{
  "best_metric": 0.7461190223693848,
  "best_model_checkpoint": "experts/mistralic-expert-26/checkpoint-400",
  "epoch": 0.05355648535564853,
  "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": 0.8469,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8006,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.793,
      "step": 3
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8309,
      "step": 4
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8613,
      "step": 5
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7609,
      "step": 6
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7405,
      "step": 7
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6562,
      "step": 8
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8598,
      "step": 9
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.8061,
      "step": 10
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6911,
      "step": 11
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7309,
      "step": 12
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6352,
      "step": 13
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7019,
      "step": 14
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.601,
      "step": 15
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7527,
      "step": 16
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9178,
      "step": 17
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8868,
      "step": 18
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9064,
      "step": 19
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.646,
      "step": 20
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8691,
      "step": 21
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9415,
      "step": 22
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7216,
      "step": 23
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.765,
      "step": 24
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7059,
      "step": 25
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7747,
      "step": 26
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6623,
      "step": 27
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7587,
      "step": 28
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9956,
      "step": 29
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7459,
      "step": 30
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9004,
      "step": 31
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6704,
      "step": 32
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6312,
      "step": 33
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6526,
      "step": 34
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.632,
      "step": 35
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7005,
      "step": 36
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.847,
      "step": 37
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7952,
      "step": 38
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7577,
      "step": 39
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7908,
      "step": 40
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9948,
      "step": 41
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5511,
      "step": 42
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6293,
      "step": 43
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8038,
      "step": 44
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6822,
      "step": 45
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8786,
      "step": 46
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7836,
      "step": 47
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7824,
      "step": 48
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7181,
      "step": 49
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8472,
      "step": 50
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7194,
      "step": 51
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6755,
      "step": 52
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9392,
      "step": 53
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7776,
      "step": 54
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7357,
      "step": 55
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7,
      "step": 56
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.801,
      "step": 57
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8782,
      "step": 58
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8814,
      "step": 59
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7144,
      "step": 60
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5467,
      "step": 61
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.879,
      "step": 62
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7626,
      "step": 63
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6348,
      "step": 64
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7493,
      "step": 65
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.703,
      "step": 66
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6604,
      "step": 67
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7255,
      "step": 68
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9477,
      "step": 69
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7842,
      "step": 70
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7174,
      "step": 71
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6468,
      "step": 72
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9514,
      "step": 73
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6043,
      "step": 74
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7093,
      "step": 75
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6341,
      "step": 76
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6099,
      "step": 77
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6215,
      "step": 78
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8154,
      "step": 79
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6901,
      "step": 80
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7121,
      "step": 81
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7695,
      "step": 82
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7517,
      "step": 83
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8143,
      "step": 84
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5674,
      "step": 85
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6716,
      "step": 86
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7034,
      "step": 87
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.665,
      "step": 88
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7358,
      "step": 89
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.65,
      "step": 90
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5839,
      "step": 91
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7898,
      "step": 92
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8046,
      "step": 93
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9556,
      "step": 94
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6294,
      "step": 95
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7129,
      "step": 96
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9083,
      "step": 97
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6225,
      "step": 98
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9371,
      "step": 99
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9608,
      "step": 100
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7615,
      "step": 101
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.85,
      "step": 102
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8309,
      "step": 103
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6994,
      "step": 104
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7748,
      "step": 105
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6932,
      "step": 106
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7905,
      "step": 107
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7735,
      "step": 108
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6634,
      "step": 109
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.844,
      "step": 110
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9069,
      "step": 111
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.894,
      "step": 112
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7108,
      "step": 113
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.797,
      "step": 114
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.712,
      "step": 115
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7004,
      "step": 116
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8133,
      "step": 117
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7058,
      "step": 118
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5617,
      "step": 119
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6603,
      "step": 120
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7919,
      "step": 121
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8013,
      "step": 122
