{
  "best_metric": 0.9129472374916077,
  "best_model_checkpoint": "experts/mistralic-expert-18/checkpoint-200",
  "epoch": 0.059608076894419194,
  "eval_steps": 200,
  "global_step": 400,
  "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": 0.8806,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8314,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9094,
      "step": 3
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9642,
      "step": 4
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.0596,
      "step": 5
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.2488,
      "step": 6
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.0475,
      "step": 7
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9753,
      "step": 8
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.0299,
      "step": 9
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.1919,
      "step": 10
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9262,
      "step": 11
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.1809,
      "step": 12
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8364,
      "step": 13
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9607,
      "step": 14
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9229,
      "step": 15
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9485,
      "step": 16
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9056,
      "step": 17
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8531,
      "step": 18
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8962,
      "step": 19
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.016,
      "step": 20
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9926,
      "step": 21
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8967,
      "step": 22
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.1022,
      "step": 23
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8736,
      "step": 24
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.1224,
      "step": 25
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7499,
      "step": 26
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.838,
      "step": 27
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.1863,
      "step": 28
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.1471,
      "step": 29
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7989,
      "step": 30
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.0997,
      "step": 31
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8971,
      "step": 32
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7339,
      "step": 33
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.1044,
      "step": 34
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.1162,
      "step": 35
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9012,
      "step": 36
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9764,
      "step": 37
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8379,
      "step": 38
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.859,
      "step": 39
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7498,
      "step": 40
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7806,
      "step": 41
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8827,
      "step": 42
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8112,
      "step": 43
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.785,
      "step": 44
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9393,
      "step": 45
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.953,
      "step": 46
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6196,
      "step": 47
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.807,
      "step": 48
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9234,
      "step": 49
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8608,
      "step": 50
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8236,
      "step": 51
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7865,
      "step": 52
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7982,
      "step": 53
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8308,
      "step": 54
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8724,
      "step": 55
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9934,
      "step": 56
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9431,
      "step": 57
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8924,
      "step": 58
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9541,
      "step": 59
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9559,
      "step": 60
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6952,
      "step": 61
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7242,
      "step": 62
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0335,
      "step": 63
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9005,
      "step": 64
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.2233,
      "step": 65
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8545,
      "step": 66
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0107,
      "step": 67
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0311,
      "step": 68
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8858,
      "step": 69
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.1565,
      "step": 70
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.849,
      "step": 71
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7856,
      "step": 72
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9566,
      "step": 73
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9554,
      "step": 74
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.1567,
      "step": 75
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9642,
      "step": 76
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.961,
      "step": 77
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0184,
      "step": 78
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9995,
      "step": 79
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8882,
      "step": 80
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9038,
      "step": 81
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.2235,
      "step": 82
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9006,
      "step": 83
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9148,
      "step": 84
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0327,
      "step": 85
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.697,
      "step": 86
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9014,
      "step": 87
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8204,
      "step": 88
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.714,
      "step": 89
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.87,
      "step": 90
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8035,
      "step": 91
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9082,
      "step": 92
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0317,
      "step": 93
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0429,
      "step": 94
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7844,
      "step": 95
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.1747,
      "step": 96
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9387,
      "step": 97
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6686,
      "step": 98
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8755,
