{
  "best_metric": 0.6062986850738525,
  "best_model_checkpoint": "./output_v2/34bCodellama_CodeLlama-34b-Python-hf_codellama_blob_1/checkpoint-1000",
  "epoch": 0.8424008424008425,
  "global_step": 1200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0.0004,
      "loss": 1.5667,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0004,
      "loss": 0.9201,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0004,
      "loss": 0.6019,
      "step": 3
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0004,
      "loss": 0.861,
      "step": 4
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0004,
      "loss": 1.064,
      "step": 5
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0004,
      "loss": 1.0007,
      "step": 6
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0004,
      "loss": 1.0533,
      "step": 7
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0004,
      "loss": 0.8498,
      "step": 8
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0004,
      "loss": 0.9189,
      "step": 9
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0004,
      "loss": 0.9369,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0004,
      "loss": 1.1375,
      "step": 11
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0004,
      "loss": 0.8369,
      "step": 12
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0004,
      "loss": 0.689,
      "step": 13
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0004,
      "loss": 0.8885,
      "step": 14
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0004,
      "loss": 0.9666,
      "step": 15
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0004,
      "loss": 0.744,
      "step": 16
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0004,
      "loss": 0.8998,
      "step": 17
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0004,
      "loss": 0.8371,
      "step": 18
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0004,
      "loss": 0.7615,
      "step": 19
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0004,
      "loss": 0.6963,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0004,
      "loss": 0.5974,
      "step": 21
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0004,
      "loss": 1.4922,
      "step": 22
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0004,
      "loss": 1.1272,
      "step": 23
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0004,
      "loss": 0.4373,
      "step": 24
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0004,
      "loss": 1.0598,
      "step": 25
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0004,
      "loss": 1.191,
      "step": 26
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0004,
      "loss": 0.6499,
      "step": 27
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0004,
      "loss": 0.7526,
      "step": 28
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0004,
      "loss": 0.7252,
      "step": 29
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0004,
      "loss": 0.8555,
      "step": 30
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0004,
      "loss": 0.479,
      "step": 31
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0004,
      "loss": 0.5273,
      "step": 32
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0004,
      "loss": 0.3907,
      "step": 33
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0004,
      "loss": 0.5284,
      "step": 34
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0004,
      "loss": 0.8696,
      "step": 35
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0004,
      "loss": 0.9264,
      "step": 36
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0004,
      "loss": 0.6812,
      "step": 37
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0004,
      "loss": 0.4482,
      "step": 38
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0004,
      "loss": 1.9031,
      "step": 39
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0004,
      "loss": 1.3295,
      "step": 40
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0004,
      "loss": 0.7498,
      "step": 41
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0004,
      "loss": 1.4452,
      "step": 42
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0004,
      "loss": 1.224,
      "step": 43
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0004,
      "loss": 0.3272,
      "step": 44
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0004,
      "loss": 0.7366,
      "step": 45
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0004,
      "loss": 1.0331,
      "step": 46
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0004,
      "loss": 0.8471,
      "step": 47
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0004,
      "loss": 0.8171,
      "step": 48
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0004,
      "loss": 0.418,
      "step": 49
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0004,
      "loss": 0.6942,
      "step": 50
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0004,
      "loss": 0.1712,
      "step": 51
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0004,
      "loss": 0.5041,
      "step": 52
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0004,
      "loss": 0.4619,
      "step": 53
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0004,
      "loss": 0.5011,
      "step": 54
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0004,
      "loss": 0.5,
      "step": 55
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0004,
      "loss": 0.917,
      "step": 56
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0004,
      "loss": 0.8142,
      "step": 57
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0004,
      "loss": 0.8708,
      "step": 58
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0004,
      "loss": 0.7987,
      "step": 59
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0004,
      "loss": 0.9699,
      "step": 60
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0004,
      "loss": 1.0015,
      "step": 61
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0004,
      "loss": 0.8052,
      "step": 62
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0004,
      "loss": 0.8645,
      "step": 63
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0004,
      "loss": 1.1444,
      "step": 64
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0004,
      "loss": 1.0087,
      "step": 65
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0004,
      "loss": 0.5679,
      "step": 66
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0004,
      "loss": 0.5507,
      "step": 67
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0004,
      "loss": 1.1764,
      "step": 68
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0004,
      "loss": 0.6332,
      "step": 69
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0004,
      "loss": 0.8759,
      "step": 70
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0004,
      "loss": 0.8385,
      "step": 71
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0004,
      "loss": 0.7869,
      "step": 72
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0004,
      "loss": 1.4457,
      "step": 73
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0004,
      "loss": 0.9331,
      "step": 74
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0004,
      "loss": 0.8943,
      "step": 75
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0004,
      "loss": 0.5783,
      "step": 76
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0004,
      "loss": 0.6433,
      "step": 77
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0004,
      "loss": 0.7517,
      "step": 78
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0004,
      "loss": 0.5061,
      "step": 79
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0004,
      "loss": 0.7332,
      "step": 80
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0004,
      "loss": 1.6546,
      "step": 81
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0004,
      "loss": 0.4723,
      "step": 82
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0004,
      "loss": 0.2814,
      "step": 83
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0004,
      "loss": 0.4068,
      "step": 84
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0004,
      "loss": 1.1729,
      "step": 85
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0004,
      "loss": 0.698,
      "step": 86
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0004,
      "loss": 0.7746,
      "step": 87
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0004,
      "loss": 0.5065,
      "step": 88
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0004,
      "loss": 0.4654,
      "step": 89
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0004,
      "loss": 0.5724,
      "step": 90
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0004,
      "loss": 1.1535,
      "step": 91
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0004,
      "loss": 0.3854,
      "step": 92
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0004,
      "loss": 1.6848,
      "step": 93
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0004,
      "loss": 0.6667,
      "step": 94
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0004,
      "loss": 0.4526,
      "step": 95
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0004,
      "loss": 0.7112,
      "step": 96
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0004,
      "loss": 0.6816,
      "step": 97
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0004,
      "loss": 0.7496,
      "step": 98
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0004,
      "loss": 1.189,
      "step": 99
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0004,
      "loss": 1.4239,
      "step": 100
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0004,
      "loss": 0.4378,
      "step": 101
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0004,
      "loss": 2.3054,
      "step": 102
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0004,
      "loss": 0.1593,
      "step": 103
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0004,
      "loss": 0.2481,
      "step": 104
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0004,
      "loss": 0.6469,
      "step": 105
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0004,
      "loss": 0.7417,
      "step": 106
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0004,
      "loss": 0.7767,
      "step": 107
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0004,
      "loss": 0.831,
      "step": 108
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0004,
      "loss": 0.7954,
      "step": 109
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0004,
      "loss": 0.6376,
      "step": 110
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0004,
      "loss": 0.6208,
      "step": 111
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0004,
      "loss": 0.9038,
      "step": 112
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0004,
      "loss": 0.9994,
      "step": 113
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0004,
      "loss": 1.0233,
      "step": 114
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0004,
      "loss": 0.7573,
      "step": 115
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0004,
      "loss": 0.5333,
      "step": 116
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0004,
      "loss": 0.4886,
      "step": 117
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0004,
      "loss": 0.75,
      "step": 118
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0004,
      "loss": 0.6377,
      "step": 119
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0004,
      "loss": 0.6518,
      "step": 120
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0004,
      "loss": 0.6409,
      "step": 121
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0004,
      "loss": 0.5996,
      "step": 122
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0004,
      "loss": 0.7964,
      "step": 123
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0004,
      "loss": 0.5898,
      "step": 124
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0004,
      "loss": 0.8753,
      "step": 125
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0004,
      "loss": 0.6304,
      "step": 126
