,
      "loss": 0.5032,
      "step": 383
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0002,
      "loss": 0.5535,
      "step": 384
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0002,
      "loss": 0.5916,
      "step": 385
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0002,
      "loss": 0.6259,
      "step": 386
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0002,
      "loss": 0.6234,
      "step": 387
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0002,
      "loss": 0.5661,
      "step": 388
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0002,
      "loss": 0.3717,
      "step": 389
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0002,
      "loss": 0.4362,
      "step": 390
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0002,
      "loss": 0.6232,
      "step": 391
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0002,
      "loss": 0.6514,
      "step": 392
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0002,
      "loss": 1.2381,
      "step": 393
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0002,
      "loss": 0.599,
      "step": 394
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0002,
      "loss": 0.5788,
      "step": 395
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0002,
      "loss": 0.4485,
      "step": 396
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0002,
      "loss": 0.5755,
      "step": 397
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0002,
      "loss": 0.7999,
      "step": 398
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0002,
      "loss": 0.4548,
      "step": 399
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0002,
      "loss": 0.5872,
      "step": 400
    },
    {
      "epoch": 0.56,
      "eval_loss": 0.5880547761917114,
      "eval_runtime": 396.9467,
      "eval_samples_per_second": 2.519,
      "eval_steps_per_second": 1.26,
      "step": 400
    },
    {
      "epoch": 0.56,
      "mmlu_eval_accuracy": 0.4846205369111515,
      "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
      "mmlu_eval_accuracy_anatomy": 0.35714285714285715,
      "mmlu_eval_accuracy_astronomy": 0.4375,
      "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
      "mmlu_eval_accuracy_clinical_knowledge": 0.41379310344827586,
      "mmlu_eval_accuracy_college_biology": 0.4375,
      "mmlu_eval_accuracy_college_chemistry": 0.125,
      "mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
      "mmlu_eval_accuracy_college_mathematics": 0.36363636363636365,
      "mmlu_eval_accuracy_college_medicine": 0.36363636363636365,
      "mmlu_eval_accuracy_college_physics": 0.2727272727272727,
      "mmlu_eval_accuracy_computer_security": 0.45454545454545453,
      "mmlu_eval_accuracy_conceptual_physics": 0.38461538461538464,
      "mmlu_eval_accuracy_econometrics": 0.3333333333333333,
      "mmlu_eval_accuracy_electrical_engineering": 0.5625,
      "mmlu_eval_accuracy_elementary_mathematics": 0.3170731707317073,
      "mmlu_eval_accuracy_formal_logic": 0.35714285714285715,
      "mmlu_eval_accuracy_global_facts": 0.2,
      "mmlu_eval_accuracy_high_school_biology": 0.28125,
      "mmlu_eval_accuracy_high_school_chemistry": 0.2727272727272727,
      "mmlu_eval_accuracy_high_school_computer_science": 0.7777777777777778,
      "mmlu_eval_accuracy_high_school_european_history": 0.6111111111111112,
      "mmlu_eval_accuracy_high_school_geography": 0.7727272727272727,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.32558139534883723,
      "mmlu_eval_accuracy_high_school_mathematics": 0.4482758620689655,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.34615384615384615,
      "mmlu_eval_accuracy_high_school_physics": 0.23529411764705882,
      "mmlu_eval_accuracy_high_school_psychology": 0.75,
      "mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
      "mmlu_eval_accuracy_high_school_us_history": 0.6818181818181818,
      "mmlu_eval_accuracy_high_school_world_history": 0.5769230769230769,
      "mmlu_eval_accuracy_human_aging": 0.6956521739130435,
      "mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
      "mmlu_eval_accuracy_international_law": 0.6923076923076923,
      "mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
      "mmlu_eval_accuracy_logical_fallacies": 0.7222222222222222,
      "mmlu_eval_accuracy_machine_learning": 0.45454545454545453,
      "mmlu_eval_accuracy_management": 0.7272727272727273,
      "mmlu_eval_accuracy_marketing": 0.76,
      "mmlu_eval_accuracy_medical_genetics": 0.8181818181818182,
      "mmlu_eval_accuracy_miscellaneous": 0.6395348837209303,
      "mmlu_eval_accuracy_moral_disputes": 0.39473684210526316,
      "mmlu_eval_accuracy_moral_scenarios": 0.25,
      "mmlu_eval_accuracy_nutrition": 0.5454545454545454,
      "mmlu_eval_accuracy_philosophy": 0.6176470588235294,
      "mmlu_eval_accuracy_prehistory": 0.45714285714285713,
      "mmlu_eval_accuracy_professional_accounting": 0.41935483870967744,
      "mmlu_eval_accuracy_professional_law": 0.25882352941176473,
      "mmlu_eval_accuracy_professional_medicine": 0.45161290322580644,
      "mmlu_eval_accuracy_professional_psychology": 0.43478260869565216,
      "mmlu_eval_accuracy_public_relations": 0.5,
      "mmlu_eval_accuracy_security_studies": 0.48148148148148145,
      "mmlu_eval_accuracy_sociology": 0.6818181818181818,
      "mmlu_eval_accuracy_us_foreign_policy": 0.7272727272727273,
      "mmlu_eval_accuracy_virology": 0.5,
      "mmlu_eval_accuracy_world_religions": 0.6842105263157895,
      "mmlu_loss": 0.9576842238246305,
      "step": 400
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0002,
      "loss": 0.3718,
      "step": 401
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0002,
      "loss": 0.5159,
      "step": 402
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002,
      "loss": 0.7165,
      "step": 403
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002,
      "loss": 0.6522,
      "step": 404
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002,
      "loss": 0.4325,
      "step": 405
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002,
      "loss": 0.5079,
      "step": 406
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002,
      "loss": 0.6084,
      "step": 407
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002,
      "loss": 0.8387,
      "step": 408
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002,
      "loss": 0.3619,
      "step": 409
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 0.4273,
      "step": 410
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 0.4658,
      "step": 411
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 0.4996,
      "step": 412
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 0.6387,
      "step": 413
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 0.7449,
      "step": 414
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 0.5566,
      "step": 415
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 0.6642,
      "step": 416
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002,
      "loss": 0.8422,
      "step": 417
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002,
      "loss": 0.9808,
      "step": 418
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002,
      "loss": 1.0236,
      "step": 419
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002,
      "loss": 0.8144,
      "step": 420
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002,
      "loss": 0.5159,
      "step": 421
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002,
      "loss": 0.5916,
      "step": 422
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002,
      "loss": 0.5595,
      "step": 423
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0002,
      "loss": 0.5962,
      "step": 424
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0002,
      "loss": 0.7472,
      "step": 425
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0002,
      "loss": 0.3524,
      "step": 426
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0002,
      "loss": 0.4166,
      "step": 427
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0002,
      "loss": 0.7061,
      "step": 428
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0002,
      "loss": 0.5893,
      "step": 429
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0002,
      "loss": 0.5269,
      "step": 430
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0002,
      "loss": 0.43,
      "step": 431
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0002,
      "loss": 0.6153,
      "step": 432
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0002,
      "loss": 0.6293,
      "step": 433
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0002,
      "loss": 0.4561,
      "step": 434
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0002,
      "loss": 0.9173,
      "step": 435
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0002,
      "loss": 0.2933,
      "step": 436
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0002,
      "loss": 0.7743,
      "step": 437
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0002,
      "loss": 0.8235,
      "step": 438
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0002,
      "loss": 0.5366,
      "step": 439
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0002,
      "loss": 0.4424,
      "step": 440
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0002,
      "loss": 0.7654,
      "step": 441
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0002,
      "loss": 0.6101,
      "step": 442
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0002,
      "loss": 0.4104,
      "step": 443
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0002,
      "loss": 0.7698,
      "step": 444
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0002,
      "loss": 0.521,
      "step": 445
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002,
      "loss": 0.6492,
      "step": 446
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002,
      "loss": 0.6252,
      "step": 447
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002,
      "loss": 0.6444,
      "step": 448
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002,
      "loss": 0.9372,
      "step": 449
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002,
      "loss": 0.5931,
      "step": 450
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002,
      "loss": 0.581,
      "step": 451
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002,
      "loss": 0.7696,
      "step": 452
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002,
      "loss": 0.6884,
      "step": 453
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002,
      "loss": 0.4991,
      "step": 454
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002,
      "loss": 0.3674,
      "step": 455
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002,
      "loss": 0.4275,
      "step": 456
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002,
      "loss": 0.6982,
      "step": 457
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002,
      "loss": 0.6223,
      "step": 458
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002,
      "loss": 0.5234,
      "step": 459
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002,
      "loss": 0.8479,
      "step": 460
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002,
      "loss": 0.5331,
      "step": 461
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002,
      "loss": 0.5743,
      "step": 462
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002,
      "loss": 0.6911,
      "step": 463
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002,
      "loss": 0.4317,
      "step": 464
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002,
      "loss": 0.4714,
      "step": 465
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002,
      "loss": 0.7314,
      "step": 466
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 0.6088,
      "step": 467
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 0.8494,
      "step": 468
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 0.7006,
      "step": 469
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 0.4734,
      "step": 470
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 0.6942,
      "step": 471
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 0.6707,
      "step": 472
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 0.4737,
      "step": 473
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002,
      "loss": 0.4141,
      "step": 474
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002,
      "loss": 0.4816,
      "step": 475
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002,
      "loss": 0.6768,
      "step": 476
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002,
      "loss": 0.4494,
      "step": 477
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002,
      "loss": 0.3933,
      "step": 478
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002,
      "loss": 0.3092,
      "step": 479
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002,
      "loss": 0.5429,
      "step": 480
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002,
      "loss": 0.5985,
      "step": 481
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002,
      "loss": 0.658,
      "step": 482
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002,
      "loss": 0.5346,
      "step": 483
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002,
      "loss": 0.2877,
      "step": 484
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002,
      "loss": 0.9522,
      "step": 485
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002,
      "loss": 0.2777,
      "step": 486
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002,
      "loss": 0.5888,
      "step": 487
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0002,
      "loss": 0.5125,
      "step": 488
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0002,
      "loss": 0.597,
      "step": 489
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0002,
      "loss": 0.6547,
      "step": 490
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0002,
      "loss": 0.4975,
      "step": 491
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0002,
      "loss": 0.556,
      "step": 492
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0002,
      "loss": 0.6408,
      "step": 493
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0002,
      "loss": 0.8527,
      "step": 494
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0002,
      "loss": 0.5432,
      "step": 495
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002,
      "loss": 0.6178,
      "step": 496
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002,
      "loss": 0.8967,
      "step": 497
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002,
      "loss": 0.5481,
      "step": 498
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002,
      "loss": 0.4747,
      "step": 499
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002,
      "loss": 0.522,
      "step": 500