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.543,
      "step": 123
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7799,
      "step": 124
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7825,
      "step": 125
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7331,
      "step": 126
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5461,
      "step": 127
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.2894,
      "step": 128
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5776,
      "step": 129
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.4728,
      "step": 130
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8037,
      "step": 131
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8163,
      "step": 132
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9718,
      "step": 133
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7634,
      "step": 134
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7856,
      "step": 135
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6623,
      "step": 136
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6417,
      "step": 137
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8078,
      "step": 138
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6473,
      "step": 139
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7786,
      "step": 140
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.822,
      "step": 141
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8149,
      "step": 142
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6739,
      "step": 143
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.681,
      "step": 144
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8158,
      "step": 145
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6431,
      "step": 146
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8103,
      "step": 147
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5391,
      "step": 148
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5818,
      "step": 149
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9752,
      "step": 150
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7014,
      "step": 151
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8224,
      "step": 152
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7193,
      "step": 153
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5579,
      "step": 154
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5877,
      "step": 155
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8186,
      "step": 156
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5337,
      "step": 157
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7355,
      "step": 158
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8316,
      "step": 159
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.795,
      "step": 160
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8623,
      "step": 161
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7517,
      "step": 162
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7901,
      "step": 163
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6109,
      "step": 164
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8499,
      "step": 165
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6832,
      "step": 166
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.563,
      "step": 167
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7447,
      "step": 168
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5976,
      "step": 169
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7516,
      "step": 170
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5842,
      "step": 171
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5788,
      "step": 172
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0914,
      "step": 173
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7697,
      "step": 174
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7694,
      "step": 175
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8007,
      "step": 176
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8465,
      "step": 177
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8914,
      "step": 178
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7133,
      "step": 179
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8848,
      "step": 180
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7373,
      "step": 181
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6826,
      "step": 182
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7722,
      "step": 183
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5657,
      "step": 184
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7213,
      "step": 185
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6559,
      "step": 186
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6655,
      "step": 187
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7012,
      "step": 188
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8487,
      "step": 189
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7951,
      "step": 190
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6477,
      "step": 191
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7438,
      "step": 192
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8507,
      "step": 193
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6214,
      "step": 194
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6879,
      "step": 195
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7684,
      "step": 196
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7626,
      "step": 197
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6479,
      "step": 198
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7303,
      "step": 199
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7522,
      "step": 200
    },
    {
      "epoch": 0.03,
      "eval_loss": 0.7530023455619812,
      "eval_runtime": 89.5744,
      "eval_samples_per_second": 11.164,
      "eval_steps_per_second": 5.582,
      "step": 200
    },
    {
      "epoch": 0.03,
      "mmlu_eval_accuracy": 0.5961120128104086,
      "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
      "mmlu_eval_accuracy_anatomy": 0.42857142857142855,
      "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": 0.5625,
      "mmlu_eval_accuracy_college_chemistry": 0.25,
      "mmlu_eval_accuracy_college_computer_science": 0.5454545454545454,
      "mmlu_eval_accuracy_college_mathematics": 0.6363636363636364,
      "mmlu_eval_accuracy_college_medicine": 0.5,
      "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
      "mmlu_eval_accuracy_computer_security": 0.7272727272727273,
      "mmlu_eval_accuracy_conceptual_physics": 0.5769230769230769,
      "mmlu_eval_accuracy_econometrics": 0.5833333333333334,
      "mmlu_eval_accuracy_electrical_engineering": 0.75,
      "mmlu_eval_accuracy_elementary_mathematics": 0.34146341463414637,
      "mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
      "mmlu_eval_accuracy_global_facts": 0.4,
      "mmlu_eval_accuracy_high_school_biology": 0.65625,
      "mmlu_eval_accuracy_high_school_chemistry": 0.36363636363636365,
      "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
      "mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.6046511627906976,
      "mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.7307692307692307,
      "mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
      "mmlu_eval_accuracy_high_school_psychology": 0.8333333333333334,
      "mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
      "mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
      "mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
      "mmlu_eval_accuracy_human_aging": 0.7391304347826086,
      "mmlu_eval_accuracy_human_sexuality": 0.5833333333333334,
      "mmlu_eval_accuracy_international_law": 0.8461538461538461,
      "mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
      "mmlu_eval_accuracy_logical_fallacies": 0.7222222222222222,
      "mmlu_eval_accuracy_machine_learning": 0.36363636363636365,
      "mmlu_eval_accuracy_management": 0.9090909090909091,
      "mmlu_eval_accuracy_marketing": 0.84,
      "mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
      "mmlu_eval_accuracy_miscellaneous": 0.7325581395348837,
      "mmlu_eval_accuracy_moral_disputes": 0.5789473684210527,
      "mmlu_eval_accuracy_moral_scenarios": 0.32,
      "mmlu_eval_accuracy_nutrition": 0.7575757575757576,
      "mmlu_eval_accuracy_philosophy": 0.7058823529411765,
      "mmlu_eval_accuracy_prehistory": 0.6285714285714286,
      "mmlu_eval_accuracy_professional_accounting": 0.5483870967741935,
      "mmlu_eval_accuracy_professional_law": 0.3941176470588235,
      "mmlu_eval_accuracy_professional_medicine": 0.5806451612903226,
      "mmlu_eval_accuracy_professional_psychology": 0.5362318840579711,
      "mmlu_eval_accuracy_public_relations": 0.4166666666666667,
      "mmlu_eval_accuracy_security_studies": 0.5925925925925926,
      "mmlu_eval_accuracy_sociology": 0.9090909090909091,
      "mmlu_eval_accuracy_us_foreign_policy": 0.9090909090909091,
      "mmlu_eval_accuracy_virology": 0.4444444444444444,
      "mmlu_eval_accuracy_world_religions": 0.8421052631578947,
      "mmlu_loss": 1.0777005341205521,
