{
  "best_metric": 0.7622222304344177,
  "best_model_checkpoint": "experts/mistralic-expert-4/checkpoint-200",
  "epoch": 0.0894304398859762,
  "eval_steps": 200,
  "global_step": 800,
  "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": 0.8836,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9068,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.5267,
      "step": 3
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8124,
      "step": 4
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.5854,
      "step": 5
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.83,
      "step": 6
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.0319,
      "step": 7
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.5087,
      "step": 8
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9476,
      "step": 9
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8989,
      "step": 10
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.2456,
      "step": 11
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7858,
      "step": 12
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.678,
      "step": 13
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8737,
      "step": 14
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.2075,
      "step": 15
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9291,
      "step": 16
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7075,
      "step": 17
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7562,
      "step": 18
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.0055,
      "step": 19
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9101,
      "step": 20
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.725,
      "step": 21
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.5941,
      "step": 22
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.0636,
      "step": 23
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8149,
      "step": 24
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.1276,
      "step": 25
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.896,
      "step": 26
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9211,
      "step": 27
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.728,
      "step": 28
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.5779,
      "step": 29
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.5838,
      "step": 30
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8043,
      "step": 31
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7809,
      "step": 32
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.5959,
      "step": 33
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.0109,
      "step": 34
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7994,
      "step": 35
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8555,
      "step": 36
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6883,
      "step": 37
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.5677,
      "step": 38
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.087,
      "step": 39
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.2933,
      "step": 40
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8544,
      "step": 41
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8421,
      "step": 42
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7714,
      "step": 43
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9672,
      "step": 44
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.4563,
      "step": 45
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6979,
      "step": 46
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6694,
      "step": 47
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0412,
      "step": 48
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9186,
      "step": 49
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5598,
      "step": 50
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.893,
      "step": 51
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8274,
      "step": 52
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5906,
      "step": 53
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6503,
      "step": 54
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8578,
      "step": 55
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6271,
      "step": 56
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.4822,
      "step": 57
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.4955,
      "step": 58
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8042,
      "step": 59
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.1222,
      "step": 60
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0774,
      "step": 61
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0726,
      "step": 62
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.4331,
      "step": 63
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.3862,
      "step": 64
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9063,
      "step": 65
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6699,
      "step": 66
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6879,
      "step": 67
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7378,
      "step": 68
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7149,
      "step": 69
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9253,
      "step": 70
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.1755,
      "step": 71
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8339,
      "step": 72
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9048,
      "step": 73
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5383,
      "step": 74
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9042,
      "step": 75
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5539,
      "step": 76
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5496,
      "step": 77
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9809,
      "step": 78
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7466,
      "step": 79
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7701,
      "step": 80
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7614,
      "step": 81
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5188,
      "step": 82
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8464,
      "step": 83
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9676,
      "step": 84
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8974,
      "step": 85
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7232,
      "step": 86
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.54,
      "step": 87
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6168,
      "step": 88
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.2568,
      "step": 89
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8187,
      "step": 90
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9073,
      "step": 91
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8315,
      "step": 92
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9093,
      "step": 93
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.709,
      "step": 94
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0193,
      "step": 95
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7218,
      "step": 96
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9009,
      "step": 97
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5433,
      "step": 98
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0513,
      "step": 99
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8954,
      "step": 100
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6808,
      "step": 101
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0333,
      "step": 102
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6295,
      "step": 103
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.4441,
      "step": 104
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0074,
      "step": 105
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0852,
      "step": 106
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7527,
      "step": 107
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7105,
      "step": 108
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8738,
      "step": 109
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.3121,
      "step": 110
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.3928,
      "step": 111
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9031,
      "step": 112
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.4751,
      "step": 113
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5742,
      "step": 114
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9761,
      "step": 115
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8362,
      "step": 116
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8315,
      "step": 117
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6396,
      "step": 118
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6226,
      "step": 119
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8304,
      "step": 120
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6335,
      "step": 121
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7912,
      "step": 122
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7265,
      "step": 123
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9238,
      "step": 124