      "step": 99
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0841,
      "step": 100
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8801,
      "step": 101
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9329,
      "step": 102
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.193,
      "step": 103
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8709,
      "step": 104
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9117,
      "step": 105
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5743,
      "step": 106
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8605,
      "step": 107
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0642,
      "step": 108
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9255,
      "step": 109
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9245,
      "step": 110
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0698,
      "step": 111
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.687,
      "step": 112
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8338,
      "step": 113
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8229,
      "step": 114
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9505,
      "step": 115
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7175,
      "step": 116
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7723,
      "step": 117
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8191,
      "step": 118
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.237,
      "step": 119
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7911,
      "step": 120
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9756,
      "step": 121
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.061,
      "step": 122
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8589,
      "step": 123
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8144,
      "step": 124
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5933,
      "step": 125
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9877,
      "step": 126
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7459,
      "step": 127
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0653,
      "step": 128
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8961,
      "step": 129
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8666,
      "step": 130
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9499,
      "step": 131
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9341,
      "step": 132
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5623,
      "step": 133
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8098,
      "step": 134
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6733,
      "step": 135
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.043,
      "step": 136
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0224,
      "step": 137
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.015,
      "step": 138
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8273,
      "step": 139
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6916,
      "step": 140
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7127,
      "step": 141
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.909,
      "step": 142
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0009,
      "step": 143
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0373,
      "step": 144
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8498,
      "step": 145
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.1059,
      "step": 146
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0539,
      "step": 147
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8912,
      "step": 148
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9306,
      "step": 149
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6859,
      "step": 150
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.069,
      "step": 151
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0221,
      "step": 152
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6557,
      "step": 153
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.1022,
      "step": 154
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8669,
      "step": 155
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.05,
      "step": 156
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5794,
      "step": 157
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7873,
      "step": 158
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9018,
      "step": 159
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0947,
      "step": 160
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7592,
      "step": 161
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5523,
      "step": 162
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8171,
      "step": 163
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9278,
      "step": 164
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9385,
      "step": 165
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9932,
      "step": 166
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6516,
      "step": 167
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9681,
      "step": 168
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7592,
      "step": 169
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0066,
      "step": 170
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.98,
      "step": 171
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7544,
      "step": 172
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.4535,
      "step": 173
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.1337,
      "step": 174
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8768,
      "step": 175
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6215,
      "step": 176
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9664,
      "step": 177
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8782,
      "step": 178
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6777,
      "step": 179
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9082,
      "step": 180
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.1479,
      "step": 181
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0668,
      "step": 182
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7604,
      "step": 183
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7257,
      "step": 184
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8318,
      "step": 185
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.1311,
      "step": 186
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9256,
      "step": 187
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.777,
      "step": 188
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8604,
      "step": 189
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0517,
      "step": 190
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0416,
      "step": 191
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0257,
      "step": 192
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8573,
      "step": 193
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7623,
      "step": 194
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9769,
      "step": 195
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0114,
      "step": 196
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9433,
      "step": 197
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8473,
      "step": 198