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0004,
      "loss": 0.5428,
      "step": 127
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0004,
      "loss": 0.8571,
      "step": 128
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0004,
      "loss": 0.8339,
      "step": 129
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0004,
      "loss": 0.6087,
      "step": 130
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0004,
      "loss": 0.3547,
      "step": 131
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0004,
      "loss": 0.3494,
      "step": 132
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0004,
      "loss": 0.5394,
      "step": 133
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0004,
      "loss": 0.4315,
      "step": 134
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0004,
      "loss": 0.3575,
      "step": 135
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0004,
      "loss": 0.4983,
      "step": 136
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0004,
      "loss": 0.4292,
      "step": 137
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0004,
      "loss": 0.6315,
      "step": 138
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0004,
      "loss": 0.9827,
      "step": 139
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0004,
      "loss": 1.4456,
      "step": 140
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0004,
      "loss": 0.5151,
      "step": 141
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0004,
      "loss": 0.4651,
      "step": 142
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0004,
      "loss": 0.8682,
      "step": 143
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0004,
      "loss": 1.1543,
      "step": 144
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0004,
      "loss": 0.3378,
      "step": 145
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0004,
      "loss": 0.5381,
      "step": 146
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0004,
      "loss": 0.5189,
      "step": 147
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0004,
      "loss": 0.3988,
      "step": 148
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0004,
      "loss": 0.9694,
      "step": 149
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0004,
      "loss": 1.2404,
      "step": 150
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0004,
      "loss": 0.3767,
      "step": 151
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0004,
      "loss": 0.5564,
      "step": 152
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0004,
      "loss": 0.2665,
      "step": 153
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0004,
      "loss": 0.172,
      "step": 154
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0004,
      "loss": 0.6584,
      "step": 155
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0004,
      "loss": 0.7588,
      "step": 156
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0004,
      "loss": 0.9431,
      "step": 157
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0004,
      "loss": 0.9456,
      "step": 158
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0004,
      "loss": 0.7471,
      "step": 159
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0004,
      "loss": 0.8477,
      "step": 160
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0004,
      "loss": 0.7679,
      "step": 161
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0004,
      "loss": 0.8937,
      "step": 162
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0004,
      "loss": 1.2858,
      "step": 163
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0004,
      "loss": 0.7806,
      "step": 164
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0004,
      "loss": 0.8034,
      "step": 165
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0004,
      "loss": 0.6372,
      "step": 166
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0004,
      "loss": 0.8217,
      "step": 167
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0004,
      "loss": 0.3706,
      "step": 168
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0004,
      "loss": 1.0851,
      "step": 169
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0004,
      "loss": 0.6719,
      "step": 170
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0004,
      "loss": 1.0366,
      "step": 171
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0004,
      "loss": 0.5603,
      "step": 172
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0004,
      "loss": 0.5651,
      "step": 173
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0004,
      "loss": 0.3804,
      "step": 174
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0004,
      "loss": 0.5946,
      "step": 175
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0004,
      "loss": 0.7649,
      "step": 176
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0004,
      "loss": 0.5035,
      "step": 177
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0004,
      "loss": 0.8066,
      "step": 178
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0004,
      "loss": 0.8046,
      "step": 179
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0004,
      "loss": 0.4233,
      "step": 180
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0004,
      "loss": 0.7945,
      "step": 181
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0004,
      "loss": 0.5722,
      "step": 182
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0004,
      "loss": 0.6088,
      "step": 183
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0004,
      "loss": 0.4229,
      "step": 184
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0004,
      "loss": 0.8723,
      "step": 185
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0004,
      "loss": 0.9287,
      "step": 186
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0004,
      "loss": 0.8082,
      "step": 187
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0004,
      "loss": 0.4144,
      "step": 188
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0004,
      "loss": 0.372,
      "step": 189
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0004,
      "loss": 0.5725,
      "step": 190
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0004,
      "loss": 0.3837,
      "step": 191
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0004,
      "loss": 0.5155,
      "step": 192
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0004,
      "loss": 0.5472,
      "step": 193
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0004,
      "loss": 0.6488,
      "step": 194
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0004,
      "loss": 1.4385,
      "step": 195
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0004,
      "loss": 0.3418,
      "step": 196
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0004,
      "loss": 0.8811,
      "step": 197
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0004,
      "loss": 0.8176,
      "step": 198
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0004,
      "loss": 0.5379,
      "step": 199
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0004,
      "loss": 1.0277,
      "step": 200
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.662463366985321,
      "eval_runtime": 262.5709,
      "eval_samples_per_second": 1.904,
      "eval_steps_per_second": 0.952,
      "step": 200
    },
    {
      "epoch": 0.14,
      "mmlu_eval_accuracy": 0.41276444254447825,
      "mmlu_eval_accuracy_abstract_algebra": 0.09090909090909091,
      "mmlu_eval_accuracy_anatomy": 0.42857142857142855,
      "mmlu_eval_accuracy_astronomy": 0.5,
      "mmlu_eval_accuracy_business_ethics": 0.2727272727272727,
      "mmlu_eval_accuracy_clinical_knowledge": 0.4482758620689655,
      "mmlu_eval_accuracy_college_biology": 0.3125,
      "mmlu_eval_accuracy_college_chemistry": 0.375,
      "mmlu_eval_accuracy_college_computer_science": 0.5454545454545454,
      "mmlu_eval_accuracy_college_mathematics": 0.2727272727272727,
      "mmlu_eval_accuracy_college_medicine": 0.6363636363636364,
      "mmlu_eval_accuracy_college_physics": 0.2727272727272727,
      "mmlu_eval_accuracy_computer_security": 0.36363636363636365,
      "mmlu_eval_accuracy_conceptual_physics": 0.3076923076923077,
      "mmlu_eval_accuracy_econometrics": 0.4166666666666667,
      "mmlu_eval_accuracy_electrical_engineering": 0.5,
      "mmlu_eval_accuracy_elementary_mathematics": 0.4146341463414634,
      "mmlu_eval_accuracy_formal_logic": 0.5714285714285714,
      "mmlu_eval_accuracy_global_facts": 0.2,
      "mmlu_eval_accuracy_high_school_biology": 0.1875,
      "mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
      "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_european_history": 0.4444444444444444,
      "mmlu_eval_accuracy_high_school_geography": 0.5909090909090909,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.5714285714285714,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.32558139534883723,
      "mmlu_eval_accuracy_high_school_mathematics": 0.4482758620689655,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.5714285714285714,
      "mmlu_loss": 0.7435930466353893,
      "step": 200
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0004,
      "loss": 1.6342,
      "step": 201
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0004,
      "loss": 0.1179,
      "step": 202
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0004,
      "loss": 0.1083,
      "step": 203
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0004,
      "loss": 0.6355,
      "step": 204
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0004,
      "loss": 0.5694,
      "step": 205
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0004,
      "loss": 0.7707,
      "step": 206
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0004,
      "loss": 0.3314,
      "step": 207
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0004,
      "loss": 1.0069,
      "step": 208
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0004,
      "loss": 0.9113,
      "step": 209
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0004,
      "loss": 0.8692,
      "step": 210
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0004,
      "loss": 0.7362,
      "step": 211
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0004,
      "loss": 0.926,
      "step": 212
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0004,
      "loss": 0.7405,
      "step": 213
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0004,
      "loss": 0.9191,
      "step": 214
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0004,
      "loss": 0.9301,
      "step": 215
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0004,
      "loss": 0.8581,
      "step": 216
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0004,
      "loss": 0.4548,
      "step": 217
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0004,
      "loss": 0.6516,
      "step": 218
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0004,
      "loss": 0.7194,
      "step": 219
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0004,
      "loss": 1.0441,
      "step": 220
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004,
      "loss": 0.9499,
      "step": 221
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004,
      "loss": 0.5447,
      "step": 222
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004,
      "loss": 0.8795,
      "step": 223
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004,
      "loss": 0.5347,
      "step": 224
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004,
      "loss": 0.5827,
      "step": 225
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004,
      "loss": 0.311,
      "step": 226
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004,
      "loss": 0.1071,
      "step": 227
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004,
      "loss": 0.6336,
      "step": 228
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004,
      "loss": 0.6212,
      "step": 229
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004,
      "loss": 0.4345,
      "step": 230
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004,
      "loss": 0.5373,
      "step": 231
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004,
      "loss": 0.9612,
      "step": 232
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004,