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.055,
      "step": 125
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6545,
      "step": 126
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9552,
      "step": 127
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7312,
      "step": 128
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.599,
      "step": 129
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9601,
      "step": 130
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6988,
      "step": 131
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9885,
      "step": 132
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5337,
      "step": 133
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7537,
      "step": 134
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7324,
      "step": 135
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8686,
      "step": 136
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8591,
      "step": 137
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7551,
      "step": 138
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9782,
      "step": 139
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.716,
      "step": 140
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6863,
      "step": 141
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.498,
      "step": 142
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.1123,
      "step": 143
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8176,
      "step": 144
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0194,
      "step": 145
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5848,
      "step": 146
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6897,
      "step": 147
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.699,
      "step": 148
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7334,
      "step": 149
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6729,
      "step": 150
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8162,
      "step": 151
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7443,
      "step": 152
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5254,
      "step": 153
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.1428,
      "step": 154
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6402,
      "step": 155
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.754,
      "step": 156
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.025,
      "step": 157
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7081,
      "step": 158
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9099,
      "step": 159
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9155,
      "step": 160
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8386,
      "step": 161
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0644,
      "step": 162
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8238,
      "step": 163
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.613,
      "step": 164
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7745,
      "step": 165
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6221,
      "step": 166
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.4419,
      "step": 167
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9317,
      "step": 168
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.4534,
      "step": 169
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6721,
      "step": 170
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7752,
      "step": 171
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7753,
      "step": 172
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7841,
      "step": 173
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7654,
      "step": 174
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7331,
      "step": 175
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8391,
      "step": 176
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6729,
      "step": 177
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7802,
      "step": 178
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9578,
      "step": 179
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.798,
      "step": 180
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5288,
      "step": 181
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.2338,
      "step": 182
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7594,
      "step": 183
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9577,
      "step": 184
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6787,
      "step": 185
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6455,
      "step": 186
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0321,
      "step": 187
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8171,
      "step": 188
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7211,
      "step": 189
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9733,
      "step": 190
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0872,
      "step": 191
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.2343,
      "step": 192
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7669,
      "step": 193
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7834,
      "step": 194
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.4672,
      "step": 195
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.958,
      "step": 196
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.4482,
      "step": 197
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7993,
      "step": 198
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0381,
      "step": 199
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9136,
      "step": 200
    },
    {
      "epoch": 0.02,
      "eval_loss": 0.7622222304344177,
      "eval_runtime": 135.897,
      "eval_samples_per_second": 7.359,
      "eval_steps_per_second": 3.679,
      "step": 200
    },
    {
      "epoch": 0.02,
      "mmlu_eval_accuracy": 0.5672989056036994,
      "mmlu_eval_accuracy_abstract_algebra": 0.18181818181818182,
      "mmlu_eval_accuracy_anatomy": 0.5714285714285714,
      "mmlu_eval_accuracy_astronomy": 0.6875,
      "mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
      "mmlu_eval_accuracy_clinical_knowledge": 0.5517241379310345,
      "mmlu_eval_accuracy_college_biology": 0.5625,
      "mmlu_eval_accuracy_college_chemistry": 0.25,
      "mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
      "mmlu_eval_accuracy_college_mathematics": 0.36363636363636365,
      "mmlu_eval_accuracy_college_medicine": 0.5909090909090909,
      "mmlu_eval_accuracy_college_physics": 0.36363636363636365,
      "mmlu_eval_accuracy_computer_security": 0.45454545454545453,
      "mmlu_eval_accuracy_conceptual_physics": 0.46153846153846156,
      "mmlu_eval_accuracy_econometrics": 0.6666666666666666,
      "mmlu_eval_accuracy_electrical_engineering": 0.625,
      "mmlu_eval_accuracy_elementary_mathematics": 0.2926829268292683,
      "mmlu_eval_accuracy_formal_logic": 0.07142857142857142,
      "mmlu_eval_accuracy_global_facts": 0.5,
      "mmlu_eval_accuracy_high_school_biology": 0.59375,
      "mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
      "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
      "mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.5348837209302325,
      "mmlu_eval_accuracy_high_school_mathematics": 0.4482758620689655,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.46153846153846156,
      "mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
      "mmlu_eval_accuracy_high_school_psychology": 0.85,
      "mmlu_eval_accuracy_high_school_statistics": 0.30434782608695654,
      "mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
      "mmlu_eval_accuracy_high_school_world_history": 0.7307692307692307,
      "mmlu_eval_accuracy_human_aging": 0.7391304347826086,
      "mmlu_eval_accuracy_human_sexuality": 0.5,
      "mmlu_eval_accuracy_international_law": 0.8461538461538461,
      "mmlu_eval_accuracy_jurisprudence": 0.7272727272727273,
      "mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
      "mmlu_eval_accuracy_machine_learning": 0.2727272727272727,
      "mmlu_eval_accuracy_management": 0.9090909090909091,
      "mmlu_eval_accuracy_marketing": 0.88,
      "mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
      "mmlu_eval_accuracy_miscellaneous": 0.7325581395348837,
      "mmlu_eval_accuracy_moral_disputes": 0.5789473684210527,
      "mmlu_eval_accuracy_moral_scenarios": 0.27,
      "mmlu_eval_accuracy_nutrition": 0.6060606060606061,
      "mmlu_eval_accuracy_philosophy": 0.5882352941176471,
      "mmlu_eval_accuracy_prehistory": 0.5142857142857142,
      "mmlu_eval_accuracy_professional_accounting": 0.4838709677419355,
      "mmlu_eval_accuracy_professional_law": 0.38235294117647056,
      "mmlu_eval_accuracy_professional_medicine": 0.6451612903225806,
      "mmlu_eval_accuracy_professional_psychology": 0.5797101449275363,
      "mmlu_eval_accuracy_public_relations": 0.4166666666666667,
      "mmlu_eval_accuracy_security_studies": 0.5555555555555556,
      "mmlu_eval_accuracy_sociology": 0.8636363636363636,
      "mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
      "mmlu_eval_accuracy_virology": 0.5,
      "mmlu_eval_accuracy_world_religions": 0.8947368421052632,
      "mmlu_loss": 1.0069586229230965,
      "step": 200
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7457,
      "step": 201