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.1912,
      "step": 199
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6054,
      "step": 200
    },
    {
      "epoch": 0.03,
      "eval_loss": 0.9129472374916077,
      "eval_runtime": 176.508,
      "eval_samples_per_second": 5.665,
      "eval_steps_per_second": 2.833,
      "step": 200
    },
    {
      "epoch": 0.03,
      "mmlu_eval_accuracy": 0.5941889078502712,
      "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
      "mmlu_eval_accuracy_anatomy": 0.5714285714285714,
      "mmlu_eval_accuracy_astronomy": 0.8125,
      "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
      "mmlu_eval_accuracy_clinical_knowledge": 0.5862068965517241,
      "mmlu_eval_accuracy_college_biology": 0.6875,
      "mmlu_eval_accuracy_college_chemistry": 0.125,
      "mmlu_eval_accuracy_college_computer_science": 0.09090909090909091,
      "mmlu_eval_accuracy_college_mathematics": 0.36363636363636365,
      "mmlu_eval_accuracy_college_medicine": 0.6818181818181818,
      "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
      "mmlu_eval_accuracy_computer_security": 0.6363636363636364,
      "mmlu_eval_accuracy_conceptual_physics": 0.5384615384615384,
      "mmlu_eval_accuracy_econometrics": 0.5833333333333334,
      "mmlu_eval_accuracy_electrical_engineering": 0.625,
      "mmlu_eval_accuracy_elementary_mathematics": 0.36585365853658536,
      "mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
      "mmlu_eval_accuracy_global_facts": 0.4,
      "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": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_geography": 0.7727272727272727,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.627906976744186,
      "mmlu_eval_accuracy_high_school_mathematics": 0.27586206896551724,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.6538461538461539,
      "mmlu_eval_accuracy_high_school_physics": 0.23529411764705882,
      "mmlu_eval_accuracy_high_school_psychology": 0.8666666666666667,
      "mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
      "mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
      "mmlu_eval_accuracy_high_school_world_history": 0.6153846153846154,
      "mmlu_eval_accuracy_human_aging": 0.7391304347826086,
      "mmlu_eval_accuracy_human_sexuality": 0.5833333333333334,
      "mmlu_eval_accuracy_international_law": 0.9230769230769231,
      "mmlu_eval_accuracy_jurisprudence": 0.5454545454545454,
      "mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
      "mmlu_eval_accuracy_machine_learning": 0.7272727272727273,
      "mmlu_eval_accuracy_management": 0.9090909090909091,
      "mmlu_eval_accuracy_marketing": 0.88,
      "mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
      "mmlu_eval_accuracy_miscellaneous": 0.686046511627907,
      "mmlu_eval_accuracy_moral_disputes": 0.5789473684210527,
      "mmlu_eval_accuracy_moral_scenarios": 0.35,
      "mmlu_eval_accuracy_nutrition": 0.7878787878787878,
      "mmlu_eval_accuracy_philosophy": 0.7647058823529411,
      "mmlu_eval_accuracy_prehistory": 0.6,
      "mmlu_eval_accuracy_professional_accounting": 0.6129032258064516,
      "mmlu_eval_accuracy_professional_law": 0.36470588235294116,
      "mmlu_eval_accuracy_professional_medicine": 0.5483870967741935,
      "mmlu_eval_accuracy_professional_psychology": 0.5797101449275363,
      "mmlu_eval_accuracy_public_relations": 0.4166666666666667,
      "mmlu_eval_accuracy_security_studies": 0.7037037037037037,
      "mmlu_eval_accuracy_sociology": 0.9090909090909091,
      "mmlu_eval_accuracy_us_foreign_policy": 0.9090909090909091,
      "mmlu_eval_accuracy_virology": 0.5555555555555556,
      "mmlu_eval_accuracy_world_religions": 0.7894736842105263,
      "mmlu_loss": 1.2055549576450577,
      "step": 200
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8703,
      "step": 201
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7398,
      "step": 202
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8275,
      "step": 203
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8609,
      "step": 204
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8893,
      "step": 205
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8933,
      "step": 206
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8941,
      "step": 207
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8109,
      "step": 208
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9929,
      "step": 209
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7378,
      "step": 210
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.2906,
      "step": 211
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0529,
      "step": 212
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8577,
      "step": 213
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0375,
      "step": 214
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.2407,
      "step": 215
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0471,
      "step": 216
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9455,
      "step": 217
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9334,
      "step": 218
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9502,
      "step": 219
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0056,
      "step": 220
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9083,
      "step": 221
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0184,
      "step": 222
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8285,
      "step": 223
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0216,
      "step": 224
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0441,
      "step": 225
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0516,
      "step": 226
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.646,
      "step": 227
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8357,
      "step": 228
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9219,
      "step": 229
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7118,
      "step": 230
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0509,
      "step": 231
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.108,
      "step": 232
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8324,
      "step": 233
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9095,
      "step": 234
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.2458,
      "step": 235
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8568,
      "step": 236
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.3849,
      "step": 237
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6962,
      "step": 238
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9087,
      "step": 239
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9527,
      "step": 240
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8144,
      "step": 241
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8777,
      "step": 242
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6796,
      "step": 243
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.1459,
      "step": 244
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7961,
      "step": 245
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0438,
      "step": 246
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8611,
      "step": 247
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9473,
      "step": 248
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7572,
      "step": 249
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0568,
      "step": 250