      "loss": 1.0251,
      "step": 233
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004,
      "loss": 0.4267,
      "step": 234
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0004,
      "loss": 0.5767,
      "step": 235
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0004,
      "loss": 0.5123,
      "step": 236
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0004,
      "loss": 0.354,
      "step": 237
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0004,
      "loss": 1.1279,
      "step": 238
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0004,
      "loss": 0.6464,
      "step": 239
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0004,
      "loss": 0.7227,
      "step": 240
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0004,
      "loss": 1.3533,
      "step": 241
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0004,
      "loss": 0.3986,
      "step": 242
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0004,
      "loss": 1.0402,
      "step": 243
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0004,
      "loss": 0.4558,
      "step": 244
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0004,
      "loss": 0.5525,
      "step": 245
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0004,
      "loss": 1.5916,
      "step": 246
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0004,
      "loss": 0.9311,
      "step": 247
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0004,
      "loss": 0.3011,
      "step": 248
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0004,
      "loss": 0.5813,
      "step": 249
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0004,
      "loss": 1.134,
      "step": 250
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0004,
      "loss": 0.3062,
      "step": 251
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0004,
      "loss": 0.2848,
      "step": 252
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0004,
      "loss": 0.1995,
      "step": 253
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0004,
      "loss": 0.8426,
      "step": 254
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0004,
      "loss": 0.8062,
      "step": 255
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0004,
      "loss": 1.0046,
      "step": 256
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0004,
      "loss": 0.6235,
      "step": 257
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0004,
      "loss": 1.0021,
      "step": 258
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0004,
      "loss": 0.8788,
      "step": 259
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0004,
      "loss": 0.883,
      "step": 260
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0004,
      "loss": 0.8167,
      "step": 261
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0004,
      "loss": 0.8068,
      "step": 262
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0004,
      "loss": 0.4259,
      "step": 263
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0004,
      "loss": 0.8339,
      "step": 264
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0004,
      "loss": 0.8149,
      "step": 265
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0004,
      "loss": 0.6023,
      "step": 266
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0004,
      "loss": 0.8297,
      "step": 267
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0004,
      "loss": 1.0205,
      "step": 268
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0004,
      "loss": 0.5643,
      "step": 269
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0004,
      "loss": 0.719,
      "step": 270
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0004,
      "loss": 0.812,
      "step": 271
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0004,
      "loss": 0.5513,
      "step": 272
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0004,
      "loss": 0.7669,
      "step": 273
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0004,
      "loss": 0.6044,
      "step": 274
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0004,
      "loss": 0.7956,
      "step": 275
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0004,
      "loss": 0.569,
      "step": 276
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0004,
      "loss": 0.5728,
      "step": 277
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0004,
      "loss": 0.534,
      "step": 278
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0004,
      "loss": 0.4182,
      "step": 279
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0004,
      "loss": 0.9406,
      "step": 280
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0004,
      "loss": 0.8158,
      "step": 281
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0004,
      "loss": 0.674,
      "step": 282
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0004,
      "loss": 0.4463,
      "step": 283
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0004,
      "loss": 0.5038,
      "step": 284
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0004,
      "loss": 0.5103,
      "step": 285
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0004,
      "loss": 0.4692,
      "step": 286
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0004,
      "loss": 0.4955,
      "step": 287
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0004,
      "loss": 1.068,
      "step": 288
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0004,
      "loss": 0.5165,
      "step": 289
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0004,
      "loss": 0.6199,
      "step": 290
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0004,
      "loss": 1.1037,
      "step": 291
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0004,
      "loss": 0.155,
      "step": 292
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004,
      "loss": 1.4112,
      "step": 293
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004,
      "loss": 0.5906,
      "step": 294
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004,
      "loss": 0.5932,
      "step": 295
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004,
      "loss": 0.8826,
      "step": 296
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004,
      "loss": 0.5626,
      "step": 297
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004,
      "loss": 0.3403,
      "step": 298
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004,
      "loss": 1.3782,
      "step": 299
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004,
      "loss": 1.2619,
      "step": 300
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004,
      "loss": 0.4879,
      "step": 301
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004,
      "loss": 0.426,
      "step": 302
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004,
      "loss": 0.2672,
      "step": 303
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004,
      "loss": 0.4627,
      "step": 304
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004,
      "loss": 0.8008,
      "step": 305
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0004,
      "loss": 0.8396,
      "step": 306
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0004,
      "loss": 0.826,
      "step": 307
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0004,
      "loss": 0.9369,
      "step": 308
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0004,
      "loss": 0.8878,
      "step": 309
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0004,
      "loss": 0.8131,
      "step": 310
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0004,
      "loss": 0.8733,
      "step": 311
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0004,
      "loss": 0.6789,
      "step": 312
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0004,
      "loss": 0.9527,
      "step": 313
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0004,
      "loss": 0.4861,
      "step": 314
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0004,
      "loss": 0.6534,
      "step": 315
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0004,
      "loss": 0.7253,
      "step": 316
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0004,
      "loss": 0.626,
      "step": 317
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0004,
      "loss": 0.7153,
      "step": 318
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0004,
      "loss": 0.5901,
      "step": 319
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0004,
      "loss": 0.5152,
      "step": 320
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0004,
      "loss": 0.6235,
      "step": 321
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0004,
      "loss": 0.7336,
      "step": 322
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0004,
      "loss": 0.8437,
      "step": 323
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0004,
      "loss": 0.8042,
      "step": 324
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0004,
      "loss": 0.524,
      "step": 325
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0004,
      "loss": 0.882,
      "step": 326
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0004,
      "loss": 0.5542,
      "step": 327
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0004,
      "loss": 0.5666,
      "step": 328
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0004,
      "loss": 0.7223,
      "step": 329
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0004,
      "loss": 0.4367,
      "step": 330
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0004,
      "loss": 0.6358,
      "step": 331
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0004,
      "loss": 0.6793,
      "step": 332
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0004,
      "loss": 0.3485,
      "step": 333
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0004,
      "loss": 0.6813,
      "step": 334
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0004,
      "loss": 0.8295,
      "step": 335
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0004,
      "loss": 0.4777,
      "step": 336
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0004,
      "loss": 0.8557,
      "step": 337
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0004,
      "loss": 0.7415,
      "step": 338
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0004,
      "loss": 0.5843,
      "step": 339
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0004,
      "loss": 0.4585,
      "step": 340
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0004,
      "loss": 0.2676,
      "step": 341
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0004,
      "loss": 0.2997,
      "step": 342
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0004,
      "loss": 0.5732,
      "step": 343
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0004,
      "loss": 0.3759,
      "step": 344
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0004,
      "loss": 0.4395,
      "step": 345
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0004,
      "loss": 0.4864,
      "step": 346
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0004,
      "loss": 0.3492,
      "step": 347
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0004,
      "loss": 0.5787,
      "step": 348
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0004,
      "loss": 0.8384,
      "step": 349
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0004,
      "loss": 1.2399,
      "step": 350
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0004,
      "loss": 0.248,
      "step": 351
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0004,
      "loss": 0.2635,
      "step": 352
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0004,
      "loss": 0.3762,
      "step": 353
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0004,
      "loss": 0.5572,
      "step": 354
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0004,
      "loss": 1.0953,
      "step": 355
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0004,
      "loss": 0.8513,
      "step": 356
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0004,
      "loss": 0.8504,
      "step": 357
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0004,
      "loss": 0.9141,
      "step": 358
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0004,
      "loss": 0.7751,
      "step": 359