      "step": 200
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6374,
      "step": 201
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7244,
      "step": 202
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6978,
      "step": 203
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7097,
      "step": 204
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5994,
      "step": 205
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7382,
      "step": 206
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6402,
      "step": 207
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6145,
      "step": 208
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7074,
      "step": 209
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6967,
      "step": 210
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8425,
      "step": 211
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7417,
      "step": 212
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.2699,
      "step": 213
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7487,
      "step": 214
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6402,
      "step": 215
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7093,
      "step": 216
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6423,
      "step": 217
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0206,
      "step": 218
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7523,
      "step": 219
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6754,
      "step": 220
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.807,
      "step": 221
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9181,
      "step": 222
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7908,
      "step": 223
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6954,
      "step": 224
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7227,
      "step": 225
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7656,
      "step": 226
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6452,
      "step": 227
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7213,
      "step": 228
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6493,
      "step": 229
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7346,
      "step": 230
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7853,
      "step": 231
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6369,
      "step": 232
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7253,
      "step": 233
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7517,
      "step": 234
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7005,
      "step": 235
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7019,
      "step": 236
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6547,
      "step": 237
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7818,
      "step": 238
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8312,
      "step": 239
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5597,
      "step": 240
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6868,
      "step": 241
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7939,
      "step": 242
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5629,
      "step": 243
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8805,
      "step": 244
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7323,
      "step": 245
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6907,
      "step": 246
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8249,
      "step": 247
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6834,
      "step": 248
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6812,
      "step": 249
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8043,
      "step": 250
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8576,
      "step": 251
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7742,
      "step": 252
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6906,
      "step": 253
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6783,
      "step": 254
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6227,
      "step": 255
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8105,
      "step": 256
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6703,
      "step": 257
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.1037,
      "step": 258
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6402,
      "step": 259
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9586,
      "step": 260
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7577,
      "step": 261
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8077,
      "step": 262
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8953,
      "step": 263
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7339,
      "step": 264
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6343,
      "step": 265
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9431,
      "step": 266
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8053,
      "step": 267
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7228,
      "step": 268
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7385,
      "step": 269
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5731,
      "step": 270
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8134,
      "step": 271
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7456,
      "step": 272
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8167,
      "step": 273
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6061,
      "step": 274
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6861,
      "step": 275
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5313,
      "step": 276
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6491,
      "step": 277
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7365,
      "step": 278
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7581,
      "step": 279
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7466,
      "step": 280
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.116,
      "step": 281
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7806,
      "step": 282
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8139,
      "step": 283
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7923,
      "step": 284
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7013,
      "step": 285
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6228,
      "step": 286
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6846,
      "step": 287
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7022,
      "step": 288
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6284,
      "step": 289
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5942,
      "step": 290
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.712,
      "step": 291
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6987,
      "step": 292
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6721,
      "step": 293
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6781,
      "step": 294
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6876,
      "step": 295
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6842,
      "step": 296
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7728,
      "step": 297
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7394,
      "step": 298
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6089,
      "step": 299
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7674,
      "step": 300
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6131,
      "step": 301
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7942,
      "step": 302
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6117,
      "step": 303
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5835,
      "step": 304
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7201,
      "step": 305
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0504,
      "step": 306
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6432,
      "step": 307
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7559,
      "step": 308
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8892,
      "step": 309
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0394,
      "step": 310
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7912,
      "step": 311
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7815,
      "step": 312
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7433,
      "step": 313
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6478,
      "step": 314
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8275,
      "step": 315
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8695,
      "step": 316
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8166,
      "step": 317
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6874,
      "step": 318
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7705,
      "step": 319
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.603,
      "step": 320
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8942,
      "step": 321
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8217,
      "step": 322
