{
  "best_metric": 0.7413733601570129,
  "best_model_checkpoint": "experts/mistralic-expert-21/checkpoint-600",
  "epoch": 0.11976047904191617,
  "eval_steps": 200,
  "global_step": 600,
  "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": 1.0613,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9108,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7569,
      "step": 3
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.787,
      "step": 4
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7733,
      "step": 5
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.3886,
      "step": 6
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.4824,
      "step": 7
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8347,
      "step": 8
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.82,
      "step": 9
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8169,
      "step": 10
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.5864,
      "step": 11
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8223,
      "step": 12
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7887,
      "step": 13
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7621,
      "step": 14
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7538,
      "step": 15
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6615,
      "step": 16
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9518,
      "step": 17
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.5117,
      "step": 18
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.0139,
      "step": 19
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7997,
      "step": 20
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7555,
      "step": 21
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6843,
      "step": 22
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7045,
      "step": 23
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.0227,
      "step": 24
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6964,
      "step": 25
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8441,
      "step": 26
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7352,
      "step": 27
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5966,
      "step": 28
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8846,
      "step": 29
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7978,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.1034,
      "step": 31
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8306,
      "step": 32
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0758,
      "step": 33
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7684,
      "step": 34
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8209,
      "step": 35
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.62,
      "step": 36
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.572,
      "step": 37
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6433,
      "step": 38
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7005,
      "step": 39
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.627,
      "step": 40
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0076,
      "step": 41
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6516,
      "step": 42
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7212,
      "step": 43
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7396,
      "step": 44
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.741,
      "step": 45
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9025,
      "step": 46
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8793,
      "step": 47
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6438,
      "step": 48
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7211,
      "step": 49
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7618,
      "step": 50
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8042,
      "step": 51
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7492,
      "step": 52
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.949,
      "step": 53
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8977,
      "step": 54
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8895,
      "step": 55
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8984,
      "step": 56
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7287,
      "step": 57
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5764,
      "step": 58
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.1263,
      "step": 59
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8061,
      "step": 60
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.4888,
      "step": 61
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7801,
      "step": 62
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9469,
      "step": 63
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6781,
      "step": 64
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0843,
      "step": 65
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6992,
      "step": 66
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8471,
      "step": 67
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7336,
      "step": 68
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.1684,
      "step": 69
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5612,
      "step": 70
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9468,
      "step": 71
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.895,
      "step": 72
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9233,
      "step": 73
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.073,
      "step": 74
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.972,
      "step": 75
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6457,
      "step": 76
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6522,
      "step": 77
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6546,
      "step": 78
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0024,
      "step": 79
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0696,
      "step": 80
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6181,
      "step": 81
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0234,
      "step": 82
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8514,
      "step": 83
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7032,
      "step": 84
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0316,
      "step": 85
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8664,
      "step": 86
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.469,
      "step": 87
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8426,
      "step": 88
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5955,
      "step": 89
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7823,
      "step": 90
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.778,
      "step": 91
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5699,
      "step": 92
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.949,
      "step": 93
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7923,
      "step": 94
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9486,
      "step": 95
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7278,
      "step": 96
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8223,
      "step": 97
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5869,
      "step": 98
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.466,
      "step": 99
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7447,
      "step": 100
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6255,
      "step": 101
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6209,
      "step": 102
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0633,
      "step": 103
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9947,