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0004,
      "loss": 0.8659,
      "step": 360
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0004,
      "loss": 0.9634,
      "step": 361
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0004,
      "loss": 0.9354,
      "step": 362
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0004,
      "loss": 0.6382,
      "step": 363
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0004,
      "loss": 0.9403,
      "step": 364
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0004,
      "loss": 0.8036,
      "step": 365
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0004,
      "loss": 0.5506,
      "step": 366
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0004,
      "loss": 0.4639,
      "step": 367
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0004,
      "loss": 0.8315,
      "step": 368
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0004,
      "loss": 0.6634,
      "step": 369
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0004,
      "loss": 0.6388,
      "step": 370
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0004,
      "loss": 0.7629,
      "step": 371
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0004,
      "loss": 0.6548,
      "step": 372
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0004,
      "loss": 0.5165,
      "step": 373
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0004,
      "loss": 0.2858,
      "step": 374
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0004,
      "loss": 0.3376,
      "step": 375
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0004,
      "loss": 0.5373,
      "step": 376
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0004,
      "loss": 0.6229,
      "step": 377
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004,
      "loss": 0.5867,
      "step": 378
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004,
      "loss": 0.5533,
      "step": 379
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004,
      "loss": 0.4804,
      "step": 380
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004,
      "loss": 0.313,
      "step": 381
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004,
      "loss": 0.7077,
      "step": 382
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004,
      "loss": 0.5241,
      "step": 383
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004,
      "loss": 0.3367,
      "step": 384
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004,
      "loss": 0.8614,
      "step": 385
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004,
      "loss": 0.4548,
      "step": 386
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004,
      "loss": 0.6221,
      "step": 387
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004,
      "loss": 0.8041,
      "step": 388
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004,
      "loss": 0.2546,
      "step": 389
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004,
      "loss": 0.5016,
      "step": 390
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004,
      "loss": 0.4664,
      "step": 391
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0004,
      "loss": 0.2124,
      "step": 392
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0004,
      "loss": 0.4019,
      "step": 393
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0004,
      "loss": 0.8749,
      "step": 394
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0004,
      "loss": 0.2319,
      "step": 395
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0004,
      "loss": 0.9602,
      "step": 396
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0004,
      "loss": 0.9182,
      "step": 397
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0004,
      "loss": 0.2092,
      "step": 398
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0004,
      "loss": 0.2311,
      "step": 399
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0004,
      "loss": 1.6764,
      "step": 400
    },
    {
      "epoch": 0.28,
      "eval_loss": 0.6333504915237427,
      "eval_runtime": 261.9428,
      "eval_samples_per_second": 1.909,
      "eval_steps_per_second": 0.954,
      "step": 400
    },
    {
      "epoch": 0.28,
      "mmlu_eval_accuracy": 0.4098082068017773,
      "mmlu_eval_accuracy_abstract_algebra": 0.09090909090909091,
      "mmlu_eval_accuracy_anatomy": 0.2857142857142857,
      "mmlu_eval_accuracy_astronomy": 0.375,
      "mmlu_eval_accuracy_business_ethics": 0.36363636363636365,
      "mmlu_eval_accuracy_clinical_knowledge": 0.4482758620689655,
      "mmlu_eval_accuracy_college_biology": 0.4375,
      "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": 0.5,
      "mmlu_eval_accuracy_college_physics": 0.36363636363636365,
      "mmlu_eval_accuracy_computer_security": 0.5454545454545454,
      "mmlu_eval_accuracy_conceptual_physics": 0.34615384615384615,
      "mmlu_eval_accuracy_econometrics": 0.16666666666666666,
      "mmlu_eval_accuracy_electrical_engineering": 0.5625,
      "mmlu_eval_accuracy_elementary_mathematics": 0.4146341463414634,
      "mmlu_eval_accuracy_formal_logic": 0.35714285714285715,
      "mmlu_eval_accuracy_global_facts": 0.2,
      "mmlu_eval_accuracy_high_school_biology": 0.3125,
      "mmlu_eval_accuracy_high_school_chemistry": 0.36363636363636365,
      "mmlu_eval_accuracy_high_school_computer_science": 0.7777777777777778,
      "mmlu_eval_accuracy_high_school_european_history": 0.6111111111111112,
      "mmlu_eval_accuracy_high_school_geography": 0.7272727272727273,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.5714285714285714,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.3953488372093023,
      "mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.35714285714285715,
      "mmlu_loss": 0.6790264239534736,
      "step": 400
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0004,
      "loss": 0.3098,
      "step": 401
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0004,
      "loss": 0.0669,
      "step": 402
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0004,
      "loss": 0.3434,
      "step": 403
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0004,
      "loss": 0.6937,
      "step": 404
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0004,
      "loss": 0.6958,
      "step": 405
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0004,
      "loss": 0.7289,
      "step": 406
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0004,
      "loss": 0.804,
      "step": 407
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0004,
      "loss": 0.7795,
      "step": 408
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0004,
      "loss": 0.4095,
      "step": 409
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0004,
      "loss": 0.5514,
      "step": 410
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0004,
      "loss": 0.9134,
      "step": 411
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0004,
      "loss": 0.6661,
      "step": 412
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0004,
      "loss": 0.8238,
      "step": 413
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0004,
      "loss": 0.6895,
      "step": 414
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0004,
      "loss": 0.6924,
      "step": 415
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0004,
      "loss": 0.5246,
      "step": 416
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0004,
      "loss": 0.6762,
      "step": 417
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0004,
      "loss": 0.4997,
      "step": 418
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0004,
      "loss": 0.7983,
      "step": 419
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0004,
      "loss": 0.5747,
      "step": 420
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0004,
      "loss": 1.1613,
      "step": 421
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0004,
      "loss": 0.6814,
      "step": 422
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0004,
      "loss": 0.97,
      "step": 423
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0004,
      "loss": 0.4104,
      "step": 424
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0004,
      "loss": 0.5966,
      "step": 425
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0004,
      "loss": 0.7571,
      "step": 426
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0004,
      "loss": 1.2513,
      "step": 427
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0004,
      "loss": 0.594,
      "step": 428
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0004,
      "loss": 0.9536,
      "step": 429
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0004,
      "loss": 0.8351,
      "step": 430
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0004,
      "loss": 0.3044,
      "step": 431
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0004,
      "loss": 0.893,
      "step": 432
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0004,
      "loss": 0.8969,
      "step": 433
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0004,
      "loss": 0.7612,
      "step": 434
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0004,
      "loss": 0.2974,
      "step": 435
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0004,
      "loss": 0.5277,
      "step": 436
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0004,
      "loss": 0.3194,
      "step": 437
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0004,
      "loss": 1.1187,
      "step": 438
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0004,
      "loss": 0.5409,
      "step": 439
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0004,
      "loss": 0.4445,
      "step": 440
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0004,
      "loss": 0.3914,
      "step": 441
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0004,
      "loss": 2.116,
      "step": 442
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0004,
      "loss": 0.4477,
      "step": 443
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0004,
      "loss": 1.4711,
      "step": 444
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0004,
      "loss": 0.6186,
      "step": 445
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0004,
      "loss": 0.3651,
      "step": 446
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0004,
      "loss": 0.2939,
      "step": 447
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0004,
      "loss": 0.2754,
      "step": 448
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0004,
      "loss": 0.7015,
      "step": 449
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0004,
      "loss": 1.2934,
      "step": 450
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0004,
      "loss": 0.3308,
      "step": 451
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0004,
      "loss": 0.3002,
      "step": 452
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0004,
      "loss": 0.3875,
      "step": 453
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0004,
      "loss": 0.1527,
      "step": 454
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0004,
      "loss": 0.791,
      "step": 455
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0004,
      "loss": 0.6592,
      "step": 456
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0004,
      "loss": 0.8466,
      "step": 457
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0004,
      "loss": 0.7768,
      "step": 458
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0004,
      "loss": 0.7784,
      "step": 459
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0004,
      "loss": 0.8584,
      "step": 460
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0004,
      "loss": 0.8313,
      "step": 461
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0004,
      "loss": 0.57,
      "step": 462
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0004,
      "loss": 0.7299,
      "step": 463
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0004,
      "loss": 0.6277,
      "step": 464
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0004,
      "loss": 0.6326,
      "step": 465
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0004,
      "loss": 0.7089,
      "step": 466
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0004,
      "loss": 0.6943,
      "step": 467