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0067,
      "step": 251
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7068,
      "step": 252
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9839,
      "step": 253
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9511,
      "step": 254
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8583,
      "step": 255
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9134,
      "step": 256
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.013,
      "step": 257
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8776,
      "step": 258
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8282,
      "step": 259
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9655,
      "step": 260
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0753,
      "step": 261
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8411,
      "step": 262
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0246,
      "step": 263
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8236,
      "step": 264
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6594,
      "step": 265
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0485,
      "step": 266
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9754,
      "step": 267
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8255,
      "step": 268
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.1425,
      "step": 269
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7942,
      "step": 270
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6286,
      "step": 271
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8562,
      "step": 272
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8846,
      "step": 273
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8653,
      "step": 274
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9637,
      "step": 275
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0031,
      "step": 276
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9643,
      "step": 277
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.14,
      "step": 278
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7588,
      "step": 279
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6927,
      "step": 280
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0515,
      "step": 281
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7698,
      "step": 282
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.067,
      "step": 283
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8873,
      "step": 284
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7697,
      "step": 285
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9438,
      "step": 286
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8233,
      "step": 287
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8528,
      "step": 288
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8835,
      "step": 289
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.1808,
      "step": 290
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0047,
      "step": 291
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0198,
      "step": 292
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.2443,
      "step": 293
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8848,
      "step": 294
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8259,
      "step": 295
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6864,
      "step": 296
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.4134,
      "step": 297
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.4718,
      "step": 298
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9678,
      "step": 299
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8294,
      "step": 300
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8161,
      "step": 301
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7389,
      "step": 302
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9217,
      "step": 303
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0356,
      "step": 304
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.966,
      "step": 305
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9029,
      "step": 306
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9289,
      "step": 307
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8617,
      "step": 308
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8378,
      "step": 309
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9235,
      "step": 310
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.899,
      "step": 311
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9035,
      "step": 312
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7387,
      "step": 313
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9732,
      "step": 314
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8665,
      "step": 315
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6973,
      "step": 316
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.2004,
      "step": 317
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9254,
      "step": 318
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8353,
      "step": 319
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7832,
      "step": 320
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9201,
      "step": 321
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8728,
      "step": 322
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8538,
      "step": 323
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9381,
      "step": 324
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.2398,
      "step": 325
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9402,
      "step": 326
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6942,
      "step": 327
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7964,
      "step": 328
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7974,
      "step": 329
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7256,
      "step": 330
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8479,
      "step": 331
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.3347,
      "step": 332
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8153,
      "step": 333
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8961,
      "step": 334
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8019,
      "step": 335
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9081,
      "step": 336
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6419,
      "step": 337
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.1492,
      "step": 338
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9803,
      "step": 339
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7756,
      "step": 340
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8267,
      "step": 341
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8829,
      "step": 342
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7999,
      "step": 343
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6872,
      "step": 344
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6968,
      "step": 345
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8884,
      "step": 346
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7844,
      "step": 347
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7488,
      "step": 348
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0249,
      "step": 349
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9262,