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002,
      "loss": 0.2957,
      "step": 501
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002,
      "loss": 0.5323,
      "step": 502
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002,
      "loss": 0.3375,
      "step": 503
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002,
      "loss": 0.4292,
      "step": 504
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002,
      "loss": 0.375,
      "step": 505
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002,
      "loss": 0.4781,
      "step": 506
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002,
      "loss": 0.3998,
      "step": 507
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002,
      "loss": 0.4019,
      "step": 508
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002,
      "loss": 0.5622,
      "step": 509
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0002,
      "loss": 0.5633,
      "step": 510
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0002,
      "loss": 0.6693,
      "step": 511
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0002,
      "loss": 0.4402,
      "step": 512
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0002,
      "loss": 1.0258,
      "step": 513
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0002,
      "loss": 0.5866,
      "step": 514
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0002,
      "loss": 0.738,
      "step": 515
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0002,
      "loss": 0.4765,
      "step": 516
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 0.748,
      "step": 517
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 0.6294,
      "step": 518
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 0.7203,
      "step": 519
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 0.5033,
      "step": 520
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 0.7007,
      "step": 521
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 0.5197,
      "step": 522
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 0.8251,
      "step": 523
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0002,
      "loss": 0.6264,
      "step": 524
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0002,
      "loss": 0.422,
      "step": 525
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0002,
      "loss": 0.3288,
      "step": 526
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0002,
      "loss": 0.7641,
      "step": 527
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0002,
      "loss": 0.8521,
      "step": 528
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0002,
      "loss": 0.3265,
      "step": 529
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0002,
      "loss": 0.6372,
      "step": 530
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0002,
      "loss": 0.5141,
      "step": 531
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0002,
      "loss": 0.5788,
      "step": 532
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0002,
      "loss": 0.6268,
      "step": 533
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0002,
      "loss": 0.3942,
      "step": 534
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0002,
      "loss": 0.5028,
      "step": 535
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0002,
      "loss": 0.6894,
      "step": 536
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0002,
      "loss": 0.5062,
      "step": 537
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0002,
      "loss": 0.3966,
      "step": 538
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0002,
      "loss": 0.5734,
      "step": 539
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0002,
      "loss": 0.67,
      "step": 540
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0002,
      "loss": 0.6711,
      "step": 541
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0002,
      "loss": 0.8185,
      "step": 542
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0002,
      "loss": 0.6515,
      "step": 543
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0002,
      "loss": 0.4948,
      "step": 544
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0002,
      "loss": 0.5231,
      "step": 545
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0002,
      "loss": 0.5238,
      "step": 546
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0002,
      "loss": 0.4768,
      "step": 547
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0002,
      "loss": 0.5402,
      "step": 548
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0002,
      "loss": 0.4775,
      "step": 549
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0002,
      "loss": 0.7488,
      "step": 550
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0002,
      "loss": 0.4362,
      "step": 551
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0002,
      "loss": 0.4829,
      "step": 552
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0002,
      "loss": 0.5964,
      "step": 553
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0002,
      "loss": 0.6663,
      "step": 554
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0002,
      "loss": 0.7621,
      "step": 555
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0002,
      "loss": 0.4115,
      "step": 556
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0002,
      "loss": 0.6349,
      "step": 557
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0002,
      "loss": 0.6089,
      "step": 558
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0002,
      "loss": 0.4772,
      "step": 559
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0002,
      "loss": 0.5788,
      "step": 560
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0002,
      "loss": 0.3905,
      "step": 561
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0002,
      "loss": 0.509,
      "step": 562
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0002,
      "loss": 0.355,
      "step": 563
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0002,
      "loss": 0.7796,
      "step": 564
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0002,
      "loss": 0.6693,
      "step": 565
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0002,
      "loss": 0.6663,
      "step": 566
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 0.5275,
      "step": 567
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 0.8837,
      "step": 568
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 0.4286,
      "step": 569
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 1.0252,
      "step": 570
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 0.5022,
      "step": 571
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 0.6754,
      "step": 572
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 0.4,
      "step": 573
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0002,
      "loss": 0.7761,
      "step": 574
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0002,
      "loss": 0.6813,
      "step": 575
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0002,
      "loss": 0.5806,
      "step": 576
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0002,
      "loss": 0.5755,
      "step": 577
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0002,
      "loss": 0.7661,
      "step": 578
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0002,
      "loss": 0.8814,
      "step": 579
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0002,
      "loss": 0.5416,
      "step": 580
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002,
      "loss": 0.4392,
      "step": 581
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002,
      "loss": 0.5973,
      "step": 582
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002,
      "loss": 0.4373,
      "step": 583
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002,
      "loss": 0.5291,
      "step": 584
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002,
      "loss": 0.485,
      "step": 585
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002,
      "loss": 0.353,
      "step": 586
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002,
      "loss": 0.9446,
      "step": 587
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0002,
      "loss": 0.4602,
      "step": 588
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0002,
      "loss": 0.7173,
      "step": 589
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0002,
      "loss": 0.4454,
      "step": 590
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0002,
      "loss": 0.4104,
      "step": 591
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0002,
      "loss": 0.3425,
      "step": 592
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0002,
      "loss": 0.4651,
      "step": 593
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0002,
      "loss": 0.6563,
      "step": 594
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002,
      "loss": 0.5574,
      "step": 595
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002,
      "loss": 0.4641,
      "step": 596
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002,
      "loss": 0.576,
      "step": 597
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002,
      "loss": 0.4755,
      "step": 598
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002,
      "loss": 0.7452,
      "step": 599
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002,
      "loss": 0.6964,
      "step": 600
    },
    {
      "epoch": 0.84,
      "eval_loss": 0.5618544220924377,
      "eval_runtime": 397.2406,
      "eval_samples_per_second": 2.517,
      "eval_steps_per_second": 1.259,
      "step": 600
    },
    {
      "epoch": 0.84,
      "mmlu_eval_accuracy": 0.5192617436513831,
      "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
      "mmlu_eval_accuracy_anatomy": 0.42857142857142855,
      "mmlu_eval_accuracy_astronomy": 0.5,
      "mmlu_eval_accuracy_business_ethics": 0.45454545454545453,
      "mmlu_eval_accuracy_clinical_knowledge": 0.41379310344827586,
      "mmlu_eval_accuracy_college_biology": 0.5,
      "mmlu_eval_accuracy_college_chemistry": 0.125,
      "mmlu_eval_accuracy_college_computer_science": 0.7272727272727273,
      "mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
      "mmlu_eval_accuracy_college_medicine": 0.5,
      "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
      "mmlu_eval_accuracy_computer_security": 0.5454545454545454,
      "mmlu_eval_accuracy_conceptual_physics": 0.34615384615384615,
      "mmlu_eval_accuracy_econometrics": 0.25,
      "mmlu_eval_accuracy_electrical_engineering": 0.4375,
      "mmlu_eval_accuracy_elementary_mathematics": 0.4146341463414634,
      "mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
      "mmlu_eval_accuracy_global_facts": 0.2,
      "mmlu_eval_accuracy_high_school_biology": 0.375,
      "mmlu_eval_accuracy_high_school_chemistry": 0.3181818181818182,
      "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_european_history": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.8095238095238095,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.4186046511627907,
      "mmlu_eval_accuracy_high_school_mathematics": 0.41379310344827586,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.5,
      "mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
      "mmlu_eval_accuracy_high_school_psychology": 0.7666666666666667,
      "mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
      "mmlu_eval_accuracy_high_school_us_history": 0.6818181818181818,
      "mmlu_eval_accuracy_high_school_world_history": 0.6153846153846154,
      "mmlu_eval_accuracy_human_aging": 0.6086956521739131,
      "mmlu_eval_accuracy_human_sexuality": 0.5,
      "mmlu_eval_accuracy_international_law": 0.6923076923076923,
      "mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
      "mmlu_eval_accuracy_logical_fallacies": 0.7777777777777778,
      "mmlu_eval_accuracy_machine_learning": 0.6363636363636364,
      "mmlu_eval_accuracy_management": 0.9090909090909091,
      "mmlu_eval_accuracy_marketing": 0.72,
      "mmlu_eval_accuracy_medical_genetics": 0.8181818181818182,
      "mmlu_eval_accuracy_miscellaneous": 0.6511627906976745,
      "mmlu_eval_accuracy_moral_disputes": 0.5,
      "mmlu_eval_accuracy_moral_scenarios": 0.28,
      "mmlu_eval_accuracy_nutrition": 0.696969696969697,
      "mmlu_eval_accuracy_philosophy": 0.5294117647058824,
      "mmlu_eval_accuracy_prehistory": 0.4857142857142857,
      "mmlu_eval_accuracy_professional_accounting": 0.41935483870967744,
      "mmlu_eval_accuracy_professional_law": 0.4235294117647059,
      "mmlu_eval_accuracy_professional_medicine": 0.41935483870967744,
      "mmlu_eval_accuracy_professional_psychology": 0.5072463768115942,
      "mmlu_eval_accuracy_public_relations": 0.5,
      "mmlu_eval_accuracy_security_studies": 0.48148148148148145,
      "mmlu_eval_accuracy_sociology": 0.7727272727272727,
      "mmlu_eval_accuracy_us_foreign_policy": 0.6363636363636364,
      "mmlu_eval_accuracy_virology": 0.5555555555555556,
      "mmlu_eval_accuracy_world_religions": 0.7368421052631579,
      "mmlu_loss": 0.7100075650522509,
      "step": 600
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002,
      "loss": 0.5679,
      "step": 601
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0002,
      "loss": 0.3122,
      "step": 602
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0002,
      "loss": 0.4759,
      "step": 603
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0002,
      "loss": 0.429,
      "step": 604
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0002,
      "loss": 0.7318,
      "step": 605
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0002,
      "loss": 0.5422,
      "step": 606
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0002,
      "loss": 0.7202,
      "step": 607
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0002,
      "loss": 0.6671,
      "step": 608
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0002,
      "loss": 0.7528,
      "step": 609
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0002,
      "loss": 0.6146,
      "step": 610
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0002,
      "loss": 1.0854,
      "step": 611
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0002,
      "loss": 0.3982,
      "step": 612
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0002,
      "loss": 0.4596,
      "step": 613
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0002,
      "loss": 0.4383,
      "step": 614
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0002,
      "loss": 0.7133,
      "step": 615
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0002,
      "loss": 0.4333,
      "step": 616
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 0.3464,
      "step": 617
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 0.4612,
      "step": 618
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 0.4464,
      "step": 619