      "step": 104
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7474,
      "step": 105
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8626,
      "step": 106
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8522,
      "step": 107
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6075,
      "step": 108
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.912,
      "step": 109
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.012,
      "step": 110
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7406,
      "step": 111
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6897,
      "step": 112
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.543,
      "step": 113
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.4828,
      "step": 114
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9805,
      "step": 115
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7141,
      "step": 116
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9519,
      "step": 117
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8587,
      "step": 118
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7717,
      "step": 119
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6015,
      "step": 120
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7259,
      "step": 121
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0637,
      "step": 122
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7039,
      "step": 123
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.1593,
      "step": 124
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9464,
      "step": 125
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5241,
      "step": 126
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8832,
      "step": 127
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7735,
      "step": 128
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9548,
      "step": 129
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6403,
      "step": 130
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9322,
      "step": 131
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7969,
      "step": 132
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.917,
      "step": 133
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6374,
      "step": 134
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9057,
      "step": 135
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5936,
      "step": 136
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.5973,
      "step": 137
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6922,
      "step": 138
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.885,
      "step": 139
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6669,
      "step": 140
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7749,
      "step": 141
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8642,
      "step": 142
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7119,
      "step": 143
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.4539,
      "step": 144
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6143,
      "step": 145
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9801,
      "step": 146
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8143,
      "step": 147
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9165,
      "step": 148
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8678,
      "step": 149
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9648,
      "step": 150
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7353,
      "step": 151
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7707,
      "step": 152
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0535,
      "step": 153
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.74,
      "step": 154
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6083,
      "step": 155
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6862,
      "step": 156
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6716,
      "step": 157
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6058,
      "step": 158
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7035,
      "step": 159
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7014,
      "step": 160
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6037,
      "step": 161
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8274,
      "step": 162
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6332,
      "step": 163
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6243,
      "step": 164
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7386,
      "step": 165
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8272,
      "step": 166
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6769,
      "step": 167
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8564,
      "step": 168
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.794,
      "step": 169
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6748,
      "step": 170
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6602,
      "step": 171
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8841,
      "step": 172
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.748,
      "step": 173
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0154,
      "step": 174
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.721,
      "step": 175
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9784,
      "step": 176
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5253,
      "step": 177
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8122,
      "step": 178
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6862,
      "step": 179
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8924,
      "step": 180
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8975,
      "step": 181
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0512,
      "step": 182
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8278,
      "step": 183
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6857,
      "step": 184
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6204,
      "step": 185
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7821,
      "step": 186
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9541,
      "step": 187
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8025,
      "step": 188
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0036,
      "step": 189
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9583,
      "step": 190
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5459,
      "step": 191
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8891,
      "step": 192
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6371,
      "step": 193
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7171,
      "step": 194
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8905,
      "step": 195
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5407,
      "step": 196
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8385,
      "step": 197
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8422,
      "step": 198
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8084,
      "step": 199
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6382,
      "step": 200
    },
    {
      "epoch": 0.04,
      "eval_loss": 0.7527853846549988,
      "eval_runtime": 193.5321,
      "eval_samples_per_second": 5.167,
      "eval_steps_per_second": 2.584,
      "step": 200
    },
    {
      "epoch": 0.04,
      "mmlu_eval_accuracy": 0.6069425201408959,
      "mmlu_eval_accuracy_abstract_algebra": 0.45454545454545453,
      "mmlu_eval_accuracy_anatomy": 0.5,
      "mmlu_eval_accuracy_astronomy": 0.75,
      "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": 0.5454545454545454,
      "mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
