{
  "best_metric": 0.6861502528190613,
  "best_model_checkpoint": "experts/mistralic-expert-29/checkpoint-200",
  "epoch": 0.050498674409796745,
  "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": 0.8281,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6834,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6795,
      "step": 3
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.5884,
      "step": 4
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.836,
      "step": 5
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7988,
      "step": 6
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.531,
      "step": 7
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6883,
      "step": 8
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7717,
      "step": 9
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7416,
      "step": 10
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.5991,
      "step": 11
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6144,
      "step": 12
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.5785,
      "step": 13
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7473,
      "step": 14
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6916,
      "step": 15
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9093,
      "step": 16
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9202,
      "step": 17
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.5683,
      "step": 18
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6056,
      "step": 19
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8613,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7247,
      "step": 21
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5612,
      "step": 22
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6764,
      "step": 23
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8115,
      "step": 24
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5642,
      "step": 25
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5383,
      "step": 26
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5503,
      "step": 27
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6636,
      "step": 28
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7039,
      "step": 29
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7322,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6291,
      "step": 31
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6892,
      "step": 32
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8208,
      "step": 33
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6776,
      "step": 34
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5404,
      "step": 35
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6605,
      "step": 36
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8808,
      "step": 37
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7159,
      "step": 38
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6436,
      "step": 39
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5478,
      "step": 40
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.879,
      "step": 41
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8325,
      "step": 42
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7474,
      "step": 43
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6632,
      "step": 44
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6072,
      "step": 45
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7562,
      "step": 46
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6429,
      "step": 47
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6926,
      "step": 48
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7064,
      "step": 49
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7125,
      "step": 50
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6166,
      "step": 51
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5655,
      "step": 52
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7084,
      "step": 53
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6504,
      "step": 54
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6086,
      "step": 55
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8146,
      "step": 56
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.706,
      "step": 57
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6723,
      "step": 58
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.4581,
      "step": 59
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7367,
      "step": 60
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7673,
      "step": 61
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.513,
      "step": 62
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7892,
      "step": 63
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.662,
      "step": 64
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.656,
      "step": 65
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.559,
      "step": 66
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6614,
      "step": 67
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7893,
      "step": 68
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7972,
      "step": 69
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8093,
      "step": 70
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8171,
      "step": 71
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6491,
      "step": 72
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6915,
      "step": 73
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5994,
      "step": 74
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5994,
      "step": 75
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.608,
      "step": 76
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6353,
      "step": 77
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6389,
      "step": 78
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.539,
      "step": 79
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6736,
      "step": 80
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7245,
      "step": 81
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7079,
      "step": 82
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7158,
      "step": 83
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5509,
      "step": 84
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.688,
      "step": 85
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9429,
      "step": 86
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7074,
      "step": 87
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7848,
      "step": 88
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7461,
      "step": 89
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7644,
      "step": 90
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.893,
      "step": 91
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9111,
      "step": 92
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.729,
      "step": 93
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6529,
      "step": 94
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9489,
      "step": 95
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6918,
      "step": 96
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5863,
      "step": 97
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6953,
      "step": 98
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6269,
      "step": 99
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5336,
      "step": 100
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6729,
      "step": 101
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.57,
      "step": 102
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7521,
      "step": 103
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6486,
      "step": 104
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6212,
      "step": 105
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8345,
      "step": 106
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6985,
      "step": 107
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6532,
      "step": 108
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7224,
      "step": 109
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6173,
      "step": 110
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8925,
      "step": 111
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6129,
      "step": 112
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6461,
      "step": 113
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5614,
      "step": 114
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7075,
      "step": 115
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5555,
      "step": 116
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6579,
      "step": 117
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7032,
      "step": 118
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7109,
      "step": 119
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.631,
      "step": 120
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5974,
      "step": 121
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8038,
      "step": 122