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 1.2875,
      "step": 620
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 0.3944,
      "step": 621
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 0.7394,
      "step": 622
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 0.6781,
      "step": 623
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0002,
      "loss": 0.6873,
      "step": 624
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0002,
      "loss": 0.5449,
      "step": 625
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0002,
      "loss": 0.5689,
      "step": 626
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0002,
      "loss": 0.3604,
      "step": 627
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0002,
      "loss": 0.5394,
      "step": 628
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0002,
      "loss": 0.5808,
      "step": 629
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0002,
      "loss": 0.4736,
      "step": 630
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0002,
      "loss": 0.8236,
      "step": 631
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0002,
      "loss": 0.3955,
      "step": 632
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0002,
      "loss": 0.709,
      "step": 633
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0002,
      "loss": 0.6104,
      "step": 634
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0002,
      "loss": 0.6482,
      "step": 635
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0002,
      "loss": 1.2371,
      "step": 636
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0002,
      "loss": 0.8039,
      "step": 637
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0002,
      "loss": 0.436,
      "step": 638
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0002,
      "loss": 0.4842,
      "step": 639
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0002,
      "loss": 1.2579,
      "step": 640
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0002,
      "loss": 0.5642,
      "step": 641
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0002,
      "loss": 0.6673,
      "step": 642
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0002,
      "loss": 0.5527,
      "step": 643
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0002,
      "loss": 0.7593,
      "step": 644
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0002,
      "loss": 0.7278,
      "step": 645
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0002,
      "loss": 0.4657,
      "step": 646
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0002,
      "loss": 0.3836,
      "step": 647
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0002,
      "loss": 0.4199,
      "step": 648
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0002,
      "loss": 0.8412,
      "step": 649
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0002,
      "loss": 0.3981,
      "step": 650
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0002,
      "loss": 0.3243,
      "step": 651
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0002,
      "loss": 0.3592,
      "step": 652
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0002,
      "loss": 0.7613,
      "step": 653
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0002,
      "loss": 0.525,
      "step": 654
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0002,
      "loss": 0.4201,
      "step": 655
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0002,
      "loss": 0.2961,
      "step": 656
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0002,
      "loss": 0.6587,
      "step": 657
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0002,
      "loss": 0.622,
      "step": 658
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0002,
      "loss": 0.631,
      "step": 659
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0002,
      "loss": 0.6036,
      "step": 660
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0002,
      "loss": 0.7581,
      "step": 661
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0002,
      "loss": 0.3218,
      "step": 662
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0002,
      "loss": 0.781,
      "step": 663
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0002,
      "loss": 0.3392,
      "step": 664
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0002,
      "loss": 0.713,
      "step": 665
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0002,
      "loss": 0.6273,
      "step": 666
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0002,
      "loss": 0.4926,
      "step": 667
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0002,
      "loss": 0.6179,
      "step": 668
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0002,
      "loss": 0.6791,
      "step": 669
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0002,
      "loss": 0.5779,
      "step": 670
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0002,
      "loss": 0.4175,
      "step": 671
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0002,
      "loss": 0.5178,
      "step": 672
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0002,
      "loss": 0.4956,
      "step": 673
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 0.5958,
      "step": 674
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 0.606,
      "step": 675
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 0.4487,
      "step": 676
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 0.4269,
      "step": 677
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 0.573,
      "step": 678
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 0.5211,
      "step": 679
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 0.7573,
      "step": 680
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0002,
      "loss": 0.6439,
      "step": 681
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0002,
      "loss": 0.6417,
      "step": 682
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0002,
      "loss": 0.6871,
      "step": 683
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0002,
      "loss": 0.5566,
      "step": 684
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0002,
      "loss": 0.4648,
      "step": 685
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0002,
      "loss": 0.3402,
      "step": 686
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0002,
      "loss": 0.4145,
      "step": 687
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0002,
      "loss": 0.5936,
      "step": 688
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0002,
      "loss": 0.525,
      "step": 689
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0002,
      "loss": 0.7718,
      "step": 690
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0002,
      "loss": 0.5453,
      "step": 691
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0002,
      "loss": 0.7362,
      "step": 692
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0002,
      "loss": 0.8271,
      "step": 693
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0002,
      "loss": 0.4735,
      "step": 694
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0002,
      "loss": 0.5554,
      "step": 695
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0002,
      "loss": 0.4636,
      "step": 696
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0002,
      "loss": 0.3986,
      "step": 697
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0002,
      "loss": 0.3132,
      "step": 698
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0002,
      "loss": 0.6233,
      "step": 699
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0002,
      "loss": 0.6888,
      "step": 700
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0002,
      "loss": 0.6325,
      "step": 701
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0002,
      "loss": 0.5242,
      "step": 702
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0002,
      "loss": 0.4286,
      "step": 703
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0002,
      "loss": 0.3178,
      "step": 704
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0002,
      "loss": 0.524,
      "step": 705
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0002,
      "loss": 0.5989,
      "step": 706
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0002,
      "loss": 0.3145,
      "step": 707
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0002,
      "loss": 0.669,
      "step": 708
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0002,
      "loss": 0.4825,
      "step": 709
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0002,
      "loss": 0.7439,
      "step": 710
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0002,
      "loss": 0.5895,
      "step": 711
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0002,
      "loss": 0.3931,
      "step": 712
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0002,
      "loss": 0.3991,
      "step": 713
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0002,
      "loss": 0.456,
      "step": 714
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0002,
      "loss": 0.6363,
      "step": 715
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0002,
      "loss": 0.4406,
      "step": 716
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0002,
      "loss": 0.5629,
      "step": 717
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0002,
      "loss": 0.4307,
      "step": 718
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0002,
      "loss": 0.5148,
      "step": 719
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0002,
      "loss": 0.3994,
      "step": 720
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0002,
      "loss": 0.5414,
      "step": 721
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0002,
      "loss": 0.5367,
      "step": 722
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 0.3362,
      "step": 723
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 0.2285,
      "step": 724
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 0.5469,
      "step": 725
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 0.3811,
      "step": 726
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 0.5973,
      "step": 727
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 0.5417,
      "step": 728
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 0.5022,
      "step": 729
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 0.517,
      "step": 730
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0002,
      "loss": 0.5841,
      "step": 731
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0002,
      "loss": 0.5231,
      "step": 732
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0002,
      "loss": 0.2776,
      "step": 733
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0002,
      "loss": 0.4693,
      "step": 734
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0002,
      "loss": 0.513,
      "step": 735
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0002,
      "loss": 0.7212,
      "step": 736
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0002,
      "loss": 0.7135,
      "step": 737
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0002,
      "loss": 0.3157,
      "step": 738
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0002,
      "loss": 0.427,
      "step": 739
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0002,
      "loss": 0.5311,
      "step": 740
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0002,
      "loss": 0.4643,
      "step": 741
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0002,
      "loss": 0.6661,
      "step": 742
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0002,
      "loss": 0.599,
      "step": 743
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0002,
      "loss": 0.5501,
      "step": 744
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0002,
      "loss": 0.4711,
      "step": 745
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0002,
      "loss": 0.474,
      "step": 746
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0002,
      "loss": 0.5778,
      "step": 747
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0002,
      "loss": 0.5552,
      "step": 748
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0002,
      "loss": 0.2903,
      "step": 749
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0002,
      "loss": 0.4615,
      "step": 750
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0002,
      "loss": 0.445,
      "step": 751
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0002,
      "loss": 0.5324,
      "step": 752
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0002,
      "loss": 0.3845,
      "step": 753
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0002,
      "loss": 0.4967,
      "step": 754
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0002,
      "loss": 0.5775,
      "step": 755
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0002,
      "loss": 0.221,
      "step": 756
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0002,
      "loss": 0.2956,
      "step": 757
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0002,
      "loss": 0.1737,
      "step": 758
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0002,
      "loss": 0.4234,
      "step": 759
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0002,
      "loss": 0.4899,
      "step": 760
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0002,
      "loss": 0.333,
      "step": 761
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0002,
      "loss": 0.46,
      "step": 762
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0002,
      "loss": 0.388,
      "step": 763
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0002,
      "loss": 0.5441,
      "step": 764
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0002,
      "loss": 0.5468,
      "step": 765
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0002,
      "loss": 0.3186,
      "step": 766
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0002,
      "loss": 0.3704,
      "step": 767
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0002,
      "loss": 0.453,
      "step": 768
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0002,
      "loss": 0.5149,
      "step": 769
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0002,
      "loss": 0.4192,
      "step": 770
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0002,
      "loss": 0.2592,
      "step": 771
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0002,
      "loss": 0.4866,
      "step": 772
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 0.3055,
      "step": 773
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 0.3449,
      "step": 774
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 0.2876,
      "step": 775
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 0.3794,
      "step": 776
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 0.6325,