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7409,
      "step": 202
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.4745,
      "step": 203
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7993,
      "step": 204
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8555,
      "step": 205
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0264,
      "step": 206
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0129,
      "step": 207
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6486,
      "step": 208
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6592,
      "step": 209
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.2253,
      "step": 210
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.2569,
      "step": 211
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5429,
      "step": 212
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.4723,
      "step": 213
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5976,
      "step": 214
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9262,
      "step": 215
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9181,
      "step": 216
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.4718,
      "step": 217
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0941,
      "step": 218
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9409,
      "step": 219
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7036,
      "step": 220
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8129,
      "step": 221
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9511,
      "step": 222
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8059,
      "step": 223
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5923,
      "step": 224
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6317,
      "step": 225
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7596,
      "step": 226
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.356,
      "step": 227
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6839,
      "step": 228
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7192,
      "step": 229
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7333,
      "step": 230
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6442,
      "step": 231
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9199,
      "step": 232
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0557,
      "step": 233
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0344,
      "step": 234
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8674,
      "step": 235
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7501,
      "step": 236
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9032,
      "step": 237
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8459,
      "step": 238
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7982,
      "step": 239
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5675,
      "step": 240
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7399,
      "step": 241
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.4875,
      "step": 242
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9398,
      "step": 243
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5853,
      "step": 244
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6324,
      "step": 245
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7644,
      "step": 246
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6404,
      "step": 247
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6217,
      "step": 248
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.745,
      "step": 249
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7691,
      "step": 250
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9053,
      "step": 251
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0101,
      "step": 252
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9381,
      "step": 253
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0886,
      "step": 254
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7413,
      "step": 255
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7652,
      "step": 256
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8463,
      "step": 257
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0698,
      "step": 258
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9377,
      "step": 259
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.903,
      "step": 260
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7064,
      "step": 261
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.992,
      "step": 262
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8681,
      "step": 263
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.2352,
      "step": 264
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.979,
      "step": 265
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7337,
      "step": 266
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.3681,
      "step": 267
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8407,
      "step": 268
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6767,
      "step": 269
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7463,
      "step": 270
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0491,
      "step": 271
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8992,
      "step": 272
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7818,
      "step": 273
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6515,
      "step": 274
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7413,
      "step": 275
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6952,
      "step": 276
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6319,
      "step": 277
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7387,
      "step": 278
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5565,
      "step": 279
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7401,
      "step": 280
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.1749,
      "step": 281
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7653,
      "step": 282
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8655,
      "step": 283
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7638,
      "step": 284
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6062,
      "step": 285
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.4701,
      "step": 286
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8371,
      "step": 287
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8219,
      "step": 288
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6089,
      "step": 289
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.541,
      "step": 290
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7023,
      "step": 291
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9798,
      "step": 292
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8186,
      "step": 293
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.3499,
      "step": 294
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6607,
      "step": 295
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6573,
      "step": 296
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0429,
      "step": 297
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.4692,
      "step": 298
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8158,
      "step": 299
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5983,
      "step": 300
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.149,
      "step": 301
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8579,
      "step": 302
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7426,
      "step": 303
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8366,
      "step": 304
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0115,
      "step": 305
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.542,
      "step": 306
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8848,
      "step": 307
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.655,
      "step": 308
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9583,
      "step": 309
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5102,
      "step": 310
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.4342,
      "step": 311
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5187,
      "step": 312
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8127,
      "step": 313
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6748,
      "step": 314
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0238,
      "step": 315
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8009,
      "step": 316
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8022,
      "step": 317
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9432,
      "step": 318
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7482,
      "step": 319
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7513,
      "step": 320
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.814,
      "step": 321
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.1215,
      "step": 322
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6266,
      "step": 323
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.946,
      "step": 324
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.922,
      "step": 325
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8452,