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0004,
      "loss": 0.5197,
      "step": 468
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0004,
      "loss": 0.79,
      "step": 469
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0004,
      "loss": 0.5237,
      "step": 470
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0004,
      "loss": 0.5653,
      "step": 471
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0004,
      "loss": 0.5509,
      "step": 472
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0004,
      "loss": 0.9335,
      "step": 473
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0004,
      "loss": 0.816,
      "step": 474
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0004,
      "loss": 0.7155,
      "step": 475
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0004,
      "loss": 0.5283,
      "step": 476
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0004,
      "loss": 0.1808,
      "step": 477
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0004,
      "loss": 0.5369,
      "step": 478
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0004,
      "loss": 0.6086,
      "step": 479
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0004,
      "loss": 0.6205,
      "step": 480
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0004,
      "loss": 1.3518,
      "step": 481
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0004,
      "loss": 0.6041,
      "step": 482
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0004,
      "loss": 0.4709,
      "step": 483
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0004,
      "loss": 0.8976,
      "step": 484
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0004,
      "loss": 0.2544,
      "step": 485
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0004,
      "loss": 1.0326,
      "step": 486
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0004,
      "loss": 0.2799,
      "step": 487
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0004,
      "loss": 1.4241,
      "step": 488
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0004,
      "loss": 0.3467,
      "step": 489
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0004,
      "loss": 0.8382,
      "step": 490
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0004,
      "loss": 0.2025,
      "step": 491
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0004,
      "loss": 0.4382,
      "step": 492
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0004,
      "loss": 0.5643,
      "step": 493
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0004,
      "loss": 0.6231,
      "step": 494
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0004,
      "loss": 1.0425,
      "step": 495
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0004,
      "loss": 0.8847,
      "step": 496
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0004,
      "loss": 0.6748,
      "step": 497
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0004,
      "loss": 0.3893,
      "step": 498
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0004,
      "loss": 0.3734,
      "step": 499
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0004,
      "loss": 1.6683,
      "step": 500
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0004,
      "loss": 0.2681,
      "step": 501
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0004,
      "loss": 0.2262,
      "step": 502
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0004,
      "loss": 0.5853,
      "step": 503
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0004,
      "loss": 0.928,
      "step": 504
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0004,
      "loss": 0.8761,
      "step": 505
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0004,
      "loss": 0.8541,
      "step": 506
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0004,
      "loss": 0.4984,
      "step": 507
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0004,
      "loss": 0.9247,
      "step": 508
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0004,
      "loss": 0.8992,
      "step": 509
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0004,
      "loss": 0.8758,
      "step": 510
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0004,
      "loss": 0.6687,
      "step": 511
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0004,
      "loss": 0.7952,
      "step": 512
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0004,
      "loss": 0.8696,
      "step": 513
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0004,
      "loss": 0.761,
      "step": 514
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0004,
      "loss": 0.8351,
      "step": 515
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0004,
      "loss": 0.8167,
      "step": 516
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0004,
      "loss": 0.6166,
      "step": 517
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0004,
      "loss": 0.7103,
      "step": 518
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0004,
      "loss": 0.9162,
      "step": 519
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004,
      "loss": 0.567,
      "step": 520
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004,
      "loss": 0.9573,
      "step": 521
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004,
      "loss": 0.4096,
      "step": 522
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004,
      "loss": 0.6648,
      "step": 523
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004,
      "loss": 0.6851,
      "step": 524
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004,
      "loss": 0.4083,
      "step": 525
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004,
      "loss": 0.8178,
      "step": 526
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004,
      "loss": 1.1736,
      "step": 527
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004,
      "loss": 0.5406,
      "step": 528
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004,
      "loss": 0.552,
      "step": 529
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004,
      "loss": 0.6497,
      "step": 530
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004,
      "loss": 0.4467,
      "step": 531
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004,
      "loss": 0.9224,
      "step": 532
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004,
      "loss": 0.9014,
      "step": 533
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0004,
      "loss": 0.8595,
      "step": 534
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0004,
      "loss": 0.343,
      "step": 535
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0004,
      "loss": 0.6365,
      "step": 536
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0004,
      "loss": 0.8377,
      "step": 537
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0004,
      "loss": 0.5699,
      "step": 538
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0004,
      "loss": 1.0938,
      "step": 539
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0004,
      "loss": 0.2896,
      "step": 540
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0004,
      "loss": 1.2605,
      "step": 541
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0004,
      "loss": 0.5222,
      "step": 542
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0004,
      "loss": 0.327,
      "step": 543
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0004,
      "loss": 0.7968,
      "step": 544
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0004,
      "loss": 0.646,
      "step": 545
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0004,
      "loss": 0.8071,
      "step": 546
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0004,
      "loss": 0.3894,
      "step": 547
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0004,
      "loss": 0.4803,
      "step": 548
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0004,
      "loss": 0.3228,
      "step": 549
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0004,
      "loss": 1.3897,
      "step": 550
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0004,
      "loss": 0.1901,
      "step": 551
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0004,
      "loss": 0.321,
      "step": 552
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0004,
      "loss": 0.8044,
      "step": 553
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0004,
      "loss": 0.8365,
      "step": 554
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0004,
      "loss": 0.7818,
      "step": 555
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0004,
      "loss": 0.4648,
      "step": 556
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0004,
      "loss": 0.5499,
      "step": 557
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0004,
      "loss": 0.7847,
      "step": 558
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0004,
      "loss": 0.7892,
      "step": 559
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0004,
      "loss": 0.8517,
      "step": 560
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0004,
      "loss": 0.7914,
      "step": 561
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0004,
      "loss": 0.893,
      "step": 562
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0004,
      "loss": 0.8116,
      "step": 563
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0004,
      "loss": 0.8543,
      "step": 564
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0004,
      "loss": 0.8607,
      "step": 565
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0004,
      "loss": 0.6274,
      "step": 566
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0004,
      "loss": 0.7551,
      "step": 567
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0004,
      "loss": 0.7328,
      "step": 568
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0004,
      "loss": 0.7561,
      "step": 569
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0004,
      "loss": 0.7465,
      "step": 570
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0004,
      "loss": 0.697,
      "step": 571
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0004,
      "loss": 0.8938,
      "step": 572
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0004,
      "loss": 0.8216,
      "step": 573
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0004,
      "loss": 0.5859,
      "step": 574
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0004,
      "loss": 0.4194,
      "step": 575
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0004,
      "loss": 0.603,
      "step": 576
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0004,
      "loss": 0.6666,
      "step": 577
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0004,
      "loss": 0.4037,
      "step": 578
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0004,
      "loss": 0.6723,
      "step": 579
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0004,
      "loss": 0.5837,
      "step": 580
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0004,
      "loss": 0.534,
      "step": 581
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0004,
      "loss": 0.44,
      "step": 582
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0004,
      "loss": 0.8701,
      "step": 583
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0004,
      "loss": 0.6575,
      "step": 584
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0004,
      "loss": 0.3981,
      "step": 585
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0004,
      "loss": 0.4929,
      "step": 586
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0004,
      "loss": 0.6346,
      "step": 587
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0004,
      "loss": 0.88,
      "step": 588
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0004,
      "loss": 0.3083,
      "step": 589
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0004,
      "loss": 0.741,
      "step": 590
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0004,
      "loss": 0.3114,
      "step": 591
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0004,
      "loss": 0.1633,
      "step": 592
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0004,
      "loss": 1.5786,
      "step": 593
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0004,
      "loss": 0.9141,