      "mmlu_eval_accuracy_college_medicine": 0.5909090909090909,
      "mmlu_eval_accuracy_college_physics": 0.5454545454545454,
      "mmlu_eval_accuracy_computer_security": 0.5454545454545454,
      "mmlu_eval_accuracy_conceptual_physics": 0.5,
      "mmlu_eval_accuracy_econometrics": 0.75,
      "mmlu_eval_accuracy_electrical_engineering": 0.6875,
      "mmlu_eval_accuracy_elementary_mathematics": 0.3902439024390244,
      "mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
      "mmlu_eval_accuracy_global_facts": 0.4,
      "mmlu_eval_accuracy_high_school_biology": 0.625,
      "mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
      "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
      "mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.627906976744186,
      "mmlu_eval_accuracy_high_school_mathematics": 0.3103448275862069,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.6538461538461539,
      "mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
      "mmlu_eval_accuracy_high_school_psychology": 0.8666666666666667,
      "mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
      "mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
      "mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
      "mmlu_eval_accuracy_human_aging": 0.7391304347826086,
      "mmlu_eval_accuracy_human_sexuality": 0.5833333333333334,
      "mmlu_eval_accuracy_international_law": 0.9230769230769231,
      "mmlu_eval_accuracy_jurisprudence": 0.6363636363636364,
      "mmlu_eval_accuracy_logical_fallacies": 0.7222222222222222,
      "mmlu_eval_accuracy_machine_learning": 0.2727272727272727,
      "mmlu_eval_accuracy_management": 0.8181818181818182,
      "mmlu_eval_accuracy_marketing": 0.96,
      "mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
      "mmlu_eval_accuracy_miscellaneous": 0.7325581395348837,
      "mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
      "mmlu_eval_accuracy_moral_scenarios": 0.36,
      "mmlu_eval_accuracy_nutrition": 0.7272727272727273,
      "mmlu_eval_accuracy_philosophy": 0.7058823529411765,
      "mmlu_eval_accuracy_prehistory": 0.5714285714285714,
      "mmlu_eval_accuracy_professional_accounting": 0.6129032258064516,
      "mmlu_eval_accuracy_professional_law": 0.4117647058823529,
      "mmlu_eval_accuracy_professional_medicine": 0.6451612903225806,
      "mmlu_eval_accuracy_professional_psychology": 0.5942028985507246,
      "mmlu_eval_accuracy_public_relations": 0.5,
      "mmlu_eval_accuracy_security_studies": 0.6296296296296297,
      "mmlu_eval_accuracy_sociology": 0.9090909090909091,
      "mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
      "mmlu_eval_accuracy_virology": 0.5555555555555556,
      "mmlu_eval_accuracy_world_religions": 0.8947368421052632,
      "mmlu_loss": 1.424748288122232,
      "step": 200
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.514,
      "step": 201
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9316,
      "step": 202
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5804,
      "step": 203
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.4997,
      "step": 204
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.1044,
      "step": 205
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8998,
      "step": 206
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9126,
      "step": 207
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7343,
      "step": 208
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8555,
      "step": 209
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8367,
      "step": 210
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9339,
      "step": 211
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5325,
      "step": 212
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.97,
      "step": 213
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7861,
      "step": 214
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7172,
      "step": 215
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7829,
      "step": 216
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9416,
      "step": 217
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9772,
      "step": 218
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9562,
      "step": 219
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9923,
      "step": 220
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.773,
      "step": 221
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8065,
      "step": 222
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.4085,
      "step": 223
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7975,
      "step": 224
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7796,
      "step": 225
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9226,
      "step": 226
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9331,
      "step": 227
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8537,
      "step": 228
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5119,
      "step": 229
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9528,
      "step": 230
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8085,
      "step": 231
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6984,
      "step": 232
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7458,
      "step": 233
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5282,
      "step": 234
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7395,
      "step": 235
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7395,
      "step": 236
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7806,
      "step": 237
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0737,
      "step": 238
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8094,
      "step": 239
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8435,
      "step": 240
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6897,
      "step": 241
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9283,
      "step": 242
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.77,
      "step": 243
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7093,
      "step": 244
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6659,
      "step": 245
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6623,
      "step": 246
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.2126,
      "step": 247
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8251,
      "step": 248
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6103,
      "step": 249
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.734,
      "step": 250
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5445,
      "step": 251
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6698,
      "step": 252
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6505,
      "step": 253
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6185,
      "step": 254
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8441,
      "step": 255
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7706,
      "step": 256
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7232,
      "step": 257
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.566,
      "step": 258
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7725,
      "step": 259
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.772,
      "step": 260
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8432,
      "step": 261
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7987,
      "step": 262
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5326,
      "step": 263
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6139,
      "step": 264
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8207,
      "step": 265
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5063,
      "step": 266
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5334,
      "step": 267
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9008,