      "step": 326
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8053,
      "step": 327
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6888,
      "step": 328
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7282,
      "step": 329
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6211,
      "step": 330
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0571,
      "step": 331
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6081,
      "step": 332
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8934,
      "step": 333
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6127,
      "step": 334
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5351,
      "step": 335
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7114,
      "step": 336
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7449,
      "step": 337
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.901,
      "step": 338
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0089,
      "step": 339
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9088,
      "step": 340
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6527,
      "step": 341
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.4576,
      "step": 342
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6783,
      "step": 343
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0807,
      "step": 344
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9711,
      "step": 345
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.773,
      "step": 346
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.1481,
      "step": 347
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7278,
      "step": 348
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8854,
      "step": 349
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5881,
      "step": 350
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5783,
      "step": 351
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5469,
      "step": 352
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0707,
      "step": 353
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9178,
      "step": 354
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8676,
      "step": 355
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.2651,
      "step": 356
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.3317,
      "step": 357
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9464,
      "step": 358
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.4135,
      "step": 359
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8236,
      "step": 360
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6812,
      "step": 361
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6053,
      "step": 362
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6821,
      "step": 363
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9269,
      "step": 364
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0339,
      "step": 365
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8589,
      "step": 366
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7864,
      "step": 367
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.3367,
      "step": 368
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5649,
      "step": 369
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8276,
      "step": 370
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9586,
      "step": 371
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.4708,
      "step": 372
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7661,
      "step": 373
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8253,
      "step": 374
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9087,
      "step": 375
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0339,
      "step": 376
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8213,
      "step": 377
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8447,
      "step": 378
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7706,
      "step": 379
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8645,
      "step": 380
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6842,
      "step": 381
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5427,
      "step": 382
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8191,
      "step": 383
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7911,
      "step": 384
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8604,
      "step": 385
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8688,
      "step": 386
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9298,
      "step": 387
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9718,
      "step": 388
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5817,
      "step": 389
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7653,
      "step": 390
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7,
      "step": 391
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0167,
      "step": 392
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8315,
      "step": 393
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9008,
      "step": 394
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9822,
      "step": 395
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0273,
      "step": 396
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6305,
      "step": 397
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5161,
      "step": 398
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7984,
      "step": 399
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5132,
      "step": 400
    },
    {
      "epoch": 0.04,
      "eval_loss": 0.7738097906112671,
      "eval_runtime": 135.3891,
      "eval_samples_per_second": 7.386,
      "eval_steps_per_second": 3.693,
      "step": 400
    },
    {
      "epoch": 0.04,
      "mmlu_eval_accuracy": 0.5629561210475079,
      "mmlu_eval_accuracy_abstract_algebra": 0.18181818181818182,
      "mmlu_eval_accuracy_anatomy": 0.5,
      "mmlu_eval_accuracy_astronomy": 0.625,
      "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
      "mmlu_eval_accuracy_clinical_knowledge": 0.5172413793103449,
      "mmlu_eval_accuracy_college_biology": 0.5625,
      "mmlu_eval_accuracy_college_chemistry": 0.375,
      "mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
      "mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
      "mmlu_eval_accuracy_college_medicine": 0.6363636363636364,
      "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
      "mmlu_eval_accuracy_computer_security": 0.45454545454545453,
      "mmlu_eval_accuracy_conceptual_physics": 0.46153846153846156,
      "mmlu_eval_accuracy_econometrics": 0.5833333333333334,
      "mmlu_eval_accuracy_electrical_engineering": 0.5625,
      "mmlu_eval_accuracy_elementary_mathematics": 0.34146341463414637,
      "mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
      "mmlu_eval_accuracy_global_facts": 0.4,
      "mmlu_eval_accuracy_high_school_biology": 0.59375,
      "mmlu_eval_accuracy_high_school_chemistry": 0.36363636363636365,
      "mmlu_eval_accuracy_high_school_computer_science": 0.7777777777777778,
      "mmlu_eval_accuracy_high_school_european_history": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_geography": 0.7727272727272727,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.5813953488372093,
      "mmlu_eval_accuracy_high_school_mathematics": 0.41379310344827586,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.5,
      "mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
      "mmlu_eval_accuracy_high_school_psychology": 0.85,
      "mmlu_eval_accuracy_high_school_statistics": 0.4782608695652174,
      "mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
      "mmlu_eval_accuracy_high_school_world_history": 0.6153846153846154,
      "mmlu_eval_accuracy_human_aging": 0.7391304347826086,
      "mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
      "mmlu_eval_accuracy_international_law": 0.7692307692307693,
      "mmlu_eval_accuracy_jurisprudence": 0.6363636363636364,
      "mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
      "mmlu_eval_accuracy_machine_learning": 0.2727272727272727,
      "mmlu_eval_accuracy_management": 0.9090909090909091,
      "mmlu_eval_accuracy_marketing": 0.88,
      "mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
      "mmlu_eval_accuracy_miscellaneous": 0.7674418604651163,
      "mmlu_eval_accuracy_moral_disputes": 0.5,
      "mmlu_eval_accuracy_moral_scenarios": 0.24,
      "mmlu_eval_accuracy_nutrition": 0.5757575757575758,
      "mmlu_eval_accuracy_philosophy": 0.7352941176470589,
      "mmlu_eval_accuracy_prehistory": 0.45714285714285713,
      "mmlu_eval_accuracy_professional_accounting": 0.5806451612903226,
      "mmlu_eval_accuracy_professional_law": 0.37058823529411766,
      "mmlu_eval_accuracy_professional_medicine": 0.7096774193548387,
      "mmlu_eval_accuracy_professional_psychology": 0.5942028985507246,
      "mmlu_eval_accuracy_public_relations": 0.4166666666666667,
      "mmlu_eval_accuracy_security_studies": 0.48148148148148145,
      "mmlu_eval_accuracy_sociology": 0.8181818181818182,
      "mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
      "mmlu_eval_accuracy_virology": 0.4444444444444444,
      "mmlu_eval_accuracy_world_religions": 0.7894736842105263,
      "mmlu_loss": 1.259490801697456,
      "step": 400
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6662,
      "step": 401