      "step": 594
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0004,
      "loss": 0.4553,
      "step": 595
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0004,
      "loss": 0.4153,
      "step": 596
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0004,
      "loss": 0.8077,
      "step": 597
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0004,
      "loss": 0.5165,
      "step": 598
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0004,
      "loss": 1.2809,
      "step": 599
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0004,
      "loss": 0.1944,
      "step": 600
    },
    {
      "epoch": 0.42,
      "eval_loss": 0.6178187727928162,
      "eval_runtime": 262.0082,
      "eval_samples_per_second": 1.908,
      "eval_steps_per_second": 0.954,
      "step": 600
    },
    {
      "epoch": 0.42,
      "mmlu_eval_accuracy": 0.3933105593085208,
      "mmlu_eval_accuracy_abstract_algebra": 0.18181818181818182,
      "mmlu_eval_accuracy_anatomy": 0.35714285714285715,
      "mmlu_eval_accuracy_astronomy": 0.5,
      "mmlu_eval_accuracy_business_ethics": 0.36363636363636365,
      "mmlu_eval_accuracy_clinical_knowledge": 0.5517241379310345,
      "mmlu_eval_accuracy_college_biology": 0.4375,
      "mmlu_eval_accuracy_college_chemistry": 0.25,
      "mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
      "mmlu_eval_accuracy_college_mathematics": 0.09090909090909091,
      "mmlu_eval_accuracy_college_medicine": 0.5,
      "mmlu_eval_accuracy_college_physics": 0.6363636363636364,
      "mmlu_eval_accuracy_computer_security": 0.36363636363636365,
      "mmlu_eval_accuracy_conceptual_physics": 0.3076923076923077,
      "mmlu_eval_accuracy_econometrics": 0.3333333333333333,
      "mmlu_eval_accuracy_electrical_engineering": 0.5,
      "mmlu_eval_accuracy_elementary_mathematics": 0.21951219512195122,
      "mmlu_eval_accuracy_formal_logic": 0.35714285714285715,
      "mmlu_eval_accuracy_global_facts": 0.3,
      "mmlu_eval_accuracy_high_school_biology": 0.3125,
      "mmlu_eval_accuracy_high_school_chemistry": 0.18181818181818182,
      "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_european_history": 0.5555555555555556,
      "mmlu_eval_accuracy_high_school_geography": 0.7272727272727273,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.42857142857142855,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.37209302325581395,
      "mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.42857142857142855,
      "mmlu_loss": 0.770003160238266,
      "step": 600
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0004,
      "loss": 0.2039,
      "step": 601
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0004,
      "loss": 0.3086,
      "step": 602
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0004,
      "loss": 0.4326,
      "step": 603
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0004,
      "loss": 0.8444,
      "step": 604
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0004,
      "loss": 0.6902,
      "step": 605
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004,
      "loss": 0.8338,
      "step": 606
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004,
      "loss": 0.9103,
      "step": 607
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004,
      "loss": 0.9047,
      "step": 608
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004,
      "loss": 0.82,
      "step": 609
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004,
      "loss": 0.4912,
      "step": 610
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004,
      "loss": 0.414,
      "step": 611
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004,
      "loss": 0.735,
      "step": 612
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004,
      "loss": 0.7284,
      "step": 613
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004,
      "loss": 0.5756,
      "step": 614
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004,
      "loss": 0.7305,
      "step": 615
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004,
      "loss": 0.5331,
      "step": 616
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004,
      "loss": 0.6001,
      "step": 617
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004,
      "loss": 0.3613,
      "step": 618
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004,
      "loss": 0.6111,
      "step": 619
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0004,
      "loss": 0.3517,
      "step": 620
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0004,
      "loss": 0.7228,
      "step": 621
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0004,
      "loss": 0.512,
      "step": 622
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0004,
      "loss": 0.4729,
      "step": 623
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0004,
      "loss": 0.6152,
      "step": 624
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0004,
      "loss": 0.6118,
      "step": 625
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0004,
      "loss": 0.9996,
      "step": 626
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0004,
      "loss": 0.2561,
      "step": 627
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0004,
      "loss": 0.375,
      "step": 628
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0004,
      "loss": 0.5303,
      "step": 629
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0004,
      "loss": 0.3277,
      "step": 630
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0004,
      "loss": 0.286,
      "step": 631
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0004,
      "loss": 0.4288,
      "step": 632
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0004,
      "loss": 0.652,
      "step": 633
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0004,
      "loss": 0.6116,
      "step": 634
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0004,
      "loss": 0.9324,
      "step": 635
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0004,
      "loss": 0.4822,
      "step": 636
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0004,
      "loss": 0.6417,
      "step": 637
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0004,
      "loss": 0.4586,
      "step": 638
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0004,
      "loss": 0.2927,
      "step": 639
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0004,
      "loss": 0.7887,
      "step": 640
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0004,
      "loss": 0.6607,
      "step": 641
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0004,
      "loss": 0.3745,
      "step": 642
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0004,
      "loss": 0.4677,
      "step": 643
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0004,
      "loss": 0.5891,
      "step": 644
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0004,
      "loss": 0.3206,
      "step": 645
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0004,
      "loss": 0.4792,
      "step": 646
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0004,
      "loss": 0.4851,
      "step": 647
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0004,
      "loss": 0.7103,
      "step": 648
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0004,
      "loss": 0.5212,
      "step": 649
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0004,
      "loss": 0.2019,
      "step": 650
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0004,
      "loss": 0.084,
      "step": 651
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0004,
      "loss": 0.4259,
      "step": 652
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0004,
      "loss": 0.7373,
      "step": 653
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0004,
      "loss": 0.8373,
      "step": 654
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0004,
      "loss": 0.887,
      "step": 655
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0004,
      "loss": 0.913,
      "step": 656
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0004,
      "loss": 0.9228,
      "step": 657
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0004,
      "loss": 0.7886,
      "step": 658
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0004,
      "loss": 0.8073,
      "step": 659
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0004,
      "loss": 0.7421,
      "step": 660
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0004,
      "loss": 1.0695,
      "step": 661
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0004,
      "loss": 0.719,
      "step": 662
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0004,
      "loss": 0.8423,
      "step": 663
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0004,
      "loss": 0.6688,
      "step": 664
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0004,
      "loss": 0.412,
      "step": 665
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0004,
      "loss": 0.7414,
      "step": 666
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0004,
      "loss": 0.7357,
      "step": 667
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0004,
      "loss": 0.5231,
      "step": 668
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0004,
      "loss": 0.6011,
      "step": 669
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0004,
      "loss": 0.4634,
      "step": 670
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0004,
      "loss": 0.2916,
      "step": 671
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0004,
      "loss": 0.9223,
      "step": 672
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0004,
      "loss": 0.8498,
      "step": 673
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0004,
      "loss": 0.6754,
      "step": 674
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0004,
      "loss": 0.5132,
      "step": 675
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0004,
      "loss": 0.6087,
      "step": 676
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0004,
      "loss": 0.7169,
      "step": 677
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0004,
      "loss": 0.4727,
      "step": 678
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0004,
      "loss": 0.8692,
      "step": 679
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0004,
      "loss": 0.1934,
      "step": 680
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0004,
      "loss": 0.2617,
      "step": 681
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0004,
      "loss": 0.6538,
      "step": 682
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0004,
      "loss": 0.3225,
      "step": 683
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0004,
      "loss": 0.7903,
      "step": 684
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0004,
      "loss": 0.6982,
      "step": 685
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0004,
      "loss": 0.3023,
      "step": 686
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0004,
      "loss": 0.5437,
      "step": 687
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0004,
      "loss": 0.2446,
      "step": 688
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0004,
      "loss": 0.4276,
      "step": 689
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0004,
      "loss": 0.5117,
      "step": 690
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0004,
      "loss": 0.7232,
      "step": 691
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0004,
      "loss": 0.2523,
      "step": 692
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0004,
      "loss": 0.4367,
      "step": 693
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0004,
      "loss": 0.2121,
      "step": 694
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0004,
      "loss": 0.384,
      "step": 695
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0004,
      "loss": 0.4618,
      "step": 696
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0004,
      "loss": 0.3786,
      "step": 697
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0004,
      "loss": 0.3153,
      "step": 698
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0004,
      "loss": 1.2427,
      "step": 699
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0004,
      "loss": 2.3096,
      "step": 700
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0004,
      "loss": 0.128,