      "step": 777
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 0.4825,
      "step": 778
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 0.0846,
      "step": 779
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0002,
      "loss": 0.5878,
      "step": 780
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0002,
      "loss": 0.4087,
      "step": 781
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0002,
      "loss": 0.66,
      "step": 782
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0002,
      "loss": 0.525,
      "step": 783
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0002,
      "loss": 0.362,
      "step": 784
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0002,
      "loss": 0.5427,
      "step": 785
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0002,
      "loss": 0.5045,
      "step": 786
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0002,
      "loss": 0.551,
      "step": 787
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0002,
      "loss": 0.3347,
      "step": 788
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0002,
      "loss": 0.5667,
      "step": 789
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0002,
      "loss": 0.4494,
      "step": 790
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0002,
      "loss": 0.3801,
      "step": 791
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0002,
      "loss": 0.5401,
      "step": 792
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0002,
      "loss": 0.3756,
      "step": 793
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0002,
      "loss": 0.7711,
      "step": 794
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0002,
      "loss": 0.4846,
      "step": 795
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0002,
      "loss": 0.2548,
      "step": 796
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0002,
      "loss": 0.8394,
      "step": 797
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0002,
      "loss": 0.4294,
      "step": 798
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0002,
      "loss": 0.4749,
      "step": 799
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0002,
      "loss": 0.4082,
      "step": 800
    },
    {
      "epoch": 1.12,
      "eval_loss": 0.5686381459236145,
      "eval_runtime": 397.1001,
      "eval_samples_per_second": 2.518,
      "eval_steps_per_second": 1.259,
      "step": 800
    },
    {
      "epoch": 1.12,
      "mmlu_eval_accuracy": 0.4887946989603133,
      "mmlu_eval_accuracy_abstract_algebra": 0.18181818181818182,
      "mmlu_eval_accuracy_anatomy": 0.5,
      "mmlu_eval_accuracy_astronomy": 0.25,
      "mmlu_eval_accuracy_business_ethics": 0.2727272727272727,
      "mmlu_eval_accuracy_clinical_knowledge": 0.4482758620689655,
      "mmlu_eval_accuracy_college_biology": 0.4375,
      "mmlu_eval_accuracy_college_chemistry": 0.25,
      "mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
      "mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
      "mmlu_eval_accuracy_college_medicine": 0.36363636363636365,
      "mmlu_eval_accuracy_college_physics": 0.36363636363636365,
      "mmlu_eval_accuracy_computer_security": 0.5454545454545454,
      "mmlu_eval_accuracy_conceptual_physics": 0.38461538461538464,
      "mmlu_eval_accuracy_econometrics": 0.3333333333333333,
      "mmlu_eval_accuracy_electrical_engineering": 0.3125,
      "mmlu_eval_accuracy_elementary_mathematics": 0.36585365853658536,
      "mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
      "mmlu_eval_accuracy_global_facts": 0.3,
      "mmlu_eval_accuracy_high_school_biology": 0.375,
      "mmlu_eval_accuracy_high_school_chemistry": 0.2727272727272727,
      "mmlu_eval_accuracy_high_school_computer_science": 0.7777777777777778,
      "mmlu_eval_accuracy_high_school_european_history": 0.8333333333333334,
      "mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.37209302325581395,
      "mmlu_eval_accuracy_high_school_mathematics": 0.3793103448275862,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.46153846153846156,
      "mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
      "mmlu_eval_accuracy_high_school_psychology": 0.8166666666666667,
      "mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
      "mmlu_eval_accuracy_high_school_us_history": 0.6363636363636364,
      "mmlu_eval_accuracy_high_school_world_history": 0.5769230769230769,
      "mmlu_eval_accuracy_human_aging": 0.6086956521739131,
      "mmlu_eval_accuracy_human_sexuality": 0.5,
      "mmlu_eval_accuracy_international_law": 0.6923076923076923,
      "mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
      "mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
      "mmlu_eval_accuracy_machine_learning": 0.5454545454545454,
      "mmlu_eval_accuracy_management": 0.7272727272727273,
      "mmlu_eval_accuracy_marketing": 0.72,
      "mmlu_eval_accuracy_medical_genetics": 0.8181818181818182,
      "mmlu_eval_accuracy_miscellaneous": 0.6511627906976745,
      "mmlu_eval_accuracy_moral_disputes": 0.4473684210526316,
      "mmlu_eval_accuracy_moral_scenarios": 0.25,
      "mmlu_eval_accuracy_nutrition": 0.5757575757575758,
      "mmlu_eval_accuracy_philosophy": 0.6176470588235294,
      "mmlu_eval_accuracy_prehistory": 0.45714285714285713,
      "mmlu_eval_accuracy_professional_accounting": 0.3225806451612903,
      "mmlu_eval_accuracy_professional_law": 0.37058823529411766,
      "mmlu_eval_accuracy_professional_medicine": 0.45161290322580644,
      "mmlu_eval_accuracy_professional_psychology": 0.5362318840579711,
      "mmlu_eval_accuracy_public_relations": 0.4166666666666667,
      "mmlu_eval_accuracy_security_studies": 0.48148148148148145,
      "mmlu_eval_accuracy_sociology": 0.7272727272727273,
      "mmlu_eval_accuracy_us_foreign_policy": 0.6363636363636364,
      "mmlu_eval_accuracy_virology": 0.4444444444444444,
      "mmlu_eval_accuracy_world_religions": 0.631578947368421,
      "mmlu_loss": 0.827487740422509,
      "step": 800
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0002,
      "loss": 0.7257,
      "step": 801
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0002,
      "loss": 0.4948,
      "step": 802
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0002,
      "loss": 0.3793,
      "step": 803
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0002,
      "loss": 0.7008,
      "step": 804
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0002,
      "loss": 0.582,
      "step": 805
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0002,
      "loss": 0.4917,
      "step": 806
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0002,
      "loss": 0.494,
      "step": 807
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0002,
      "loss": 0.3638,
      "step": 808
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0002,
      "loss": 0.3804,
      "step": 809
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0002,
      "loss": 0.3673,
      "step": 810
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0002,
      "loss": 0.5424,
      "step": 811
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0002,
      "loss": 0.4179,
      "step": 812
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0002,
      "loss": 0.7281,
      "step": 813
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0002,
      "loss": 1.0638,
      "step": 814
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0002,
      "loss": 0.6234,
      "step": 815
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0002,
      "loss": 0.4955,
      "step": 816
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0002,
      "loss": 0.5072,
      "step": 817
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0002,
      "loss": 0.5987,
      "step": 818
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0002,
      "loss": 0.4196,
      "step": 819
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0002,
      "loss": 0.4912,
      "step": 820
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0002,
      "loss": 0.6076,
      "step": 821
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0002,
      "loss": 0.4394,
      "step": 822
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0002,
      "loss": 0.3876,
      "step": 823
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0002,
      "loss": 0.4345,
      "step": 824
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0002,
      "loss": 0.6818,
      "step": 825
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0002,
      "loss": 0.5721,
      "step": 826
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0002,
      "loss": 0.3784,
      "step": 827
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0002,
      "loss": 0.3719,
      "step": 828
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0002,
      "loss": 0.5833,
      "step": 829
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 0.3224,
      "step": 830
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 0.506,
      "step": 831
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 0.5069,
      "step": 832
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 0.662,
      "step": 833
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 0.2466,
      "step": 834
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 0.2753,
      "step": 835
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 0.6013,
      "step": 836
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0002,
      "loss": 0.43,
      "step": 837
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0002,
      "loss": 0.441,
      "step": 838
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0002,
      "loss": 0.3748,
      "step": 839
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0002,
      "loss": 0.3574,
      "step": 840
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0002,
      "loss": 0.5118,
      "step": 841
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0002,
      "loss": 0.4662,
      "step": 842
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0002,
      "loss": 0.5061,
      "step": 843
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0002,
      "loss": 0.4125,
      "step": 844
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0002,
      "loss": 0.4514,
      "step": 845
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0002,
      "loss": 0.485,
      "step": 846
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0002,
      "loss": 0.4109,
      "step": 847
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0002,
      "loss": 0.6806,
      "step": 848
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0002,
      "loss": 0.4135,
      "step": 849
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0002,
      "loss": 0.2929,
      "step": 850
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0002,
      "loss": 0.5998,
      "step": 851
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0002,
      "loss": 0.6018,
      "step": 852
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0002,
      "loss": 0.4222,
      "step": 853
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0002,
      "loss": 0.613,
      "step": 854
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0002,
      "loss": 0.5271,
      "step": 855
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0002,
      "loss": 0.2058,
      "step": 856
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0002,
      "loss": 0.5006,
      "step": 857
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0002,
      "loss": 0.3047,
      "step": 858
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0002,
      "loss": 0.4376,
      "step": 859
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0002,
      "loss": 0.6368,
      "step": 860
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0002,
      "loss": 0.9255,
      "step": 861
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0002,
      "loss": 0.8196,
      "step": 862
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0002,
      "loss": 0.4536,
      "step": 863
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0002,
      "loss": 0.3389,
      "step": 864
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0002,
      "loss": 0.3768,
      "step": 865
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0002,
      "loss": 0.532,
      "step": 866
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0002,
      "loss": 0.2816,
      "step": 867
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0002,
      "loss": 0.3547,
      "step": 868
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0002,
      "loss": 0.6835,
      "step": 869
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0002,
      "loss": 0.5265,
      "step": 870
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0002,
      "loss": 0.2499,
      "step": 871
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0002,
      "loss": 0.823,
      "step": 872
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0002,
      "loss": 0.8358,
      "step": 873
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0002,
      "loss": 0.4189,
      "step": 874
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0002,
      "loss": 0.5019,
      "step": 875
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0002,
      "loss": 0.7701,
      "step": 876
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0002,
      "loss": 0.6844,
      "step": 877
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0002,
      "loss": 0.5303,
      "step": 878
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0002,
      "loss": 0.5351,
      "step": 879
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 0.559,
      "step": 880
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 0.3725,
      "step": 881
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 0.398,
      "step": 882
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 0.5737,
      "step": 883
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 0.2967,
      "step": 884
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 0.4052,
      "step": 885
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 0.6591,
      "step": 886
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0002,
      "loss": 0.4774,
      "step": 887
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0002,
      "loss": 0.4082,
      "step": 888
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0002,
      "loss": 0.5407,
      "step": 889
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0002,
      "loss": 0.4064,
      "step": 890
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0002,
      "loss": 0.4437,
      "step": 891
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0002,
      "loss": 0.6069,
      "step": 892
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0002,
      "loss": 0.4905,
      "step": 893
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0002,
      "loss": 0.3571,
      "step": 894
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0002,
      "loss": 0.6318,