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6748,
      "step": 123
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8089,
      "step": 124
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5831,
      "step": 125
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7566,
      "step": 126
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6766,
      "step": 127
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7793,
      "step": 128
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.638,
      "step": 129
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6085,
      "step": 130
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6366,
      "step": 131
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.918,
      "step": 132
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6851,
      "step": 133
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6124,
      "step": 134
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7642,
      "step": 135
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.642,
      "step": 136
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9157,
      "step": 137
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6619,
      "step": 138
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8664,
      "step": 139
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6881,
      "step": 140
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6909,
      "step": 141
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7242,
      "step": 142
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.635,
      "step": 143
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8164,
      "step": 144
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6169,
      "step": 145
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5874,
      "step": 146
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6332,
      "step": 147
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8494,
      "step": 148
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.834,
      "step": 149
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8598,
      "step": 150
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5545,
      "step": 151
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9171,
      "step": 152
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6941,
      "step": 153
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8018,
      "step": 154
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7894,
      "step": 155
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5925,
      "step": 156
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6051,
      "step": 157
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7172,
      "step": 158
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7902,
      "step": 159
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8397,
      "step": 160
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7131,
      "step": 161
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7729,
      "step": 162
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.604,
      "step": 163
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.682,
      "step": 164
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.641,
      "step": 165
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7199,
      "step": 166
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5337,
      "step": 167
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8882,
      "step": 168
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7981,
      "step": 169
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7438,
      "step": 170
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.586,
      "step": 171
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8469,
      "step": 172
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7154,
      "step": 173
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6153,
      "step": 174
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5517,
      "step": 175
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.951,
      "step": 176
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6128,
      "step": 177
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6679,
      "step": 178
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6546,
      "step": 179
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6633,
      "step": 180
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7509,
      "step": 181
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6667,
      "step": 182
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6744,
      "step": 183
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8333,
      "step": 184
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9516,
      "step": 185
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5293,
      "step": 186
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5694,
      "step": 187
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7939,
      "step": 188
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6257,
      "step": 189
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6859,
      "step": 190
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7746,
      "step": 191
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5474,
      "step": 192
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6079,
      "step": 193
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6545,
      "step": 194
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7829,
      "step": 195
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6969,
      "step": 196
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.713,
      "step": 197
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8589,
      "step": 198
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6964,
      "step": 199
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6997,
      "step": 200
    },
    {
      "epoch": 0.05,
      "eval_loss": 0.6861502528190613,
      "eval_runtime": 95.1356,
      "eval_samples_per_second": 10.511,
      "eval_steps_per_second": 5.256,
      "step": 200
    },
    {
      "epoch": 0.05,
      "mmlu_eval_accuracy": 0.5869545490402098,
      "mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
      "mmlu_eval_accuracy_anatomy": 0.5,
      "mmlu_eval_accuracy_astronomy": 0.8125,
      "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
      "mmlu_eval_accuracy_clinical_knowledge": 0.6206896551724138,
      "mmlu_eval_accuracy_college_biology": 0.625,
      "mmlu_eval_accuracy_college_chemistry": 0.25,
      "mmlu_eval_accuracy_college_computer_science": 0.18181818181818182,
      "mmlu_eval_accuracy_college_mathematics": 0.5454545454545454,
      "mmlu_eval_accuracy_college_medicine": 0.5909090909090909,
      "mmlu_eval_accuracy_college_physics": 0.5454545454545454,
      "mmlu_eval_accuracy_computer_security": 0.6363636363636364,
      "mmlu_eval_accuracy_conceptual_physics": 0.46153846153846156,
      "mmlu_eval_accuracy_econometrics": 0.5,
      "mmlu_eval_accuracy_electrical_engineering": 0.5625,
      "mmlu_eval_accuracy_elementary_mathematics": 0.4146341463414634,
      "mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
      "mmlu_eval_accuracy_global_facts": 0.4,
      "mmlu_eval_accuracy_high_school_biology": 0.59375,
      "mmlu_eval_accuracy_high_school_chemistry": 0.45454545454545453,
      "mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
      "mmlu_eval_accuracy_high_school_european_history": 0.7222222222222222,
      "mmlu_eval_accuracy_high_school_geography": 0.9090909090909091,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.6511627906976745,
      "mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.6153846153846154,
      "mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
      "mmlu_eval_accuracy_high_school_psychology": 0.8333333333333334,
      "mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
      "mmlu_eval_accuracy_high_school_us_history": 0.6818181818181818,
      "mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
      "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": 0.6666666666666666,
      "mmlu_eval_accuracy_machine_learning": 0.36363636363636365,
      "mmlu_eval_accuracy_management": 0.9090909090909091,
      "mmlu_eval_accuracy_marketing": 0.92,
      "mmlu_eval_accuracy_medical_genetics": 1.0,
      "mmlu_eval_accuracy_miscellaneous": 0.7441860465116279,
      "mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
      "mmlu_eval_accuracy_moral_scenarios": 0.38,
      "mmlu_eval_accuracy_nutrition": 0.7272727272727273,
      "mmlu_eval_accuracy_philosophy": 0.7058823529411765,
      "mmlu_eval_accuracy_prehistory": 0.4857142857142857,
      "mmlu_eval_accuracy_professional_accounting": 0.5483870967741935,
      "mmlu_eval_accuracy_professional_law": 0.40588235294117647,
      "mmlu_eval_accuracy_professional_medicine": 0.6129032258064516,
      "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": 0.4444444444444444,
      "mmlu_eval_accuracy_world_religions": 0.7894736842105263,
      "mmlu_loss": 1.8230019673977447,
      "step": 200
    }
  ],
  "logging_steps": 1,
  "max_steps": 11880,
  "num_train_epochs": 3,
  "save_steps": 200,
  "total_flos": 2.339533720800461e+16,
  "trial_name": null,
  "trial_params": null
}