      "step": 701
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0004,
      "loss": 0.2697,
      "step": 702
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0004,
      "loss": 0.3647,
      "step": 703
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0004,
      "loss": 0.7612,
      "step": 704
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0004,
      "loss": 0.9207,
      "step": 705
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0004,
      "loss": 0.6071,
      "step": 706
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0004,
      "loss": 0.7884,
      "step": 707
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0004,
      "loss": 0.9419,
      "step": 708
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0004,
      "loss": 0.7632,
      "step": 709
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0004,
      "loss": 0.7119,
      "step": 710
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0004,
      "loss": 0.696,
      "step": 711
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0004,
      "loss": 0.6528,
      "step": 712
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0004,
      "loss": 0.9004,
      "step": 713
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0004,
      "loss": 0.6225,
      "step": 714
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0004,
      "loss": 0.766,
      "step": 715
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0004,
      "loss": 0.8218,
      "step": 716
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0004,
      "loss": 0.6443,
      "step": 717
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0004,
      "loss": 0.6947,
      "step": 718
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0004,
      "loss": 0.5489,
      "step": 719
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0004,
      "loss": 0.5443,
      "step": 720
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0004,
      "loss": 0.3092,
      "step": 721
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0004,
      "loss": 0.6232,
      "step": 722
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0004,
      "loss": 0.473,
      "step": 723
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0004,
      "loss": 0.4059,
      "step": 724
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0004,
      "loss": 0.7054,
      "step": 725
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0004,
      "loss": 0.4207,
      "step": 726
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0004,
      "loss": 0.7167,
      "step": 727
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0004,
      "loss": 0.9212,
      "step": 728
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0004,
      "loss": 0.428,
      "step": 729
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0004,
      "loss": 0.5357,
      "step": 730
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0004,
      "loss": 0.5664,
      "step": 731
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0004,
      "loss": 0.9507,
      "step": 732
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0004,
      "loss": 1.1244,
      "step": 733
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0004,
      "loss": 0.5503,
      "step": 734
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0004,
      "loss": 0.259,
      "step": 735
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0004,
      "loss": 1.298,
      "step": 736
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0004,
      "loss": 0.3772,
      "step": 737
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0004,
      "loss": 1.0177,
      "step": 738
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0004,
      "loss": 1.0031,
      "step": 739
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0004,
      "loss": 0.8615,
      "step": 740
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0004,
      "loss": 0.4476,
      "step": 741
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0004,
      "loss": 1.4852,
      "step": 742
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0004,
      "loss": 0.5538,
      "step": 743
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0004,
      "loss": 0.5481,
      "step": 744
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0004,
      "loss": 1.0126,
      "step": 745
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0004,
      "loss": 0.5887,
      "step": 746
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0004,
      "loss": 0.4874,
      "step": 747
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004,
      "loss": 0.685,
      "step": 748
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004,
      "loss": 0.4105,
      "step": 749
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004,
      "loss": 0.956,
      "step": 750
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004,
      "loss": 0.1461,
      "step": 751
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004,
      "loss": 0.1716,
      "step": 752
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004,
      "loss": 0.2236,
      "step": 753
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004,
      "loss": 0.5039,
      "step": 754
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004,
      "loss": 0.5399,
      "step": 755
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004,
      "loss": 0.8579,
      "step": 756
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004,
      "loss": 0.8438,
      "step": 757
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004,
      "loss": 0.6915,
      "step": 758
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004,
      "loss": 0.9139,
      "step": 759
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004,
      "loss": 0.7129,
      "step": 760
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004,
      "loss": 0.7473,
      "step": 761
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0004,
      "loss": 0.8594,
      "step": 762
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004,
      "loss": 0.5555,
      "step": 763
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004,
      "loss": 0.9835,
      "step": 764
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004,
      "loss": 0.5186,
      "step": 765
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004,
      "loss": 0.8268,
      "step": 766
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004,
      "loss": 0.7523,
      "step": 767
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004,
      "loss": 0.5537,
      "step": 768
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004,
      "loss": 0.5165,
      "step": 769
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004,
      "loss": 0.6995,
      "step": 770
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004,
      "loss": 0.785,
      "step": 771
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004,
      "loss": 0.8304,
      "step": 772
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004,
      "loss": 0.672,
      "step": 773
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004,
      "loss": 0.785,
      "step": 774
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004,
      "loss": 0.6968,
      "step": 775
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004,
      "loss": 0.6106,
      "step": 776
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0004,
      "loss": 0.6045,
      "step": 777
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0004,
      "loss": 0.4182,
      "step": 778
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0004,
      "loss": 1.0079,
      "step": 779
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0004,
      "loss": 0.2465,
      "step": 780
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0004,
      "loss": 0.2605,
      "step": 781
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0004,
      "loss": 1.2797,
      "step": 782
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0004,
      "loss": 0.1785,
      "step": 783
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0004,
      "loss": 0.5563,
      "step": 784
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0004,
      "loss": 0.395,
      "step": 785
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0004,
      "loss": 0.256,
      "step": 786
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0004,
      "loss": 1.0868,
      "step": 787
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0004,
      "loss": 0.3505,
      "step": 788
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0004,
      "loss": 0.4305,
      "step": 789
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0004,
      "loss": 0.7121,
      "step": 790
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0004,
      "loss": 0.6441,
      "step": 791
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0004,
      "loss": 1.2003,
      "step": 792
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0004,
      "loss": 0.3511,
      "step": 793
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0004,
      "loss": 0.606,
      "step": 794
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0004,
      "loss": 1.3632,
      "step": 795
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0004,
      "loss": 0.5291,
      "step": 796
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0004,
      "loss": 1.0989,
      "step": 797
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0004,
      "loss": 1.6083,
      "step": 798
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0004,
      "loss": 0.4116,
      "step": 799
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0004,
      "loss": 2.924,
      "step": 800
    },
    {
      "epoch": 0.56,
      "eval_loss": 0.6417471766471863,
      "eval_runtime": 262.0155,
      "eval_samples_per_second": 1.908,
      "eval_steps_per_second": 0.954,
      "step": 800
    },
    {
      "epoch": 0.56,
      "mmlu_eval_accuracy": 0.30538545442719256,
      "mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
      "mmlu_eval_accuracy_anatomy": 0.35714285714285715,
      "mmlu_eval_accuracy_astronomy": 0.25,
      "mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
      "mmlu_eval_accuracy_clinical_knowledge": 0.41379310344827586,
      "mmlu_eval_accuracy_college_biology": 0.375,
      "mmlu_eval_accuracy_college_chemistry": 0.125,
      "mmlu_eval_accuracy_college_computer_science": 0.09090909090909091,
      "mmlu_eval_accuracy_college_mathematics": 0.36363636363636365,
      "mmlu_eval_accuracy_college_medicine": 0.2727272727272727,
      "mmlu_eval_accuracy_college_physics": 0.2727272727272727,
      "mmlu_eval_accuracy_computer_security": 0.2727272727272727,
      "mmlu_eval_accuracy_conceptual_physics": 0.23076923076923078,
      "mmlu_eval_accuracy_econometrics": 0.25,
      "mmlu_eval_accuracy_electrical_engineering": 0.1875,
      "mmlu_eval_accuracy_elementary_mathematics": 0.2682926829268293,
      "mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
      "mmlu_eval_accuracy_global_facts": 0.2,
      "mmlu_eval_accuracy_high_school_biology": 0.25,
      "mmlu_eval_accuracy_high_school_chemistry": 0.18181818181818182,
      "mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
      "mmlu_eval_accuracy_high_school_european_history": 0.5,
      "mmlu_eval_accuracy_high_school_geography": 0.6363636363636364,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.19047619047619047,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.27906976744186046,
      "mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.35714285714285715,
      "mmlu_loss": 1.06649634963274,
      "step": 800
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0004,
      "loss": 0.3342,
      "step": 801
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0004,
      "loss": 0.403,
      "step": 802
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0004,
      "loss": 0.525,
      "step": 803
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0004,
      "loss": 0.4384,
      "step": 804
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004,
      "loss": 2.1165,
      "step": 805
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004,
      "loss": 0.6357,
      "step": 806
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004,
      "loss": 0.8001,
      "step": 807
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004,
      "loss": 0.8867,