      "step": 350
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9092,
      "step": 351
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9186,
      "step": 352
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9488,
      "step": 353
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8253,
      "step": 354
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6747,
      "step": 355
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.1798,
      "step": 356
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.046,
      "step": 357
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0411,
      "step": 358
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0301,
      "step": 359
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8327,
      "step": 360
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8115,
      "step": 361
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.852,
      "step": 362
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9171,
      "step": 363
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8708,
      "step": 364
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6221,
      "step": 365
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0739,
      "step": 366
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0021,
      "step": 367
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0984,
      "step": 368
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6944,
      "step": 369
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9001,
      "step": 370
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8393,
      "step": 371
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8986,
      "step": 372
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8556,
      "step": 373
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9434,
      "step": 374
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9566,
      "step": 375
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8219,
      "step": 376
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8845,
      "step": 377
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9489,
      "step": 378
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8542,
      "step": 379
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9531,
      "step": 380
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8932,
      "step": 381
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8073,
      "step": 382
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.1058,
      "step": 383
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0112,
      "step": 384
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9445,
      "step": 385
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9967,
      "step": 386
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9992,
      "step": 387
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0076,
      "step": 388
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.5664,
      "step": 389
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8862,
      "step": 390
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.095,
      "step": 391
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9463,
      "step": 392
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.516,
      "step": 393
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.514,
      "step": 394
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8276,
      "step": 395
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7418,
      "step": 396
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7304,
      "step": 397
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7052,
      "step": 398
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8999,
      "step": 399
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.5368,
      "step": 400
    },
    {
      "epoch": 0.06,
      "eval_loss": 0.9132314324378967,
      "eval_runtime": 176.4034,
      "eval_samples_per_second": 5.669,
      "eval_steps_per_second": 2.834,
      "step": 400
    },
    {
      "epoch": 0.06,
      "mmlu_eval_accuracy": 0.5441688522823004,
      "mmlu_eval_accuracy_abstract_algebra": 0.18181818181818182,
      "mmlu_eval_accuracy_anatomy": 0.42857142857142855,
      "mmlu_eval_accuracy_astronomy": 0.625,
      "mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
      "mmlu_eval_accuracy_clinical_knowledge": 0.5862068965517241,
      "mmlu_eval_accuracy_college_biology": 0.625,
      "mmlu_eval_accuracy_college_chemistry": 0.125,
      "mmlu_eval_accuracy_college_computer_science": 0.09090909090909091,
      "mmlu_eval_accuracy_college_mathematics": 0.18181818181818182,
      "mmlu_eval_accuracy_college_medicine": 0.6818181818181818,
      "mmlu_eval_accuracy_college_physics": 0.5454545454545454,
      "mmlu_eval_accuracy_computer_security": 0.7272727272727273,
      "mmlu_eval_accuracy_conceptual_physics": 0.46153846153846156,
      "mmlu_eval_accuracy_econometrics": 0.5,
      "mmlu_eval_accuracy_electrical_engineering": 0.3125,
      "mmlu_eval_accuracy_elementary_mathematics": 0.3170731707317073,
      "mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
      "mmlu_eval_accuracy_global_facts": 0.4,
      "mmlu_eval_accuracy_high_school_biology": 0.625,
      "mmlu_eval_accuracy_high_school_chemistry": 0.3181818181818182,
      "mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
      "mmlu_eval_accuracy_high_school_european_history": 0.6111111111111112,
      "mmlu_eval_accuracy_high_school_geography": 0.7727272727272727,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.4883720930232558,
      "mmlu_eval_accuracy_high_school_mathematics": 0.10344827586206896,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.5769230769230769,
      "mmlu_eval_accuracy_high_school_physics": 0.29411764705882354,
      "mmlu_eval_accuracy_high_school_psychology": 0.8333333333333334,
      "mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
      "mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
      "mmlu_eval_accuracy_high_school_world_history": 0.7307692307692307,
      "mmlu_eval_accuracy_human_aging": 0.7391304347826086,
      "mmlu_eval_accuracy_human_sexuality": 0.5,
      "mmlu_eval_accuracy_international_law": 1.0,
      "mmlu_eval_accuracy_jurisprudence": 0.5454545454545454,
      "mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
      "mmlu_eval_accuracy_machine_learning": 0.36363636363636365,
      "mmlu_eval_accuracy_management": 0.8181818181818182,
      "mmlu_eval_accuracy_marketing": 0.84,
      "mmlu_eval_accuracy_medical_genetics": 0.8181818181818182,
      "mmlu_eval_accuracy_miscellaneous": 0.7325581395348837,
      "mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
      "mmlu_eval_accuracy_moral_scenarios": 0.3,
      "mmlu_eval_accuracy_nutrition": 0.7272727272727273,
      "mmlu_eval_accuracy_philosophy": 0.6470588235294118,
      "mmlu_eval_accuracy_prehistory": 0.7428571428571429,
      "mmlu_eval_accuracy_professional_accounting": 0.45161290322580644,
      "mmlu_eval_accuracy_professional_law": 0.3941176470588235,
      "mmlu_eval_accuracy_professional_medicine": 0.4838709677419355,
      "mmlu_eval_accuracy_professional_psychology": 0.5072463768115942,
      "mmlu_eval_accuracy_public_relations": 0.5833333333333334,
      "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": 0.7368421052631579,
      "mmlu_loss": 1.1913223482143163,
      "step": 400
    }
  ],
  "logging_steps": 1,
  "max_steps": 20130,
  "num_train_epochs": 3,
  "save_steps": 200,
  "total_flos": 1.1350949708326502e+17,
  "trial_name": null,
  "trial_params": null
}