      "step": 268
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6599,
      "step": 269
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8471,
      "step": 270
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6595,
      "step": 271
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5456,
      "step": 272
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0143,
      "step": 273
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.67,
      "step": 274
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6222,
      "step": 275
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7805,
      "step": 276
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6144,
      "step": 277
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7708,
      "step": 278
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7156,
      "step": 279
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7694,
      "step": 280
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6985,
      "step": 281
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.874,
      "step": 282
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.5317,
      "step": 283
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6243,
      "step": 284
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6899,
      "step": 285
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6082,
      "step": 286
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8266,
      "step": 287
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7347,
      "step": 288
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.751,
      "step": 289
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0287,
      "step": 290
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8968,
      "step": 291
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6152,
      "step": 292
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6274,
      "step": 293
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8135,
      "step": 294
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9844,
      "step": 295
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.524,
      "step": 296
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9002,
      "step": 297
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8671,
      "step": 298
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.571,
      "step": 299
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6869,
      "step": 300
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7714,
      "step": 301
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8017,
      "step": 302
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7137,
      "step": 303
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0807,
      "step": 304
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0078,
      "step": 305
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.1694,
      "step": 306
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7987,
      "step": 307
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6085,
      "step": 308
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.5241,
      "step": 309
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.5681,
      "step": 310
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.5722,
      "step": 311
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6979,
      "step": 312
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6865,
      "step": 313
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9059,
      "step": 314
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8185,
      "step": 315
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6021,
      "step": 316
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7516,
      "step": 317
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.911,
      "step": 318
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7015,
      "step": 319
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.597,
      "step": 320
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0015,
      "step": 321
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7167,
      "step": 322
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9415,
      "step": 323
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8917,
      "step": 324
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7481,
      "step": 325
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5737,
      "step": 326
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9759,
      "step": 327
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9918,
      "step": 328
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9649,
      "step": 329
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7749,
      "step": 330
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9503,
      "step": 331
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7141,
      "step": 332
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6378,
      "step": 333
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7751,
      "step": 334
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.645,
      "step": 335
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9118,
      "step": 336
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7924,
      "step": 337
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6403,
      "step": 338
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8784,
      "step": 339
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8649,
      "step": 340
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8398,
      "step": 341
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5552,
      "step": 342
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5564,
      "step": 343
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9732,
      "step": 344
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.1613,
      "step": 345
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7281,
      "step": 346
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6125,
      "step": 347
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8193,
      "step": 348
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8056,
      "step": 349
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7315,
      "step": 350
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8247,
      "step": 351
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8393,
      "step": 352
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7128,
      "step": 353
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9818,
      "step": 354
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8143,
      "step": 355
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6425,
      "step": 356
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7803,
      "step": 357
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.742,
      "step": 358
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.4946,
      "step": 359
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8821,
      "step": 360
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8141,
      "step": 361
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9041,
      "step": 362
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.1578,
      "step": 363
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5657,
      "step": 364
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7776,
      "step": 365
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7091,
      "step": 366
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6348,
      "step": 367
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6316,
      "step": 368
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6251,
      "step": 369
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.0801,
      "step": 370
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7038,
      "step": 371
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5184,