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6167,
      "step": 323
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8269,
      "step": 324
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7044,
      "step": 325
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7432,
      "step": 326
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6309,
      "step": 327
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7465,
      "step": 328
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.793,
      "step": 329
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7845,
      "step": 330
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7565,
      "step": 331
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6906,
      "step": 332
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6994,
      "step": 333
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7468,
      "step": 334
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8354,
      "step": 335
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8465,
      "step": 336
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7144,
      "step": 337
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8756,
      "step": 338
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7708,
      "step": 339
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7718,
      "step": 340
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9794,
      "step": 341
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7535,
      "step": 342
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8074,
      "step": 343
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7776,
      "step": 344
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7466,
      "step": 345
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8341,
      "step": 346
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.688,
      "step": 347
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.819,
      "step": 348
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7021,
      "step": 349
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7799,
      "step": 350
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8469,
      "step": 351
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7183,
      "step": 352
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7751,
      "step": 353
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5917,
      "step": 354
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9829,
      "step": 355
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6522,
      "step": 356
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9012,
      "step": 357
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0506,
      "step": 358
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7775,
      "step": 359
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8593,
      "step": 360
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7605,
      "step": 361
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6358,
      "step": 362
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7548,
      "step": 363
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7528,
      "step": 364
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0458,
      "step": 365
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9207,
      "step": 366
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6877,
      "step": 367
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7077,
      "step": 368
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7992,
      "step": 369
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.954,
      "step": 370
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8487,
      "step": 371
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6797,
      "step": 372
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7036,
      "step": 373
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8241,
      "step": 374
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6894,
      "step": 375
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8354,
      "step": 376
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7885,
      "step": 377
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7283,
      "step": 378
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7293,
      "step": 379
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8178,
      "step": 380
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6726,
      "step": 381
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7794,
      "step": 382
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8613,
      "step": 383
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7072,
      "step": 384
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7113,
      "step": 385
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6039,
      "step": 386
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9879,
      "step": 387
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6862,
      "step": 388
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7019,
      "step": 389
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7679,
      "step": 390
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8131,
      "step": 391
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6174,
      "step": 392
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6729,
      "step": 393
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6222,
      "step": 394
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6997,
      "step": 395
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7967,
      "step": 396
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6941,
      "step": 397
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8199,
      "step": 398
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6833,
      "step": 399
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7067,
      "step": 400
    },
    {
      "epoch": 0.05,
      "eval_loss": 0.7461190223693848,
      "eval_runtime": 89.365,
      "eval_samples_per_second": 11.19,
      "eval_steps_per_second": 5.595,
      "step": 400
    },
    {
      "epoch": 0.05,
      "mmlu_eval_accuracy": 0.5797172913046011,
      "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
      "mmlu_eval_accuracy_anatomy": 0.42857142857142855,
      "mmlu_eval_accuracy_astronomy": 0.6875,
      "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
      "mmlu_eval_accuracy_clinical_knowledge": 0.6206896551724138,
      "mmlu_eval_accuracy_college_biology": 0.4375,
      "mmlu_eval_accuracy_college_chemistry": 0.375,
      "mmlu_eval_accuracy_college_computer_science": 0.6363636363636364,
      "mmlu_eval_accuracy_college_mathematics": 0.36363636363636365,
      "mmlu_eval_accuracy_college_medicine": 0.5909090909090909,
      "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
      "mmlu_eval_accuracy_computer_security": 0.5454545454545454,
      "mmlu_eval_accuracy_conceptual_physics": 0.46153846153846156,
      "mmlu_eval_accuracy_econometrics": 0.5,
      "mmlu_eval_accuracy_electrical_engineering": 0.6875,
      "mmlu_eval_accuracy_elementary_mathematics": 0.36585365853658536,
      "mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
      "mmlu_eval_accuracy_global_facts": 0.5,
      "mmlu_eval_accuracy_high_school_biology": 0.5,
      "mmlu_eval_accuracy_high_school_chemistry": 0.2727272727272727,
      "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
      "mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.5581395348837209,
      "mmlu_eval_accuracy_high_school_mathematics": 0.4827586206896552,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.5769230769230769,
      "mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
      "mmlu_eval_accuracy_high_school_psychology": 0.7833333333333333,
      "mmlu_eval_accuracy_high_school_statistics": 0.43478260869565216,
      "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": 0.8461538461538461,
      "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": 0.88,
      "mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
      "mmlu_eval_accuracy_miscellaneous": 0.7441860465116279,
      "mmlu_eval_accuracy_moral_disputes": 0.5789473684210527,
      "mmlu_eval_accuracy_moral_scenarios": 0.3,
      "mmlu_eval_accuracy_nutrition": 0.5757575757575758,
      "mmlu_eval_accuracy_philosophy": 0.6764705882352942,
      "mmlu_eval_accuracy_prehistory": 0.45714285714285713,
      "mmlu_eval_accuracy_professional_accounting": 0.6129032258064516,
      "mmlu_eval_accuracy_professional_law": 0.38235294117647056,
      "mmlu_eval_accuracy_professional_medicine": 0.5483870967741935,
      "mmlu_eval_accuracy_professional_psychology": 0.5942028985507246,
      "mmlu_eval_accuracy_public_relations": 0.4166666666666667,
      "mmlu_eval_accuracy_security_studies": 0.6666666666666666,
      "mmlu_eval_accuracy_sociology": 0.8636363636363636,
      "mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
      "mmlu_eval_accuracy_virology": 0.5,
      "mmlu_eval_accuracy_world_religions": 0.8947368421052632,
      "mmlu_loss": 1.0156579838175686,
      "step": 400
    }
  ],
  "logging_steps": 1,
  "max_steps": 22404,
  "num_train_epochs": 3,
  "save_steps": 200,
  "total_flos": 4.447988472840192e+16,
  "trial_name": null,
  "trial_params": null
}