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8725,
      "step": 402
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.842,
      "step": 403
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5504,
      "step": 404
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9595,
      "step": 405
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7878,
      "step": 406
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8974,
      "step": 407
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.617,
      "step": 408
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8067,
      "step": 409
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8149,
      "step": 410
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5254,
      "step": 411
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7172,
      "step": 412
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7117,
      "step": 413
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7014,
      "step": 414
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0684,
      "step": 415
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9571,
      "step": 416
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6536,
      "step": 417
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6513,
      "step": 418
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5801,
      "step": 419
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0074,
      "step": 420
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7686,
      "step": 421
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5249,
      "step": 422
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.4267,
      "step": 423
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8294,
      "step": 424
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9329,
      "step": 425
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.294,
      "step": 426
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.659,
      "step": 427
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5661,
      "step": 428
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6495,
      "step": 429
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.022,
      "step": 430
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9567,
      "step": 431
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9376,
      "step": 432
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.865,
      "step": 433
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8335,
      "step": 434
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.3547,
      "step": 435
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7468,
      "step": 436
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.838,
      "step": 437
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9447,
      "step": 438
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9341,
      "step": 439
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7219,
      "step": 440
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.478,
      "step": 441
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.1492,
      "step": 442
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8927,
      "step": 443
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6739,
      "step": 444
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8425,
      "step": 445
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8269,
      "step": 446
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9041,
      "step": 447
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8533,
      "step": 448
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.2088,
      "step": 449
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.562,
      "step": 450
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5463,
      "step": 451
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.726,
      "step": 452
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8446,
      "step": 453
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6027,
      "step": 454
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9695,
      "step": 455
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7165,
      "step": 456
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9498,
      "step": 457
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9708,
      "step": 458
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.1096,
      "step": 459
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5941,
      "step": 460
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9342,
      "step": 461
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6778,
      "step": 462
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7208,
      "step": 463
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7434,
      "step": 464
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7407,
      "step": 465
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.967,
      "step": 466
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5082,
      "step": 467
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7873,
      "step": 468
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9815,
      "step": 469
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5141,
      "step": 470
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6792,
      "step": 471
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7229,
      "step": 472
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5889,
      "step": 473
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.829,
      "step": 474
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.378,
      "step": 475
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9447,
      "step": 476
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5353,
      "step": 477
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7108,
      "step": 478
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.83,
      "step": 479
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6203,
      "step": 480
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7407,
      "step": 481
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7342,
      "step": 482
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0649,
      "step": 483
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8858,
      "step": 484
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7047,
      "step": 485
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7517,
      "step": 486
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8535,
      "step": 487
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6518,
      "step": 488
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7292,
      "step": 489
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6117,
      "step": 490
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.4378,
      "step": 491
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.722,
      "step": 492
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6505,
      "step": 493
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8827,
      "step": 494
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.5225,
      "step": 495
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9396,
      "step": 496
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6725,
      "step": 497
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6589,
      "step": 498
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.4633,
      "step": 499
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8317,
      "step": 500
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8208,
      "step": 501
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.2557,
      "step": 502
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.2034,
      "step": 503
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8563,
      "step": 504
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.444,
      "step": 505
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.929,
      "step": 506
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9689,
      "step": 507
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6122,
      "step": 508
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6736,
      "step": 509
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7997,
      "step": 510
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8722,
      "step": 511
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.3435,
      "step": 512
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8396,
      "step": 513
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7301,
      "step": 514
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6976,
      "step": 515
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6821,
      "step": 516
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.4544,
      "step": 517
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9472,
      "step": 518
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7855,
      "step": 519
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.963,
      "step": 520
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.5704,
      "step": 521
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8195,
      "step": 522
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0542,
      "step": 523
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8061,
      "step": 524
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9321,
      "step": 525
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.1324,