      "step": 372
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7923,
      "step": 373
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8136,
      "step": 374
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9586,
      "step": 375
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9144,
      "step": 376
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.599,
      "step": 377
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7147,
      "step": 378
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7103,
      "step": 379
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6476,
      "step": 380
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8461,
      "step": 381
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.0367,
      "step": 382
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7077,
      "step": 383
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8001,
      "step": 384
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.2422,
      "step": 385
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8415,
      "step": 386
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.736,
      "step": 387
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8057,
      "step": 388
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6561,
      "step": 389
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8917,
      "step": 390
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6604,
      "step": 391
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6119,
      "step": 392
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7716,
      "step": 393
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7671,
      "step": 394
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6078,
      "step": 395
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9074,
      "step": 396
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7811,
      "step": 397
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9108,
      "step": 398
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7685,
      "step": 399
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.0709,
      "step": 400
    },
    {
      "epoch": 0.08,
      "eval_loss": 0.744175374507904,
      "eval_runtime": 193.6137,
      "eval_samples_per_second": 5.165,
      "eval_steps_per_second": 2.582,
      "step": 400
    },
    {
      "epoch": 0.08,
      "mmlu_eval_accuracy": 0.6043681736715669,
      "mmlu_eval_accuracy_abstract_algebra": 0.45454545454545453,
      "mmlu_eval_accuracy_anatomy": 0.5,
      "mmlu_eval_accuracy_astronomy": 0.75,
      "mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
      "mmlu_eval_accuracy_clinical_knowledge": 0.5862068965517241,
      "mmlu_eval_accuracy_college_biology": 0.5,
      "mmlu_eval_accuracy_college_chemistry": 0.375,
      "mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
      "mmlu_eval_accuracy_college_mathematics": 0.5454545454545454,
      "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": 0.5384615384615384,
      "mmlu_eval_accuracy_econometrics": 0.5833333333333334,
      "mmlu_eval_accuracy_electrical_engineering": 0.5625,
      "mmlu_eval_accuracy_elementary_mathematics": 0.43902439024390244,
      "mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
      "mmlu_eval_accuracy_global_facts": 0.3,
      "mmlu_eval_accuracy_high_school_biology": 0.46875,
      "mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
      "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_european_history": 0.8333333333333334,
      "mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.5813953488372093,
      "mmlu_eval_accuracy_high_school_mathematics": 0.3793103448275862,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.6538461538461539,
      "mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
      "mmlu_eval_accuracy_high_school_psychology": 0.8666666666666667,
      "mmlu_eval_accuracy_high_school_statistics": 0.2608695652173913,
      "mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
      "mmlu_eval_accuracy_high_school_world_history": 0.7307692307692307,
      "mmlu_eval_accuracy_human_aging": 0.7391304347826086,
      "mmlu_eval_accuracy_human_sexuality": 0.5833333333333334,
      "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": 0.9090909090909091,
      "mmlu_eval_accuracy_miscellaneous": 0.7325581395348837,
      "mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
      "mmlu_eval_accuracy_moral_scenarios": 0.32,
      "mmlu_eval_accuracy_nutrition": 0.7575757575757576,
      "mmlu_eval_accuracy_philosophy": 0.7058823529411765,
      "mmlu_eval_accuracy_prehistory": 0.5714285714285714,
      "mmlu_eval_accuracy_professional_accounting": 0.6129032258064516,
      "mmlu_eval_accuracy_professional_law": 0.38823529411764707,
      "mmlu_eval_accuracy_professional_medicine": 0.6129032258064516,
      "mmlu_eval_accuracy_professional_psychology": 0.6086956521739131,
      "mmlu_eval_accuracy_public_relations": 0.5833333333333334,
      "mmlu_eval_accuracy_security_studies": 0.6666666666666666,
      "mmlu_eval_accuracy_sociology": 0.8636363636363636,
      "mmlu_eval_accuracy_us_foreign_policy": 0.9090909090909091,
      "mmlu_eval_accuracy_virology": 0.5555555555555556,
      "mmlu_eval_accuracy_world_religions": 0.7894736842105263,
      "mmlu_loss": 1.4941282213979228,
      "step": 400
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8267,
      "step": 401
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7679,
      "step": 402
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8803,
      "step": 403
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6909,
      "step": 404
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7483,
      "step": 405
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.0523,
      "step": 406
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9839,
      "step": 407
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.5975,
      "step": 408
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 1.0406,
      "step": 409
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9047,
      "step": 410
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7649,
      "step": 411
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7999,
      "step": 412
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7923,
      "step": 413
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7002,
      "step": 414
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8975,
      "step": 415
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7059,
      "step": 416
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6865,
      "step": 417
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7177,
      "step": 418
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8824,
      "step": 419
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.787,
      "step": 420
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8307,
      "step": 421
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7682,
      "step": 422
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8036,
      "step": 423
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9209,
      "step": 424
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9188,
      "step": 425
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7904,
      "step": 426
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.9804,
      "step": 427
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6341,
      "step": 428
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.8869,
      "step": 429
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.8143,
      "step": 430
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.8974,
      "step": 431
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6402,
      "step": 432
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5342,
      "step": 433
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7838,
