{
  "best_metric": 0.7577090859413147,
  "best_model_checkpoint": "experts/mistralic-expert-15/checkpoint-200",
  "epoch": 0.03286095707537482,
  "eval_steps": 200,
  "global_step": 200,
  "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": 0.8743,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.0512,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.0398,
      "step": 3
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8428,
      "step": 4
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8539,
      "step": 5
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6822,
      "step": 6
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7817,
      "step": 7
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8251,
      "step": 8
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.641,
      "step": 9
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9213,
      "step": 10
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8051,
      "step": 11
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9494,
      "step": 12
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7953,
      "step": 13
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6879,
      "step": 14
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7866,
      "step": 15
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7557,
      "step": 16
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6147,
      "step": 17
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7999,
      "step": 18
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9288,
      "step": 19
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6545,
      "step": 20
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6837,
      "step": 21
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8564,
      "step": 22
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6611,
      "step": 23
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6956,
      "step": 24
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.5079,
      "step": 25
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6142,
      "step": 26
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7226,
      "step": 27
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7037,
      "step": 28
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8852,
      "step": 29
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7209,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8991,
      "step": 31
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5839,
      "step": 32
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6147,
      "step": 33
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9075,
      "step": 34
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6618,
      "step": 35
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8367,
      "step": 36
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7012,
      "step": 37
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9249,
      "step": 38
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5889,
      "step": 39
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8997,
      "step": 40
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7283,
      "step": 41
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8314,
      "step": 42
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7005,
      "step": 43
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5693,
      "step": 44
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8334,
      "step": 45
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9775,
      "step": 46
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7298,
      "step": 47
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9334,
      "step": 48
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7225,
      "step": 49
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8088,
      "step": 50
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7466,
      "step": 51
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.4508,
      "step": 52
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.4574,
      "step": 53
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6886,
      "step": 54
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7615,
      "step": 55
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5989,
      "step": 56
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0519,
      "step": 57
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9321,
      "step": 58
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6666,
      "step": 59
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5781,
      "step": 60
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9561,
      "step": 61
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5952,
      "step": 62
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7159,
      "step": 63
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6358,
      "step": 64
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.4376,
      "step": 65
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6991,
      "step": 66
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6598,
      "step": 67
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7651,
      "step": 68
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7757,
      "step": 69
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7273,
      "step": 70
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8523,
      "step": 71
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6573,
      "step": 72
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5618,
      "step": 73
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6906,
      "step": 74
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7202,
      "step": 75
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8629,
      "step": 76
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6166,
      "step": 77
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8685,
      "step": 78
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8057,
      "step": 79
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8424,
      "step": 80
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9356,
      "step": 81
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8011,
      "step": 82
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5423,
      "step": 83
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6254,
      "step": 84
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7161,
      "step": 85
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6061,
      "step": 86
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5743,
      "step": 87
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.609,
      "step": 88
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7292,
      "step": 89
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9335,
      "step": 90
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.678,
      "step": 91
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7919,
      "step": 92
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6743,
      "step": 93
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9325,
      "step": 94
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6499,
      "step": 95
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5631,
      "step": 96
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8618,
      "step": 97
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.683,
      "step": 98
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.635,
      "step": 99
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7615,
      "step": 100
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8628,
      "step": 101
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7456,
      "step": 102
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8235,
      "step": 103
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.916,
      "step": 104
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0372,
      "step": 105
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9372,
      "step": 106
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6709,
      "step": 107
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.1999,
      "step": 108
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7315,
      "step": 109
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7147,
      "step": 110
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6933,
      "step": 111
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.844,
      "step": 112
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7172,
      "step": 113
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7477,
      "step": 114
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.835,
      "step": 115
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0017,
      "step": 116
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5665,
      "step": 117
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6992,
      "step": 118
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9693,
      "step": 119
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.574,
      "step": 120
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8181,
      "step": 121
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6321,
      "step": 122
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.85,