      "step": 526
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7763,
      "step": 527
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.3291,
      "step": 528
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7649,
      "step": 529
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9375,
      "step": 530
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.5123,
      "step": 531
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9329,
      "step": 532
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9119,
      "step": 533
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9945,
      "step": 534
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8338,
      "step": 535
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6888,
      "step": 536
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7925,
      "step": 537
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6036,
      "step": 538
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6694,
      "step": 539
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8962,
      "step": 540
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.5671,
      "step": 541
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.1594,
      "step": 542
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.3611,
      "step": 543
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0575,
      "step": 544
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6943,
      "step": 545
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8019,
      "step": 546
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.1535,
      "step": 547
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.95,
      "step": 548
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.1333,
      "step": 549
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9413,
      "step": 550
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7724,
      "step": 551
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9963,
      "step": 552
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9392,
      "step": 553
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.479,
      "step": 554
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.954,
      "step": 555
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8864,
      "step": 556
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6997,
      "step": 557
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.409,
      "step": 558
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8893,
      "step": 559
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0069,
      "step": 560
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.021,
      "step": 561
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.604,
      "step": 562
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.4664,
      "step": 563
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.465,
      "step": 564
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.5493,
      "step": 565
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6715,
      "step": 566
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7808,
      "step": 567
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8187,
      "step": 568
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.4409,
      "step": 569
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.5817,
      "step": 570
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8494,
      "step": 571
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.4059,
      "step": 572
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.5549,
      "step": 573
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.619,
      "step": 574
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7614,
      "step": 575
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.574,
      "step": 576
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.3832,
      "step": 577
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0101,
      "step": 578
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7803,
      "step": 579
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8475,
      "step": 580
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6601,
      "step": 581
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.0982,
      "step": 582
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5729,
      "step": 583
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.921,
      "step": 584
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7795,
      "step": 585
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8713,
      "step": 586
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8538,
      "step": 587
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5769,
      "step": 588
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9782,
      "step": 589
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.1549,
      "step": 590
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7921,
      "step": 591
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.4679,
      "step": 592
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7376,
      "step": 593
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.3555,
      "step": 594
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.2566,
      "step": 595
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7228,
      "step": 596
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.3707,
      "step": 597
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7439,
      "step": 598
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.0021,
      "step": 599
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7533,
      "step": 600
    },
    {
      "epoch": 0.07,
      "eval_loss": 0.7750256061553955,
      "eval_runtime": 135.4384,
      "eval_samples_per_second": 7.383,
      "eval_steps_per_second": 3.692,
      "step": 600
    },
    {
      "epoch": 0.07,
      "mmlu_eval_accuracy": 0.5783145240167142,
      "mmlu_eval_accuracy_abstract_algebra": 0.18181818181818182,
      "mmlu_eval_accuracy_anatomy": 0.5714285714285714,
      "mmlu_eval_accuracy_astronomy": 0.75,
      "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
      "mmlu_eval_accuracy_clinical_knowledge": 0.5172413793103449,
      "mmlu_eval_accuracy_college_biology": 0.6875,
      "mmlu_eval_accuracy_college_chemistry": 0.375,
      "mmlu_eval_accuracy_college_computer_science": 0.2727272727272727,
      "mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
      "mmlu_eval_accuracy_college_medicine": 0.6363636363636364,
      "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
      "mmlu_eval_accuracy_computer_security": 0.36363636363636365,
      "mmlu_eval_accuracy_conceptual_physics": 0.5384615384615384,
      "mmlu_eval_accuracy_econometrics": 0.4166666666666667,
      "mmlu_eval_accuracy_electrical_engineering": 0.5,
      "mmlu_eval_accuracy_elementary_mathematics": 0.4146341463414634,
      "mmlu_eval_accuracy_formal_logic": 0.07142857142857142,
      "mmlu_eval_accuracy_global_facts": 0.4,
      "mmlu_eval_accuracy_high_school_biology": 0.5625,
      "mmlu_eval_accuracy_high_school_chemistry": 0.3181818181818182,
      "mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
      "mmlu_eval_accuracy_high_school_european_history": 0.7222222222222222,
      "mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.6190476190476191,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.5813953488372093,
      "mmlu_eval_accuracy_high_school_mathematics": 0.3793103448275862,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.5,
      "mmlu_eval_accuracy_high_school_physics": 0.29411764705882354,
      "mmlu_eval_accuracy_high_school_psychology": 0.8,
      "mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
      "mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
      "mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
      "mmlu_eval_accuracy_human_aging": 0.8260869565217391,
      "mmlu_eval_accuracy_human_sexuality": 0.8333333333333334,
      "mmlu_eval_accuracy_international_law": 0.8461538461538461,
      "mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
      "mmlu_eval_accuracy_logical_fallacies": 0.7222222222222222,
      "mmlu_eval_accuracy_machine_learning": 0.5454545454545454,
      "mmlu_eval_accuracy_management": 0.6363636363636364,
      "mmlu_eval_accuracy_marketing": 0.8,
      "mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
      "mmlu_eval_accuracy_miscellaneous": 0.7441860465116279,
      "mmlu_eval_accuracy_moral_disputes": 0.5789473684210527,
      "mmlu_eval_accuracy_moral_scenarios": 0.33,
      "mmlu_eval_accuracy_nutrition": 0.7575757575757576,
      "mmlu_eval_accuracy_philosophy": 0.6470588235294118,
      "mmlu_eval_accuracy_prehistory": 0.6571428571428571,
      "mmlu_eval_accuracy_professional_accounting": 0.5483870967741935,
      "mmlu_eval_accuracy_professional_law": 0.3764705882352941,
      "mmlu_eval_accuracy_professional_medicine": 0.5806451612903226,
      "mmlu_eval_accuracy_professional_psychology": 0.5507246376811594,
      "mmlu_eval_accuracy_public_relations": 0.4166666666666667,
      "mmlu_eval_accuracy_security_studies": 0.6296296296296297,
      "mmlu_eval_accuracy_sociology": 0.9090909090909091,
      "mmlu_eval_accuracy_us_foreign_policy": 0.9090909090909091,
      "mmlu_eval_accuracy_virology": 0.6111111111111112,
      "mmlu_eval_accuracy_world_religions": 0.8947368421052632,
      "mmlu_loss": 0.9808103412623194,
      "step": 600
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6714,