      "step": 895
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0002,
      "loss": 0.4658,
      "step": 896
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0002,
      "loss": 0.5483,
      "step": 897
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0002,
      "loss": 0.5362,
      "step": 898
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0002,
      "loss": 0.7549,
      "step": 899
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0002,
      "loss": 0.3347,
      "step": 900
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0002,
      "loss": 0.5958,
      "step": 901
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0002,
      "loss": 0.4665,
      "step": 902
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0002,
      "loss": 0.4061,
      "step": 903
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0002,
      "loss": 0.2981,
      "step": 904
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0002,
      "loss": 0.6599,
      "step": 905
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0002,
      "loss": 0.3867,
      "step": 906
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0002,
      "loss": 0.4789,
      "step": 907
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0002,
      "loss": 0.6472,
      "step": 908
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0002,
      "loss": 0.6347,
      "step": 909
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0002,
      "loss": 0.5878,
      "step": 910
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0002,
      "loss": 0.7185,
      "step": 911
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0002,
      "loss": 0.4526,
      "step": 912
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0002,
      "loss": 0.7132,
      "step": 913
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0002,
      "loss": 0.9166,
      "step": 914
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0002,
      "loss": 0.2947,
      "step": 915
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0002,
      "loss": 0.422,
      "step": 916
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0002,
      "loss": 0.368,
      "step": 917
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0002,
      "loss": 0.4067,
      "step": 918
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0002,
      "loss": 0.5581,
      "step": 919
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0002,
      "loss": 0.779,
      "step": 920
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0002,
      "loss": 0.5046,
      "step": 921
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0002,
      "loss": 0.2822,
      "step": 922
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0002,
      "loss": 0.5238,
      "step": 923
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0002,
      "loss": 0.4115,
      "step": 924
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0002,
      "loss": 0.5919,
      "step": 925
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0002,
      "loss": 0.3671,
      "step": 926
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0002,
      "loss": 0.3481,
      "step": 927
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0002,
      "loss": 0.4921,
      "step": 928
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0002,
      "loss": 0.6569,
      "step": 929
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 0.5033,
      "step": 930
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 0.5091,
      "step": 931
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 0.4866,
      "step": 932
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 0.4251,
      "step": 933
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 0.4045,
      "step": 934
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 0.3586,
      "step": 935
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 0.4785,
      "step": 936
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0002,
      "loss": 0.4966,
      "step": 937
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0002,
      "loss": 0.3425,
      "step": 938
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0002,
      "loss": 0.4925,
      "step": 939
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0002,
      "loss": 0.4038,
      "step": 940
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0002,
      "loss": 0.5039,
      "step": 941
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0002,
      "loss": 0.715,
      "step": 942
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0002,
      "loss": 0.6889,
      "step": 943
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0002,
      "loss": 0.4278,
      "step": 944
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0002,
      "loss": 0.4127,
      "step": 945
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0002,
      "loss": 0.5168,
      "step": 946
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0002,
      "loss": 0.5567,
      "step": 947
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0002,
      "loss": 0.4922,
      "step": 948
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0002,
      "loss": 0.4343,
      "step": 949
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0002,
      "loss": 0.7558,
      "step": 950
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0002,
      "loss": 0.5109,
      "step": 951
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0002,
      "loss": 0.4131,
      "step": 952
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0002,
      "loss": 0.5395,
      "step": 953
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0002,
      "loss": 0.4339,
      "step": 954
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0002,
      "loss": 0.3693,
      "step": 955
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0002,
      "loss": 0.2442,
      "step": 956
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0002,
      "loss": 0.3236,
      "step": 957
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0002,
      "loss": 0.5036,
      "step": 958
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0002,
      "loss": 0.5788,
      "step": 959
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0002,
      "loss": 0.6294,
      "step": 960
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0002,
      "loss": 0.6333,
      "step": 961
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0002,
      "loss": 0.2901,
      "step": 962
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0002,
      "loss": 0.7099,
      "step": 963
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0002,
      "loss": 0.4176,
      "step": 964
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0002,
      "loss": 0.1454,
      "step": 965
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0002,
      "loss": 0.5429,
      "step": 966
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0002,
      "loss": 0.5377,
      "step": 967
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0002,
      "loss": 0.2129,
      "step": 968
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0002,
      "loss": 0.5728,
      "step": 969
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0002,
      "loss": 0.4328,
      "step": 970
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0002,
      "loss": 0.3753,
      "step": 971
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0002,
      "loss": 0.6478,
      "step": 972
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0002,
      "loss": 0.3985,
      "step": 973
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0002,
      "loss": 0.3202,
      "step": 974
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0002,
      "loss": 0.22,
      "step": 975
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0002,
      "loss": 0.5935,
      "step": 976
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0002,
      "loss": 0.8957,
      "step": 977
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0002,
      "loss": 0.5153,
      "step": 978
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0002,
      "loss": 0.4195,
      "step": 979
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 0.3145,
      "step": 980
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 0.559,
      "step": 981
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 0.5841,
      "step": 982
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 0.3964,
      "step": 983
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 0.7424,
      "step": 984
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 0.3926,
      "step": 985
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 0.3537,
      "step": 986
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0002,
      "loss": 0.4892,
      "step": 987
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0002,
      "loss": 0.5922,
      "step": 988
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0002,
      "loss": 0.7276,
      "step": 989
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0002,
      "loss": 0.507,
      "step": 990
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0002,
      "loss": 0.6271,
      "step": 991
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0002,
      "loss": 0.5251,
      "step": 992
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0002,
      "loss": 0.5694,
      "step": 993
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0002,
      "loss": 0.2735,
      "step": 994
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0002,
      "loss": 0.4638,
      "step": 995
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0002,
      "loss": 0.4626,
      "step": 996
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0002,
      "loss": 0.4226,
      "step": 997
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0002,
      "loss": 0.4178,
      "step": 998
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0002,
      "loss": 0.4433,
      "step": 999
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0002,
      "loss": 0.5467,
      "step": 1000
    },
    {
      "epoch": 1.4,
      "eval_loss": 0.5548574328422546,
      "eval_runtime": 397.7064,
      "eval_samples_per_second": 2.514,
      "eval_steps_per_second": 1.257,
      "step": 1000
    },
    {
      "epoch": 1.4,
      "mmlu_eval_accuracy": 0.4721227283910478,
      "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
      "mmlu_eval_accuracy_anatomy": 0.42857142857142855,
      "mmlu_eval_accuracy_astronomy": 0.3125,
      "mmlu_eval_accuracy_business_ethics": 0.36363636363636365,
      "mmlu_eval_accuracy_clinical_knowledge": 0.3793103448275862,
      "mmlu_eval_accuracy_college_biology": 0.5,
      "mmlu_eval_accuracy_college_chemistry": 0.125,
      "mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
      "mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
      "mmlu_eval_accuracy_college_medicine": 0.45454545454545453,
      "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
      "mmlu_eval_accuracy_computer_security": 0.6363636363636364,
      "mmlu_eval_accuracy_conceptual_physics": 0.38461538461538464,
      "mmlu_eval_accuracy_econometrics": 0.4166666666666667,
      "mmlu_eval_accuracy_electrical_engineering": 0.3125,
      "mmlu_eval_accuracy_elementary_mathematics": 0.36585365853658536,
      "mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
      "mmlu_eval_accuracy_global_facts": 0.2,
      "mmlu_eval_accuracy_high_school_biology": 0.4375,
      "mmlu_eval_accuracy_high_school_chemistry": 0.09090909090909091,
      "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_european_history": 0.6111111111111112,
      "mmlu_eval_accuracy_high_school_geography": 0.7272727272727273,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.3488372093023256,
      "mmlu_eval_accuracy_high_school_mathematics": 0.3448275862068966,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.5384615384615384,
      "mmlu_eval_accuracy_high_school_physics": 0.23529411764705882,
      "mmlu_eval_accuracy_high_school_psychology": 0.8,
      "mmlu_eval_accuracy_high_school_statistics": 0.21739130434782608,
      "mmlu_eval_accuracy_high_school_us_history": 0.5909090909090909,
      "mmlu_eval_accuracy_high_school_world_history": 0.6153846153846154,
      "mmlu_eval_accuracy_human_aging": 0.6521739130434783,
      "mmlu_eval_accuracy_human_sexuality": 0.3333333333333333,
      "mmlu_eval_accuracy_international_law": 0.8461538461538461,
      "mmlu_eval_accuracy_jurisprudence": 0.5454545454545454,
      "mmlu_eval_accuracy_logical_fallacies": 0.7222222222222222,
      "mmlu_eval_accuracy_machine_learning": 0.18181818181818182,
      "mmlu_eval_accuracy_management": 0.6363636363636364,
      "mmlu_eval_accuracy_marketing": 0.76,
      "mmlu_eval_accuracy_medical_genetics": 0.7272727272727273,
      "mmlu_eval_accuracy_miscellaneous": 0.627906976744186,
      "mmlu_eval_accuracy_moral_disputes": 0.5,
      "mmlu_eval_accuracy_moral_scenarios": 0.28,
      "mmlu_eval_accuracy_nutrition": 0.6363636363636364,
      "mmlu_eval_accuracy_philosophy": 0.5882352941176471,
      "mmlu_eval_accuracy_prehistory": 0.5714285714285714,
      "mmlu_eval_accuracy_professional_accounting": 0.3225806451612903,
      "mmlu_eval_accuracy_professional_law": 0.4117647058823529,
      "mmlu_eval_accuracy_professional_medicine": 0.3225806451612903,
      "mmlu_eval_accuracy_professional_psychology": 0.42028985507246375,
      "mmlu_eval_accuracy_public_relations": 0.4166666666666667,
      "mmlu_eval_accuracy_security_studies": 0.37037037037037035,
      "mmlu_eval_accuracy_sociology": 0.6363636363636364,
      "mmlu_eval_accuracy_us_foreign_policy": 0.5454545454545454,
      "mmlu_eval_accuracy_virology": 0.5555555555555556,
      "mmlu_eval_accuracy_world_religions": 0.631578947368421,
      "mmlu_loss": 0.6330746299643119,
      "step": 1000
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0002,
      "loss": 0.512,
      "step": 1001
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0002,
      "loss": 0.4605,
      "step": 1002
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0002,
      "loss": 0.4642,
      "step": 1003
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0002,
      "loss": 0.5197,
      "step": 1004
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0002,
      "loss": 0.7914,
      "step": 1005
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0002,
      "loss": 0.4228,
      "step": 1006
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0002,
      "loss": 0.608,
      "step": 1007
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0002,
      "loss": 0.5949,
      "step": 1008
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0002,
      "loss": 0.4641,
      "step": 1009
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0002,
      "loss": 0.5826,
      "step": 1010
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0002,
      "loss": 0.447,
      "step": 1011
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0002,
      "loss": 0.3685,
      "step": 1012