      "step": 123
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6986,
      "step": 124
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7242,
      "step": 125
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7909,
      "step": 126
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7143,
      "step": 127
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5493,
      "step": 128
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7625,
      "step": 129
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7199,
      "step": 130
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.3807,
      "step": 131
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.322,
      "step": 132
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.582,
      "step": 133
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.575,
      "step": 134
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8918,
      "step": 135
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.1352,
      "step": 136
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9177,
      "step": 137
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9293,
      "step": 138
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5509,
      "step": 139
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9554,
      "step": 140
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5094,
      "step": 141
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5929,
      "step": 142
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9733,
      "step": 143
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5117,
      "step": 144
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6024,
      "step": 145
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6401,
      "step": 146
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.924,
      "step": 147
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9399,
      "step": 148
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.972,
      "step": 149
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6407,
      "step": 150
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8665,
      "step": 151
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7225,
      "step": 152
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6445,
      "step": 153
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7775,
      "step": 154
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7517,
      "step": 155
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9464,
      "step": 156
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5912,
      "step": 157
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6161,
      "step": 158
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8201,
      "step": 159
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6348,
      "step": 160
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0841,
      "step": 161
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8405,
      "step": 162
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9317,
      "step": 163
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7158,
      "step": 164
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9286,
      "step": 165
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9918,
      "step": 166
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9909,
      "step": 167
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7513,
      "step": 168
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8818,
      "step": 169
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8644,
      "step": 170
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7351,
      "step": 171
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7574,
      "step": 172
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6465,
      "step": 173
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7679,
      "step": 174
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8849,
      "step": 175
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0791,
      "step": 176
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8741,
      "step": 177
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7822,
      "step": 178
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8417,
      "step": 179
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7317,
      "step": 180
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0245,
      "step": 181
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6746,
      "step": 182
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6059,
      "step": 183
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9795,
      "step": 184
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9011,
      "step": 185
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6384,
      "step": 186
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7801,
      "step": 187
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.81,
      "step": 188
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8018,
      "step": 189
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6842,
      "step": 190
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7468,
      "step": 191
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8598,
      "step": 192
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9457,
      "step": 193
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5614,
      "step": 194
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6668,
      "step": 195
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8057,
      "step": 196
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6429,
      "step": 197
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6506,
      "step": 198
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8127,
      "step": 199
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9115,
      "step": 200
    },
    {
      "epoch": 0.03,
      "eval_loss": 0.7577090859413147,
      "eval_runtime": 131.6982,
      "eval_samples_per_second": 7.593,
      "eval_steps_per_second": 3.797,
      "step": 200
    },
    {
      "epoch": 0.03,
      "mmlu_eval_accuracy": 0.5965101027344765,
      "mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
      "mmlu_eval_accuracy_anatomy": 0.5,
      "mmlu_eval_accuracy_astronomy": 0.75,
      "mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
      "mmlu_eval_accuracy_clinical_knowledge": 0.6206896551724138,
      "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": 0.36363636363636365,
      "mmlu_eval_accuracy_college_medicine": 0.6363636363636364,
      "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
      "mmlu_eval_accuracy_computer_security": 0.7272727272727273,
      "mmlu_eval_accuracy_conceptual_physics": 0.5,
      "mmlu_eval_accuracy_econometrics": 0.5,
      "mmlu_eval_accuracy_electrical_engineering": 0.5625,
      "mmlu_eval_accuracy_elementary_mathematics": 0.4878048780487805,
      "mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
      "mmlu_eval_accuracy_global_facts": 0.3,
      "mmlu_eval_accuracy_high_school_biology": 0.5625,
      "mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
      "mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
      "mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
      "mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.6511627906976745,
      "mmlu_eval_accuracy_high_school_mathematics": 0.20689655172413793,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.6153846153846154,
      "mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
      "mmlu_eval_accuracy_high_school_psychology": 0.8333333333333334,
      "mmlu_eval_accuracy_high_school_statistics": 0.43478260869565216,
      "mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
      "mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
      "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": 0.6363636363636364,
      "mmlu_eval_accuracy_logical_fallacies": 0.7222222222222222,
      "mmlu_eval_accuracy_machine_learning": 0.5454545454545454,
      "mmlu_eval_accuracy_management": 0.9090909090909091,
      "mmlu_eval_accuracy_marketing": 0.88,
      "mmlu_eval_accuracy_medical_genetics": 1.0,
      "mmlu_eval_accuracy_miscellaneous": 0.7558139534883721,
      "mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
      "mmlu_eval_accuracy_moral_scenarios": 0.34,
      "mmlu_eval_accuracy_nutrition": 0.7272727272727273,
      "mmlu_eval_accuracy_philosophy": 0.7647058823529411,
      "mmlu_eval_accuracy_prehistory": 0.6,
      "mmlu_eval_accuracy_professional_accounting": 0.5806451612903226,
      "mmlu_eval_accuracy_professional_law": 0.4,
      "mmlu_eval_accuracy_professional_medicine": 0.5806451612903226,
      "mmlu_eval_accuracy_professional_psychology": 0.6086956521739131,
      "mmlu_eval_accuracy_public_relations": 0.5,
      "mmlu_eval_accuracy_security_studies": 0.6666666666666666,
      "mmlu_eval_accuracy_sociology": 0.9090909090909091,
      "mmlu_eval_accuracy_us_foreign_policy": 0.9090909090909091,
      "mmlu_eval_accuracy_virology": 0.5,
      "mmlu_eval_accuracy_world_religions": 0.8421052631578947,
      "mmlu_loss": 1.1431999449708132,
      "step": 200
    }
  ],
  "logging_steps": 1,
  "max_steps": 18258,
  "num_train_epochs": 3,
  "save_steps": 200,
  "total_flos": 4.117576204320768e+16,
  "trial_name": null,
  "trial_params": null
}