      "step": 601
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5294,
      "step": 602
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7082,
      "step": 603
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6699,
      "step": 604
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8466,
      "step": 605
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.0917,
      "step": 606
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6288,
      "step": 607
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7341,
      "step": 608
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8254,
      "step": 609
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5146,
      "step": 610
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6109,
      "step": 611
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6973,
      "step": 612
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.575,
      "step": 613
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8646,
      "step": 614
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9915,
      "step": 615
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.4296,
      "step": 616
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6696,
      "step": 617
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5158,
      "step": 618
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5821,
      "step": 619
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8987,
      "step": 620
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5515,
      "step": 621
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5665,
      "step": 622
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6828,
      "step": 623
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6333,
      "step": 624
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.1513,
      "step": 625
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6114,
      "step": 626
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.3815,
      "step": 627
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6126,
      "step": 628
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5456,
      "step": 629
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.827,
      "step": 630
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6445,
      "step": 631
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.703,
      "step": 632
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7183,
      "step": 633
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5367,
      "step": 634
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6124,
      "step": 635
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5082,
      "step": 636
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9187,
      "step": 637
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8237,
      "step": 638
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7072,
      "step": 639
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8225,
      "step": 640
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.0131,
      "step": 641
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9881,
      "step": 642
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.0252,
      "step": 643
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8965,
      "step": 644
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9925,
      "step": 645
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8711,
      "step": 646
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5546,
      "step": 647
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6453,
      "step": 648
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.0059,
      "step": 649
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.617,
      "step": 650
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7537,
      "step": 651
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.0014,
      "step": 652
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6775,
      "step": 653
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8886,
      "step": 654
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.1783,
      "step": 655
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7332,
      "step": 656
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7638,
      "step": 657
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7362,
      "step": 658
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8912,
      "step": 659
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7011,
      "step": 660
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5246,
      "step": 661
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.2745,
      "step": 662
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6328,
      "step": 663
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8244,
      "step": 664
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8167,
      "step": 665
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.3225,
      "step": 666
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8041,
      "step": 667
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.0157,
      "step": 668
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5485,
      "step": 669
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6974,
      "step": 670
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6497,
      "step": 671
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9076,
      "step": 672
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.563,
      "step": 673
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6542,
      "step": 674
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.5499,
      "step": 675
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8955,
      "step": 676
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.133,
      "step": 677
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.1901,
      "step": 678
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6697,
      "step": 679
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8408,
      "step": 680
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8049,
      "step": 681
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9487,
      "step": 682
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.5676,
      "step": 683
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6208,
      "step": 684
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7667,
      "step": 685
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7008,
      "step": 686
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.0697,
      "step": 687
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.4353,
      "step": 688
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.0255,
      "step": 689
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8203,
      "step": 690
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.611,
      "step": 691
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7725,
      "step": 692
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9597,
      "step": 693
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6575,
      "step": 694
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.962,
      "step": 695
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8652,
      "step": 696
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.4212,
      "step": 697
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.3521,
      "step": 698
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6675,
      "step": 699
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.0333,
      "step": 700
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6589,
      "step": 701
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7,
      "step": 702
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9109,
      "step": 703
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.925,
      "step": 704
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9993,
      "step": 705
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8407,
      "step": 706
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.0704,
      "step": 707
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.1131,
      "step": 708
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8542,
      "step": 709
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6608,
      "step": 710
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7028,
      "step": 711
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9139,
      "step": 712
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6866,
      "step": 713
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9173,
      "step": 714
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.956,
      "step": 715
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7077,
      "step": 716
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8455,
      "step": 717
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.5813,
      "step": 718
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7635,
      "step": 719
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8366,
      "step": 720
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8558,
      "step": 721
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8632,
      "step": 722
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.0833,
      "step": 723
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6886,
      "step": 724
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6994,
      "step": 725