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0002,
      "loss": 0.5129,
      "step": 1013
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0002,
      "loss": 0.6513,
      "step": 1014
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0002,
      "loss": 0.4045,
      "step": 1015
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0002,
      "loss": 0.8535,
      "step": 1016
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0002,
      "loss": 0.4578,
      "step": 1017
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0002,
      "loss": 0.3557,
      "step": 1018
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0002,
      "loss": 0.5615,
      "step": 1019
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0002,
      "loss": 0.5109,
      "step": 1020
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0002,
      "loss": 0.5499,
      "step": 1021
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0002,
      "loss": 0.7445,
      "step": 1022
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0002,
      "loss": 0.4995,
      "step": 1023
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0002,
      "loss": 0.6331,
      "step": 1024
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0002,
      "loss": 0.4429,
      "step": 1025
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0002,
      "loss": 0.535,
      "step": 1026
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0002,
      "loss": 0.4315,
      "step": 1027
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0002,
      "loss": 0.4712,
      "step": 1028
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0002,
      "loss": 0.7811,
      "step": 1029
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0002,
      "loss": 0.2913,
      "step": 1030
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0002,
      "loss": 0.2231,
      "step": 1031
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0002,
      "loss": 0.6743,
      "step": 1032
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0002,
      "loss": 0.3777,
      "step": 1033
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0002,
      "loss": 0.3063,
      "step": 1034
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0002,
      "loss": 0.4412,
      "step": 1035
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0002,
      "loss": 0.402,
      "step": 1036
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 0.3651,
      "step": 1037
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 0.624,
      "step": 1038
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 0.467,
      "step": 1039
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 0.3748,
      "step": 1040
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 0.4512,
      "step": 1041
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 0.4396,
      "step": 1042
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 0.3602,
      "step": 1043
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0002,
      "loss": 0.5191,
      "step": 1044
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0002,
      "loss": 0.3586,
      "step": 1045
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0002,
      "loss": 0.4412,
      "step": 1046
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0002,
      "loss": 0.5509,
      "step": 1047
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0002,
      "loss": 0.1959,
      "step": 1048
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0002,
      "loss": 0.7194,
      "step": 1049
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0002,
      "loss": 0.5329,
      "step": 1050
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0002,
      "loss": 0.3183,
      "step": 1051
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0002,
      "loss": 0.529,
      "step": 1052
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0002,
      "loss": 0.5259,
      "step": 1053
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0002,
      "loss": 0.4782,
      "step": 1054
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0002,
      "loss": 0.5058,
      "step": 1055
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0002,
      "loss": 0.4638,
      "step": 1056
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0002,
      "loss": 0.5469,
      "step": 1057
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0002,
      "loss": 0.7656,
      "step": 1058
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0002,
      "loss": 0.2239,
      "step": 1059
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0002,
      "loss": 0.2989,
      "step": 1060
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0002,
      "loss": 0.6438,
      "step": 1061
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0002,
      "loss": 0.2802,
      "step": 1062
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0002,
      "loss": 0.4715,
      "step": 1063
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0002,
      "loss": 0.5388,
      "step": 1064
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0002,
      "loss": 0.5813,
      "step": 1065
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0002,
      "loss": 0.8507,
      "step": 1066
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0002,
      "loss": 0.3369,
      "step": 1067
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0002,
      "loss": 0.5435,
      "step": 1068
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0002,
      "loss": 0.4366,
      "step": 1069
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0002,
      "loss": 0.4863,
      "step": 1070
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0002,
      "loss": 0.4534,
      "step": 1071
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0002,
      "loss": 0.5245,
      "step": 1072
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0002,
      "loss": 0.604,
      "step": 1073
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0002,
      "loss": 0.9578,
      "step": 1074
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0002,
      "loss": 0.4446,
      "step": 1075
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0002,
      "loss": 0.7155,
      "step": 1076
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0002,
      "loss": 0.4007,
      "step": 1077
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0002,
      "loss": 0.5243,
      "step": 1078
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0002,
      "loss": 0.3955,
      "step": 1079
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0002,
      "loss": 0.3289,
      "step": 1080
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0002,
      "loss": 0.3196,
      "step": 1081
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0002,
      "loss": 0.4171,
      "step": 1082
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0002,
      "loss": 0.4861,
      "step": 1083
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0002,
      "loss": 0.3964,
      "step": 1084
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0002,
      "loss": 0.4036,
      "step": 1085
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0002,
      "loss": 0.6718,
      "step": 1086
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 0.4988,
      "step": 1087
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 0.3719,
      "step": 1088
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 0.8588,
      "step": 1089
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 0.5142,
      "step": 1090
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 0.6402,
      "step": 1091
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 0.5727,
      "step": 1092
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 1.1739,
      "step": 1093
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0002,
      "loss": 0.7557,
      "step": 1094
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0002,
      "loss": 0.5089,
      "step": 1095
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0002,
      "loss": 0.3532,
      "step": 1096
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0002,
      "loss": 0.5793,
      "step": 1097
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0002,
      "loss": 0.4937,
      "step": 1098
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0002,
      "loss": 0.5031,
      "step": 1099
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0002,
      "loss": 0.2077,
      "step": 1100
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0002,
      "loss": 0.5418,
      "step": 1101
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0002,
      "loss": 0.3988,
      "step": 1102
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0002,
      "loss": 0.5441,
      "step": 1103
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0002,
      "loss": 0.6248,
      "step": 1104
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0002,
      "loss": 0.5542,
      "step": 1105
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0002,
      "loss": 0.3487,
      "step": 1106
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0002,
      "loss": 0.5826,
      "step": 1107
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0002,
      "loss": 0.3423,
      "step": 1108
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0002,
      "loss": 0.1865,
      "step": 1109
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0002,
      "loss": 0.4359,
      "step": 1110
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0002,
      "loss": 0.3346,
      "step": 1111
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0002,
      "loss": 0.4429,
      "step": 1112
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0002,
      "loss": 0.6522,
      "step": 1113
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0002,
      "loss": 0.4003,
      "step": 1114
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0002,
      "loss": 0.5939,
      "step": 1115
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0002,
      "loss": 0.2761,
      "step": 1116
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0002,
      "loss": 0.7465,
      "step": 1117
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0002,
      "loss": 0.3926,
      "step": 1118
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0002,
      "loss": 0.4905,
      "step": 1119
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0002,
      "loss": 0.1575,
      "step": 1120
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0002,
      "loss": 0.4862,
      "step": 1121
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0002,
      "loss": 0.4529,
      "step": 1122
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0002,
      "loss": 0.7323,
      "step": 1123
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0002,
      "loss": 0.3422,
      "step": 1124
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0002,
      "loss": 0.4838,
      "step": 1125
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0002,
      "loss": 0.5205,
      "step": 1126
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0002,
      "loss": 0.4127,
      "step": 1127
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0002,
      "loss": 0.4456,
      "step": 1128
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0002,
      "loss": 0.1641,
      "step": 1129
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0002,
      "loss": 0.3075,
      "step": 1130
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0002,
      "loss": 0.4241,
      "step": 1131
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0002,
      "loss": 0.4734,
      "step": 1132
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0002,
      "loss": 0.1663,
      "step": 1133
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0002,
      "loss": 0.5906,
      "step": 1134
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0002,
      "loss": 0.4918,
      "step": 1135
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0002,
      "loss": 0.2649,
      "step": 1136
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 0.5191,
      "step": 1137
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 0.5273,
      "step": 1138
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 0.4774,
      "step": 1139
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 0.3517,
      "step": 1140
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 0.3579,
      "step": 1141
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 0.4866,
      "step": 1142
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 0.5646,
      "step": 1143
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0002,
      "loss": 0.3094,
      "step": 1144
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0002,
      "loss": 0.5368,
      "step": 1145
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0002,
      "loss": 0.7069,
      "step": 1146
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0002,
      "loss": 0.528,
      "step": 1147
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0002,
      "loss": 0.4145,
      "step": 1148
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0002,
      "loss": 0.4658,
      "step": 1149
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0002,
      "loss": 0.6312,
      "step": 1150
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0002,
      "loss": 0.4069,
      "step": 1151
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0002,
      "loss": 0.5671,
      "step": 1152
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0002,
      "loss": 0.1648,
      "step": 1153
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0002,
      "loss": 0.1745,
      "step": 1154
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0002,
      "loss": 0.6023,
      "step": 1155
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0002,
      "loss": 0.5318,
      "step": 1156
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0002,
      "loss": 0.2937,
      "step": 1157
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0002,
      "loss": 0.2239,
      "step": 1158
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0002,
      "loss": 0.3798,
      "step": 1159
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0002,
      "loss": 0.4392,
      "step": 1160
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0002,
      "loss": 0.5516,
      "step": 1161
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0002,
      "loss": 0.2776,
      "step": 1162
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0002,
      "loss": 0.7169,
      "step": 1163
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0002,
      "loss": 0.5286,
      "step": 1164
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0002,
      "loss": 0.557,
      "step": 1165
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0002,
      "loss": 0.4192,
      "step": 1166
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0002,
      "loss": 0.8117,
      "step": 1167
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0002,