      "step": 434
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.8436,
      "step": 435
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 1.024,
      "step": 436
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.8074,
      "step": 437
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6338,
      "step": 438
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.9353,
      "step": 439
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6655,
      "step": 440
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.661,
      "step": 441
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7146,
      "step": 442
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.8318,
      "step": 443
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5375,
      "step": 444
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.8082,
      "step": 445
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6488,
      "step": 446
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.8286,
      "step": 447
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5802,
      "step": 448
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7761,
      "step": 449
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7238,
      "step": 450
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.9706,
      "step": 451
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5635,
      "step": 452
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7103,
      "step": 453
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.4757,
      "step": 454
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5925,
      "step": 455
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6109,
      "step": 456
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.622,
      "step": 457
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7854,
      "step": 458
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5802,
      "step": 459
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7563,
      "step": 460
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7902,
      "step": 461
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.8256,
      "step": 462
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.948,
      "step": 463
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.745,
      "step": 464
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6632,
      "step": 465
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6803,
      "step": 466
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.8442,
      "step": 467
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6172,
      "step": 468
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5671,
      "step": 469
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7855,
      "step": 470
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6973,
      "step": 471
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5741,
      "step": 472
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7252,
      "step": 473
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.7218,
      "step": 474
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.5713,
      "step": 475
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.8696,
      "step": 476
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.6925,
      "step": 477
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 1.0487,
      "step": 478
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7919,
      "step": 479
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.9096,
      "step": 480
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.8272,
      "step": 481
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.8518,
      "step": 482
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.6736,
      "step": 483
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.802,
      "step": 484
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.8978,
      "step": 485
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.657,
      "step": 486
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.5339,
      "step": 487
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7309,
      "step": 488
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.6462,
      "step": 489
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.5552,
      "step": 490
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 1.1669,
      "step": 491
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7967,
      "step": 492
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7578,
      "step": 493
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.8775,
      "step": 494
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7491,
      "step": 495
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.9756,
      "step": 496
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.8121,
      "step": 497
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7392,
      "step": 498
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.9004,
      "step": 499
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7345,
      "step": 500
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7513,
      "step": 501
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7502,
      "step": 502
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.8346,
      "step": 503
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.9917,
      "step": 504
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.8684,
      "step": 505
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.9052,
      "step": 506
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.6764,
      "step": 507
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7646,
      "step": 508
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.4939,
      "step": 509
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 1.0915,
      "step": 510
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.4308,
      "step": 511
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7663,
      "step": 512
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.6548,
      "step": 513
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.6723,
      "step": 514
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 1.0556,
      "step": 515
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7296,
      "step": 516
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.9113,
      "step": 517
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.9695,
      "step": 518
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7099,
      "step": 519
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.5564,
      "step": 520
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.8164,
      "step": 521
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.5783,
      "step": 522
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.7821,
      "step": 523
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.6573,
      "step": 524
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.8268,
      "step": 525
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.6128,
      "step": 526
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.6761,
      "step": 527
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.7277,
      "step": 528
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.5248,
      "step": 529
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.9179,
      "step": 530
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.6577,
      "step": 531
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.6512,
      "step": 532
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.7375,
      "step": 533
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.9032,
      "step": 534
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.7618,
      "step": 535
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.7835,
      "step": 536
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.6359,
      "step": 537