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7585,
      "step": 726
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6604,
      "step": 727
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.131,
      "step": 728
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.4719,
      "step": 729
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.5937,
      "step": 730
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9576,
      "step": 731
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8273,
      "step": 732
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.0008,
      "step": 733
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6292,
      "step": 734
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.4883,
      "step": 735
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8437,
      "step": 736
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7538,
      "step": 737
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.5859,
      "step": 738
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.4288,
      "step": 739
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.4325,
      "step": 740
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.5306,
      "step": 741
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6267,
      "step": 742
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8932,
      "step": 743
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8786,
      "step": 744
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8611,
      "step": 745
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.4142,
      "step": 746
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7675,
      "step": 747
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8481,
      "step": 748
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.3033,
      "step": 749
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8763,
      "step": 750
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6534,
      "step": 751
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7804,
      "step": 752
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8118,
      "step": 753
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.5996,
      "step": 754
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.362,
      "step": 755
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7265,
      "step": 756
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8745,
      "step": 757
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.621,
      "step": 758
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.5144,
      "step": 759
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.4028,
      "step": 760
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6001,
      "step": 761
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.874,
      "step": 762
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.4725,
      "step": 763
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 1.3119,
      "step": 764
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6592,
      "step": 765
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.8258,
      "step": 766
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6995,
      "step": 767
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 1.0119,
      "step": 768
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7032,
      "step": 769
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 1.1674,
      "step": 770
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5645,
      "step": 771
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5349,
      "step": 772
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 1.2056,
      "step": 773
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6296,
      "step": 774
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 1.0401,
      "step": 775
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6246,
      "step": 776
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7492,
      "step": 777
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5515,
      "step": 778
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.9114,
      "step": 779
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 1.1123,
      "step": 780
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.767,
      "step": 781
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.787,
      "step": 782
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7121,
      "step": 783
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5778,
      "step": 784
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.4818,
      "step": 785
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.9512,
      "step": 786
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.9971,
      "step": 787
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6708,
      "step": 788
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.922,
      "step": 789
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.8661,
      "step": 790
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.9268,
      "step": 791
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6024,
      "step": 792
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5103,
      "step": 793
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6788,
      "step": 794
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.411,
      "step": 795
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.8877,
      "step": 796
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.9004,
      "step": 797
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.9761,
      "step": 798
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.8112,
      "step": 799
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7737,
      "step": 800
    },
    {
      "epoch": 0.09,
      "eval_loss": 0.7735617756843567,
      "eval_runtime": 135.4598,
      "eval_samples_per_second": 7.382,
      "eval_steps_per_second": 3.691,
      "step": 800
    },
    {
      "epoch": 0.09,
      "mmlu_eval_accuracy": 0.5663508152380243,
      "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
      "mmlu_eval_accuracy_anatomy": 0.5,
      "mmlu_eval_accuracy_astronomy": 0.625,
      "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
      "mmlu_eval_accuracy_clinical_knowledge": 0.5517241379310345,
      "mmlu_eval_accuracy_college_biology": 0.625,
      "mmlu_eval_accuracy_college_chemistry": 0.25,
      "mmlu_eval_accuracy_college_computer_science": 0.0,
      "mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
      "mmlu_eval_accuracy_college_medicine": 0.5454545454545454,
      "mmlu_eval_accuracy_college_physics": 0.36363636363636365,
      "mmlu_eval_accuracy_computer_security": 0.45454545454545453,
      "mmlu_eval_accuracy_conceptual_physics": 0.46153846153846156,
      "mmlu_eval_accuracy_econometrics": 0.5,
      "mmlu_eval_accuracy_electrical_engineering": 0.4375,
      "mmlu_eval_accuracy_elementary_mathematics": 0.3170731707317073,
      "mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
      "mmlu_eval_accuracy_global_facts": 0.3,
      "mmlu_eval_accuracy_high_school_biology": 0.625,
      "mmlu_eval_accuracy_high_school_chemistry": 0.36363636363636365,
      "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
      "mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.6511627906976745,
      "mmlu_eval_accuracy_high_school_mathematics": 0.4482758620689655,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.5769230769230769,
      "mmlu_eval_accuracy_high_school_physics": 0.23529411764705882,
      "mmlu_eval_accuracy_high_school_psychology": 0.8166666666666667,
      "mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
      "mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
      "mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
      "mmlu_eval_accuracy_human_aging": 0.8260869565217391,
      "mmlu_eval_accuracy_human_sexuality": 0.6666666666666666,
      "mmlu_eval_accuracy_international_law": 1.0,
      "mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
      "mmlu_eval_accuracy_logical_fallacies": 0.7222222222222222,
      "mmlu_eval_accuracy_machine_learning": 0.36363636363636365,
      "mmlu_eval_accuracy_management": 0.7272727272727273,
      "mmlu_eval_accuracy_marketing": 0.92,
      "mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
      "mmlu_eval_accuracy_miscellaneous": 0.7558139534883721,
      "mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
      "mmlu_eval_accuracy_moral_scenarios": 0.24,
      "mmlu_eval_accuracy_nutrition": 0.696969696969697,
      "mmlu_eval_accuracy_philosophy": 0.6764705882352942,
      "mmlu_eval_accuracy_prehistory": 0.5428571428571428,
      "mmlu_eval_accuracy_professional_accounting": 0.5483870967741935,
      "mmlu_eval_accuracy_professional_law": 0.3352941176470588,
      "mmlu_eval_accuracy_professional_medicine": 0.5806451612903226,
      "mmlu_eval_accuracy_professional_psychology": 0.6231884057971014,
      "mmlu_eval_accuracy_public_relations": 0.5,
      "mmlu_eval_accuracy_security_studies": 0.6296296296296297,
      "mmlu_eval_accuracy_sociology": 0.8181818181818182,
      "mmlu_eval_accuracy_us_foreign_policy": 0.7272727272727273,
      "mmlu_eval_accuracy_virology": 0.6666666666666666,
      "mmlu_eval_accuracy_world_religions": 0.8947368421052632,
      "mmlu_loss": 1.1613865079207457,
      "step": 800
    }
  ],
  "logging_steps": 1,
  "max_steps": 26835,
  "num_train_epochs": 3,
  "save_steps": 200,
  "total_flos": 1.618004336024617e+17,
  "trial_name": null,
  "trial_params": null
}