      "loss": 0.5134,
      "step": 1168
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0002,
      "loss": 0.4967,
      "step": 1169
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0002,
      "loss": 0.4486,
      "step": 1170
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0002,
      "loss": 0.6065,
      "step": 1171
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0002,
      "loss": 0.4096,
      "step": 1172
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0002,
      "loss": 0.529,
      "step": 1173
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0002,
      "loss": 0.5445,
      "step": 1174
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0002,
      "loss": 0.3461,
      "step": 1175
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0002,
      "loss": 0.5853,
      "step": 1176
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0002,
      "loss": 0.5213,
      "step": 1177
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0002,
      "loss": 0.4355,
      "step": 1178
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0002,
      "loss": 0.6598,
      "step": 1179
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0002,
      "loss": 0.3969,
      "step": 1180
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0002,
      "loss": 0.6133,
      "step": 1181
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0002,
      "loss": 0.5464,
      "step": 1182
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0002,
      "loss": 0.4366,
      "step": 1183
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0002,
      "loss": 0.5487,
      "step": 1184
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0002,
      "loss": 0.5514,
      "step": 1185
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0002,
      "loss": 0.2731,
      "step": 1186
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0002,
      "loss": 0.4296,
      "step": 1187
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0002,
      "loss": 0.7976,
      "step": 1188
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0002,
      "loss": 0.3542,
      "step": 1189
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0002,
      "loss": 0.4839,
      "step": 1190
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0002,
      "loss": 0.5778,
      "step": 1191
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0002,
      "loss": 0.3116,
      "step": 1192
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0002,
      "loss": 0.6339,
      "step": 1193
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 0.3298,
      "step": 1194
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 0.659,
      "step": 1195
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 0.4763,
      "step": 1196
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 0.6885,
      "step": 1197
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 0.5319,
      "step": 1198
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 0.3336,
      "step": 1199
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 0.7863,
      "step": 1200
    },
    {
      "epoch": 1.68,
      "eval_loss": 0.5517624616622925,
      "eval_runtime": 398.1851,
      "eval_samples_per_second": 2.511,
      "eval_steps_per_second": 1.256,
      "step": 1200
    },
    {
      "epoch": 1.68,
      "mmlu_eval_accuracy": 0.4673485527289215,
      "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
      "mmlu_eval_accuracy_anatomy": 0.5,
      "mmlu_eval_accuracy_astronomy": 0.375,
      "mmlu_eval_accuracy_business_ethics": 0.45454545454545453,
      "mmlu_eval_accuracy_clinical_knowledge": 0.41379310344827586,
      "mmlu_eval_accuracy_college_biology": 0.5,
      "mmlu_eval_accuracy_college_chemistry": 0.0,
      "mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
      "mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
      "mmlu_eval_accuracy_college_medicine": 0.5,
      "mmlu_eval_accuracy_college_physics": 0.36363636363636365,
      "mmlu_eval_accuracy_computer_security": 0.6363636363636364,
      "mmlu_eval_accuracy_conceptual_physics": 0.5,
      "mmlu_eval_accuracy_econometrics": 0.4166666666666667,
      "mmlu_eval_accuracy_electrical_engineering": 0.1875,
      "mmlu_eval_accuracy_elementary_mathematics": 0.36585365853658536,
      "mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
      "mmlu_eval_accuracy_global_facts": 0.3,
      "mmlu_eval_accuracy_high_school_biology": 0.375,
      "mmlu_eval_accuracy_high_school_chemistry": 0.13636363636363635,
      "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_european_history": 0.4444444444444444,
      "mmlu_eval_accuracy_high_school_geography": 0.7272727272727273,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.3953488372093023,
      "mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.5384615384615384,
      "mmlu_eval_accuracy_high_school_physics": 0.23529411764705882,
      "mmlu_eval_accuracy_high_school_psychology": 0.7333333333333333,
      "mmlu_eval_accuracy_high_school_statistics": 0.2608695652173913,
      "mmlu_eval_accuracy_high_school_us_history": 0.5909090909090909,
      "mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
      "mmlu_eval_accuracy_human_aging": 0.5217391304347826,
      "mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
      "mmlu_eval_accuracy_international_law": 0.7692307692307693,
      "mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
      "mmlu_eval_accuracy_logical_fallacies": 0.7222222222222222,
      "mmlu_eval_accuracy_machine_learning": 0.2727272727272727,
      "mmlu_eval_accuracy_management": 0.8181818181818182,
      "mmlu_eval_accuracy_marketing": 0.84,
      "mmlu_eval_accuracy_medical_genetics": 0.8181818181818182,
      "mmlu_eval_accuracy_miscellaneous": 0.627906976744186,
      "mmlu_eval_accuracy_moral_disputes": 0.4473684210526316,
      "mmlu_eval_accuracy_moral_scenarios": 0.24,
      "mmlu_eval_accuracy_nutrition": 0.6363636363636364,
      "mmlu_eval_accuracy_philosophy": 0.47058823529411764,
      "mmlu_eval_accuracy_prehistory": 0.5714285714285714,
      "mmlu_eval_accuracy_professional_accounting": 0.2903225806451613,
      "mmlu_eval_accuracy_professional_law": 0.35294117647058826,
      "mmlu_eval_accuracy_professional_medicine": 0.3548387096774194,
      "mmlu_eval_accuracy_professional_psychology": 0.43478260869565216,
      "mmlu_eval_accuracy_public_relations": 0.25,
      "mmlu_eval_accuracy_security_studies": 0.4444444444444444,
      "mmlu_eval_accuracy_sociology": 0.7272727272727273,
      "mmlu_eval_accuracy_us_foreign_policy": 0.5454545454545454,
      "mmlu_eval_accuracy_virology": 0.3333333333333333,
      "mmlu_eval_accuracy_world_religions": 0.631578947368421,
      "mmlu_loss": 0.7883516700923754,
      "step": 1200
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0002,
      "loss": 0.6287,
      "step": 1201
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0002,
      "loss": 0.2449,
      "step": 1202
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0002,
      "loss": 0.3399,
      "step": 1203
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0002,
      "loss": 0.4511,
      "step": 1204
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0002,
      "loss": 0.7452,
      "step": 1205
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0002,
      "loss": 0.7483,
      "step": 1206
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0002,
      "loss": 0.4456,
      "step": 1207
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0002,
      "loss": 0.4566,
      "step": 1208
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0002,
      "loss": 0.4174,
      "step": 1209
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0002,
      "loss": 0.4166,
      "step": 1210
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0002,
      "loss": 0.5438,
      "step": 1211
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0002,
      "loss": 0.408,
      "step": 1212
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0002,
      "loss": 0.3984,
      "step": 1213
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0002,
      "loss": 0.4479,
      "step": 1214
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0002,
      "loss": 0.7566,
      "step": 1215
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0002,
      "loss": 0.6747,
      "step": 1216
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0002,
      "loss": 0.4957,
      "step": 1217
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0002,
      "loss": 1.1681,
      "step": 1218
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0002,
      "loss": 0.6068,
      "step": 1219
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0002,
      "loss": 0.3397,
      "step": 1220
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0002,
      "loss": 0.4097,
      "step": 1221
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0002,
      "loss": 0.5339,
      "step": 1222
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0002,
      "loss": 0.3092,
      "step": 1223
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0002,
      "loss": 0.482,
      "step": 1224
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0002,
      "loss": 0.4504,
      "step": 1225
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0002,
      "loss": 0.181,
      "step": 1226
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0002,
      "loss": 0.512,
      "step": 1227
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0002,
      "loss": 0.5132,
      "step": 1228
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0002,
      "loss": 0.5487,
      "step": 1229
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0002,
      "loss": 0.4302,
      "step": 1230
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0002,
      "loss": 0.4496,
      "step": 1231
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0002,
      "loss": 0.3834,
      "step": 1232
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0002,
      "loss": 0.5786,
      "step": 1233
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0002,
      "loss": 0.2782,
      "step": 1234
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0002,
      "loss": 0.2991,
      "step": 1235
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0002,
      "loss": 0.7636,
      "step": 1236
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0002,
      "loss": 0.5127,
      "step": 1237
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0002,
      "loss": 0.2507,
      "step": 1238
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0002,
      "loss": 0.5053,
      "step": 1239
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0002,
      "loss": 0.4294,
      "step": 1240
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0002,
      "loss": 0.4264,
      "step": 1241
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0002,
      "loss": 0.5081,
      "step": 1242
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 0.7329,
      "step": 1243
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 0.2069,
      "step": 1244
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 0.4969,
      "step": 1245
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 0.7081,
      "step": 1246
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 0.2848,
      "step": 1247
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 0.2537,
      "step": 1248
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 0.63,
      "step": 1249
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0002,
      "loss": 0.5571,
      "step": 1250
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0002,
      "loss": 0.5448,
      "step": 1251
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0002,
      "loss": 0.6951,
      "step": 1252
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0002,
      "loss": 0.5987,
      "step": 1253
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0002,
      "loss": 0.3836,
      "step": 1254
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0002,
      "loss": 0.4985,
      "step": 1255
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0002,
      "loss": 0.4734,
      "step": 1256
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0002,
      "loss": 0.2987,
      "step": 1257
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0002,
      "loss": 0.7252,
      "step": 1258
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0002,
      "loss": 0.399,
      "step": 1259
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0002,
      "loss": 0.6162,
      "step": 1260
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0002,
      "loss": 0.6824,
      "step": 1261
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0002,
      "loss": 0.4843,
      "step": 1262
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0002,
      "loss": 0.4097,
      "step": 1263
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0002,
      "loss": 0.4881,
      "step": 1264
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0002,
      "loss": 0.4019,
      "step": 1265
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0002,
      "loss": 0.3003,
      "step": 1266
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0002,
      "loss": 0.4741,
      "step": 1267
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0002,
      "loss": 0.2978,
      "step": 1268
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0002,
      "loss": 0.8337,
      "step": 1269
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0002,
      "loss": 0.3119,
      "step": 1270
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0002,
      "loss": 0.3243,
      "step": 1271
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0002,
      "loss": 0.6131,
      "step": 1272
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0002,
      "loss": 0.6219,
      "step": 1273
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0002,
      "loss": 0.6798,
      "step": 1274
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0002,
      "loss": 0.6259,
      "step": 1275
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0002,
      "loss": 0.2762,
      "step": 1276
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0002,
      "loss": 0.6748,
      "step": 1277
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0002,
      "loss": 0.4655,
      "step": 1278
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0002,
      "loss": 0.5338,
      "step": 1279
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0002,
      "loss": 0.317,
      "step": 1280
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0002,
      "loss": 0.623,
      "step": 1281
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0002,
      "loss": 0.2506,
      "step": 1282
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0002,
      "loss": 0.3568,
      "step": 1283
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0002,
      "loss": 0.3715,
      "step": 1284
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0002,
      "loss": 0.5987,