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 1.052,
      "step": 538
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.5695,
      "step": 539
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.8816,
      "step": 540
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.6503,
      "step": 541
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.6225,
      "step": 542
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.5026,
      "step": 543
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.8693,
      "step": 544
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.7503,
      "step": 545
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.7084,
      "step": 546
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.9936,
      "step": 547
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.8152,
      "step": 548
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.9392,
      "step": 549
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 1.0249,
      "step": 550
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.9395,
      "step": 551
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.4423,
      "step": 552
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.9268,
      "step": 553
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.9497,
      "step": 554
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.468,
      "step": 555
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.8133,
      "step": 556
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.6607,
      "step": 557
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.8391,
      "step": 558
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.8714,
      "step": 559
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.6308,
      "step": 560
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.6394,
      "step": 561
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.913,
      "step": 562
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.688,
      "step": 563
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.8688,
      "step": 564
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.7826,
      "step": 565
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.5856,
      "step": 566
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 1.0415,
      "step": 567
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.9789,
      "step": 568
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.9218,
      "step": 569
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.7543,
      "step": 570
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.53,
      "step": 571
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 1.1324,
      "step": 572
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.7376,
      "step": 573
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.9111,
      "step": 574
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.7983,
      "step": 575
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.7959,
      "step": 576
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.8756,
      "step": 577
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 1.0287,
      "step": 578
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.4868,
      "step": 579
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.6186,
      "step": 580
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.6415,
      "step": 581
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.8003,
      "step": 582
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.8524,
      "step": 583
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.8741,
      "step": 584
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.6316,
      "step": 585
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.6371,
      "step": 586
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.7191,
      "step": 587
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.7531,
      "step": 588
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.6027,
      "step": 589
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.7606,
      "step": 590
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.938,
      "step": 591
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.7439,
      "step": 592
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.9444,
      "step": 593
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.6378,
      "step": 594
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.9355,
      "step": 595
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.7833,
      "step": 596
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.7256,
      "step": 597
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.6942,
      "step": 598
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.7742,
      "step": 599
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.684,
      "step": 600
    },
    {
      "epoch": 0.12,
      "eval_loss": 0.7413733601570129,
      "eval_runtime": 193.4196,
      "eval_samples_per_second": 5.17,
      "eval_steps_per_second": 2.585,
      "step": 600
    },
    {
      "epoch": 0.12,
      "mmlu_eval_accuracy": 0.6066353154142871,
      "mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
      "mmlu_eval_accuracy_anatomy": 0.5714285714285714,
      "mmlu_eval_accuracy_astronomy": 0.8125,
      "mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
      "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": 0.36363636363636365,
      "mmlu_eval_accuracy_college_mathematics": 0.5454545454545454,
      "mmlu_eval_accuracy_college_medicine": 0.5454545454545454,
      "mmlu_eval_accuracy_college_physics": 0.5454545454545454,
      "mmlu_eval_accuracy_computer_security": 0.5454545454545454,
      "mmlu_eval_accuracy_conceptual_physics": 0.5769230769230769,
      "mmlu_eval_accuracy_econometrics": 0.5833333333333334,
      "mmlu_eval_accuracy_electrical_engineering": 0.625,
      "mmlu_eval_accuracy_elementary_mathematics": 0.4634146341463415,
      "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": 0.4090909090909091,
      "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
      "mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.627906976744186,
      "mmlu_eval_accuracy_high_school_mathematics": 0.27586206896551724,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.5769230769230769,
      "mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
      "mmlu_eval_accuracy_high_school_psychology": 0.8666666666666667,
      "mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
      "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": 0.9230769230769231,
      "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": 0.92,
      "mmlu_eval_accuracy_medical_genetics": 1.0,
      "mmlu_eval_accuracy_miscellaneous": 0.7441860465116279,
      "mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
      "mmlu_eval_accuracy_moral_scenarios": 0.33,
      "mmlu_eval_accuracy_nutrition": 0.7272727272727273,
      "mmlu_eval_accuracy_philosophy": 0.7352941176470589,
      "mmlu_eval_accuracy_prehistory": 0.6,
      "mmlu_eval_accuracy_professional_accounting": 0.5483870967741935,
      "mmlu_eval_accuracy_professional_law": 0.38823529411764707,
      "mmlu_eval_accuracy_professional_medicine": 0.6451612903225806,
      "mmlu_eval_accuracy_professional_psychology": 0.5797101449275363,
      "mmlu_eval_accuracy_public_relations": 0.5833333333333334,
      "mmlu_eval_accuracy_security_studies": 0.6666666666666666,
      "mmlu_eval_accuracy_sociology": 0.8636363636363636,
      "mmlu_eval_accuracy_us_foreign_policy": 0.9090909090909091,
      "mmlu_eval_accuracy_virology": 0.5555555555555556,
      "mmlu_eval_accuracy_world_religions": 0.8421052631578947,
      "mmlu_loss": 1.546568229210906,
      "step": 600
    }
  ],
  "logging_steps": 1,
  "max_steps": 15030,
  "num_train_epochs": 3,
  "save_steps": 200,
  "total_flos": 1.737893418864476e+17,
  "trial_name": null,
  "trial_params": null
}