      "step": 1285
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0002,
      "loss": 0.5535,
      "step": 1286
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0002,
      "loss": 0.2838,
      "step": 1287
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0002,
      "loss": 0.5816,
      "step": 1288
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0002,
      "loss": 0.5041,
      "step": 1289
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0002,
      "loss": 0.561,
      "step": 1290
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0002,
      "loss": 0.5789,
      "step": 1291
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0002,
      "loss": 0.5387,
      "step": 1292
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 0.3089,
      "step": 1293
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 0.7878,
      "step": 1294
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 0.3375,
      "step": 1295
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 0.5434,
      "step": 1296
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 0.5912,
      "step": 1297
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 0.3783,
      "step": 1298
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 0.4357,
      "step": 1299
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0002,
      "loss": 0.6958,
      "step": 1300
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0002,
      "loss": 0.3629,
      "step": 1301
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0002,
      "loss": 0.7533,
      "step": 1302
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0002,
      "loss": 0.4568,
      "step": 1303
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0002,
      "loss": 0.6572,
      "step": 1304
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0002,
      "loss": 0.6041,
      "step": 1305
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0002,
      "loss": 0.2666,
      "step": 1306
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0002,
      "loss": 0.752,
      "step": 1307
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0002,
      "loss": 0.5149,
      "step": 1308
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0002,
      "loss": 0.3519,
      "step": 1309
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0002,
      "loss": 0.37,
      "step": 1310
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0002,
      "loss": 0.358,
      "step": 1311
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0002,
      "loss": 0.4981,
      "step": 1312
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0002,
      "loss": 0.8581,
      "step": 1313
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0002,
      "loss": 0.4593,
      "step": 1314
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0002,
      "loss": 0.3166,
      "step": 1315
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0002,
      "loss": 0.686,
      "step": 1316
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0002,
      "loss": 0.6256,
      "step": 1317
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0002,
      "loss": 0.3933,
      "step": 1318
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0002,
      "loss": 0.4853,
      "step": 1319
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0002,
      "loss": 0.4779,
      "step": 1320
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0002,
      "loss": 0.368,
      "step": 1321
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0002,
      "loss": 0.4096,
      "step": 1322
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0002,
      "loss": 0.5476,
      "step": 1323
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0002,
      "loss": 0.6451,
      "step": 1324
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0002,
      "loss": 0.4243,
      "step": 1325
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0002,
      "loss": 0.6374,
      "step": 1326
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0002,
      "loss": 0.8085,
      "step": 1327
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0002,
      "loss": 0.6558,
      "step": 1328
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0002,
      "loss": 0.5206,
      "step": 1329
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0002,
      "loss": 0.5061,
      "step": 1330
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0002,
      "loss": 0.468,
      "step": 1331
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0002,
      "loss": 0.6756,
      "step": 1332
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0002,
      "loss": 0.5527,
      "step": 1333
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0002,
      "loss": 0.4959,
      "step": 1334
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0002,
      "loss": 0.3878,
      "step": 1335
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0002,
      "loss": 0.4828,
      "step": 1336
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0002,
      "loss": 0.4019,
      "step": 1337
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0002,
      "loss": 0.6582,
      "step": 1338
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0002,
      "loss": 0.619,
      "step": 1339
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0002,
      "loss": 0.9632,
      "step": 1340
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0002,
      "loss": 0.3747,
      "step": 1341
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0002,
      "loss": 0.6371,
      "step": 1342
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 0.4066,
      "step": 1343
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 0.5924,
      "step": 1344
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 0.8573,
      "step": 1345
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 0.5788,
      "step": 1346
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 0.4295,
      "step": 1347
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 0.4233,
      "step": 1348
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 0.4692,
      "step": 1349
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0002,
      "loss": 0.7578,
      "step": 1350
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0002,
      "loss": 0.3574,
      "step": 1351
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0002,
      "loss": 0.5718,
      "step": 1352
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0002,
      "loss": 0.5106,
      "step": 1353
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0002,
      "loss": 0.4731,
      "step": 1354
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0002,
      "loss": 0.982,
      "step": 1355
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0002,
      "loss": 0.481,
      "step": 1356
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0002,
      "loss": 0.4094,
      "step": 1357
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0002,
      "loss": 0.313,
      "step": 1358
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0002,
      "loss": 0.2817,
      "step": 1359
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0002,
      "loss": 0.5999,
      "step": 1360
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0002,
      "loss": 0.4088,
      "step": 1361
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0002,
      "loss": 0.3724,
      "step": 1362
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0002,
      "loss": 0.5448,
      "step": 1363
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0002,
      "loss": 0.7124,
      "step": 1364
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0002,
      "loss": 0.4248,
      "step": 1365
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0002,
      "loss": 0.2958,
      "step": 1366
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0002,
      "loss": 0.5723,
      "step": 1367
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0002,
      "loss": 0.2449,
      "step": 1368
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0002,
      "loss": 0.48,
      "step": 1369
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0002,
      "loss": 0.1805,
      "step": 1370
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0002,
      "loss": 0.6565,
      "step": 1371
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0002,
      "loss": 0.3818,
      "step": 1372
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0002,
      "loss": 0.5324,
      "step": 1373
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0002,
      "loss": 0.4995,
      "step": 1374
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0002,
      "loss": 0.5294,
      "step": 1375
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0002,
      "loss": 0.2767,
      "step": 1376
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0002,
      "loss": 0.2938,
      "step": 1377
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0002,
      "loss": 0.3204,
      "step": 1378
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0002,
      "loss": 0.497,
      "step": 1379
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0002,
      "loss": 0.5519,
      "step": 1380
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0002,
      "loss": 0.308,
      "step": 1381
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0002,
      "loss": 0.4787,
      "step": 1382
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0002,
      "loss": 0.7684,
      "step": 1383
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0002,
      "loss": 0.2952,
      "step": 1384
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0002,
      "loss": 0.7964,
      "step": 1385
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0002,
      "loss": 0.4978,
      "step": 1386
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0002,
      "loss": 0.6111,
      "step": 1387
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0002,
      "loss": 0.9022,
      "step": 1388
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0002,
      "loss": 0.5483,
      "step": 1389
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0002,
      "loss": 0.3768,
      "step": 1390
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0002,
      "loss": 0.5542,
      "step": 1391
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0002,
      "loss": 0.6502,
      "step": 1392
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0002,
      "loss": 0.5545,
      "step": 1393
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0002,
      "loss": 0.7618,
      "step": 1394
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0002,
      "loss": 0.5697,
      "step": 1395
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0002,
      "loss": 0.3605,
      "step": 1396
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0002,
      "loss": 0.4917,
      "step": 1397
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0002,
      "loss": 0.4284,
      "step": 1398
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0002,
      "loss": 0.5859,
      "step": 1399
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 0.4323,
      "step": 1400
    },
    {
      "epoch": 1.97,
      "eval_loss": 0.551896870136261,
      "eval_runtime": 398.5526,
      "eval_samples_per_second": 2.509,
      "eval_steps_per_second": 1.255,
      "step": 1400
    },
    {
      "epoch": 1.97,
      "mmlu_eval_accuracy": 0.47302293981613547,
      "mmlu_eval_accuracy_abstract_algebra": 0.45454545454545453,
      "mmlu_eval_accuracy_anatomy": 0.5714285714285714,
      "mmlu_eval_accuracy_astronomy": 0.375,
      "mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
      "mmlu_eval_accuracy_clinical_knowledge": 0.4482758620689655,
      "mmlu_eval_accuracy_college_biology": 0.375,
      "mmlu_eval_accuracy_college_chemistry": 0.0,
      "mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
      "mmlu_eval_accuracy_college_mathematics": 0.2727272727272727,
      "mmlu_eval_accuracy_college_medicine": 0.36363636363636365,
      "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
      "mmlu_eval_accuracy_computer_security": 0.45454545454545453,
      "mmlu_eval_accuracy_conceptual_physics": 0.34615384615384615,
      "mmlu_eval_accuracy_econometrics": 0.3333333333333333,
      "mmlu_eval_accuracy_electrical_engineering": 0.25,
      "mmlu_eval_accuracy_elementary_mathematics": 0.3902439024390244,
      "mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
      "mmlu_eval_accuracy_global_facts": 0.3,
      "mmlu_eval_accuracy_high_school_biology": 0.3125,
      "mmlu_eval_accuracy_high_school_chemistry": 0.22727272727272727,
      "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_european_history": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_geography": 0.7272727272727273,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.6190476190476191,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.4186046511627907,
      "mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.5,
      "mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
      "mmlu_eval_accuracy_high_school_psychology": 0.7833333333333333,
      "mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
      "mmlu_eval_accuracy_high_school_us_history": 0.5909090909090909,
      "mmlu_eval_accuracy_high_school_world_history": 0.46153846153846156,
      "mmlu_eval_accuracy_human_aging": 0.6086956521739131,
      "mmlu_eval_accuracy_human_sexuality": 0.16666666666666666,
      "mmlu_eval_accuracy_international_law": 0.7692307692307693,
      "mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
      "mmlu_eval_accuracy_logical_fallacies": 0.6111111111111112,
      "mmlu_eval_accuracy_machine_learning": 0.5454545454545454,
      "mmlu_eval_accuracy_management": 0.8181818181818182,
      "mmlu_eval_accuracy_marketing": 0.88,
      "mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
      "mmlu_eval_accuracy_miscellaneous": 0.627906976744186,
      "mmlu_eval_accuracy_moral_disputes": 0.47368421052631576,
      "mmlu_eval_accuracy_moral_scenarios": 0.29,
      "mmlu_eval_accuracy_nutrition": 0.6666666666666666,
      "mmlu_eval_accuracy_philosophy": 0.5882352941176471,
      "mmlu_eval_accuracy_prehistory": 0.4,
      "mmlu_eval_accuracy_professional_accounting": 0.3225806451612903,
      "mmlu_eval_accuracy_professional_law": 0.36470588235294116,
      "mmlu_eval_accuracy_professional_medicine": 0.5161290322580645,
      "mmlu_eval_accuracy_professional_psychology": 0.4492753623188406,
      "mmlu_eval_accuracy_public_relations": 0.4166666666666667,
      "mmlu_eval_accuracy_security_studies": 0.48148148148148145,
      "mmlu_eval_accuracy_sociology": 0.6818181818181818,
      "mmlu_eval_accuracy_us_foreign_policy": 0.5454545454545454,
      "mmlu_eval_accuracy_virology": 0.5,
      "mmlu_eval_accuracy_world_religions": 0.5789473684210527,
      "mmlu_loss": 0.7568482326552778,
      "step": 1400
    }
  ],
  "max_steps": 5000,
  "num_train_epochs": 8,
  "total_flos": 7.574037752863457e+17,
  "trial_name": null,
  "trial_params": null
}