      "step": 808
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004,
      "loss": 0.9335,
      "step": 809
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004,
      "loss": 0.5799,
      "step": 810
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004,
      "loss": 0.8978,
      "step": 811
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004,
      "loss": 0.8002,
      "step": 812
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004,
      "loss": 0.8444,
      "step": 813
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004,
      "loss": 0.8727,
      "step": 814
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004,
      "loss": 0.7608,
      "step": 815
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004,
      "loss": 0.6937,
      "step": 816
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004,
      "loss": 0.4899,
      "step": 817
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004,
      "loss": 0.655,
      "step": 818
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004,
      "loss": 0.8038,
      "step": 819
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0004,
      "loss": 0.5451,
      "step": 820
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0004,
      "loss": 1.2012,
      "step": 821
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0004,
      "loss": 0.3971,
      "step": 822
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0004,
      "loss": 0.3928,
      "step": 823
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0004,
      "loss": 0.2833,
      "step": 824
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0004,
      "loss": 0.4913,
      "step": 825
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0004,
      "loss": 0.3772,
      "step": 826
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0004,
      "loss": 0.714,
      "step": 827
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0004,
      "loss": 0.4023,
      "step": 828
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0004,
      "loss": 0.633,
      "step": 829
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0004,
      "loss": 0.3141,
      "step": 830
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0004,
      "loss": 0.4716,
      "step": 831
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0004,
      "loss": 0.3983,
      "step": 832
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0004,
      "loss": 0.2375,
      "step": 833
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0004,
      "loss": 0.375,
      "step": 834
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0004,
      "loss": 0.4347,
      "step": 835
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0004,
      "loss": 1.9594,
      "step": 836
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0004,
      "loss": 0.3025,
      "step": 837
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0004,
      "loss": 0.6105,
      "step": 838
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0004,
      "loss": 0.6737,
      "step": 839
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0004,
      "loss": 0.4223,
      "step": 840
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0004,
      "loss": 0.5873,
      "step": 841
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0004,
      "loss": 0.727,
      "step": 842
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0004,
      "loss": 0.5712,
      "step": 843
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0004,
      "loss": 1.0627,
      "step": 844
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0004,
      "loss": 0.3005,
      "step": 845
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0004,
      "loss": 0.4506,
      "step": 846
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0004,
      "loss": 0.4797,
      "step": 847
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0004,
      "loss": 0.4686,
      "step": 848
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0004,
      "loss": 0.2213,
      "step": 849
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0004,
      "loss": 0.8648,
      "step": 850
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0004,
      "loss": 0.2103,
      "step": 851
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0004,
      "loss": 0.4139,
      "step": 852
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0004,
      "loss": 0.1546,
      "step": 853
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0004,
      "loss": 0.1155,
      "step": 854
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0004,
      "loss": 0.625,
      "step": 855
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0004,
      "loss": 0.8956,
      "step": 856
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0004,
      "loss": 0.7037,
      "step": 857
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0004,
      "loss": 0.8418,
      "step": 858
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0004,
      "loss": 0.6672,
      "step": 859
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0004,
      "loss": 0.8433,
      "step": 860
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0004,
      "loss": 0.7895,
      "step": 861
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0004,
      "loss": 0.7541,
      "step": 862
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0004,
      "loss": 0.7261,
      "step": 863
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0004,
      "loss": 0.6541,
      "step": 864
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0004,
      "loss": 0.7801,
      "step": 865
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0004,
      "loss": 0.7632,
      "step": 866
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0004,
      "loss": 0.3367,
      "step": 867
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0004,
      "loss": 0.925,
      "step": 868
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0004,
      "loss": 0.5,
      "step": 869
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0004,
      "loss": 0.644,
      "step": 870
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0004,
      "loss": 0.5403,
      "step": 871
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0004,
      "loss": 0.7117,
      "step": 872
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0004,
      "loss": 0.5514,
      "step": 873
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0004,
      "loss": 0.5432,
      "step": 874
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0004,
      "loss": 0.6386,
      "step": 875
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0004,
      "loss": 0.8542,
      "step": 876
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0004,
      "loss": 0.5486,
      "step": 877
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0004,
      "loss": 0.4173,
      "step": 878
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0004,
      "loss": 0.5841,
      "step": 879
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0004,
      "loss": 0.48,
      "step": 880
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0004,
      "loss": 0.3887,
      "step": 881
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0004,
      "loss": 0.2361,
      "step": 882
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0004,
      "loss": 0.4635,
      "step": 883
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0004,
      "loss": 0.3724,
      "step": 884
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0004,
      "loss": 0.3429,
      "step": 885
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0004,
      "loss": 0.1864,
      "step": 886
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0004,
      "loss": 1.0825,
      "step": 887
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0004,
      "loss": 0.8833,
      "step": 888
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0004,
      "loss": 0.6354,
      "step": 889
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0004,
      "loss": 1.1362,
      "step": 890
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0004,
      "loss": 0.6212,
      "step": 891
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0004,
      "loss": 1.0072,
      "step": 892
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0004,
      "loss": 0.4585,
      "step": 893
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0004,
      "loss": 0.5292,
      "step": 894
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0004,
      "loss": 0.578,
      "step": 895
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0004,
      "loss": 0.5296,
      "step": 896
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0004,
      "loss": 0.1166,
      "step": 897
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0004,
      "loss": 1.0426,
      "step": 898
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0004,
      "loss": 0.3109,
      "step": 899
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0004,
      "loss": 1.5721,
      "step": 900
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0004,
      "loss": 0.3905,
      "step": 901
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0004,
      "loss": 0.1533,
      "step": 902
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0004,
      "loss": 0.214,
      "step": 903
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0004,
      "loss": 0.7675,
      "step": 904
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004,
      "loss": 0.7776,
      "step": 905
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004,
      "loss": 0.8972,
      "step": 906
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004,
      "loss": 0.9331,
      "step": 907
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004,
      "loss": 0.7464,
      "step": 908
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004,
      "loss": 0.7094,
      "step": 909
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004,
      "loss": 0.8942,
      "step": 910
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004,
      "loss": 0.8436,
      "step": 911
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004,
      "loss": 0.8331,
      "step": 912
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004,
      "loss": 0.5505,
      "step": 913
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004,
      "loss": 0.4096,
      "step": 914
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004,
      "loss": 0.4566,
      "step": 915
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004,
      "loss": 0.6529,
      "step": 916
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004,
      "loss": 0.4911,
      "step": 917
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004,
      "loss": 0.4905,
      "step": 918
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0004,
      "loss": 0.7721,
      "step": 919
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0004,
      "loss": 0.7032,
      "step": 920
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0004,
      "loss": 0.6081,
      "step": 921
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0004,
      "loss": 0.7841,
      "step": 922
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0004,
      "loss": 0.6164,
      "step": 923
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0004,
      "loss": 0.6198,
      "step": 924
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0004,
      "loss": 0.5534,
      "step": 925
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0004,
      "loss": 0.9022,
      "step": 926
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0004,
      "loss": 0.2717,
      "step": 927
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0004,
      "loss": 0.7298,
      "step": 928
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0004,
      "loss": 0.7003,
      "step": 929
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0004,
      "loss": 0.5087,
      "step": 930
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0004,
      "loss": 0.5762,
      "step": 931
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0004,
      "loss": 0.4049,
      "step": 932
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0004,
      "loss": 0.3152,
      "step": 933
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004,
      "loss": 0.7488,
      "step": 934
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004,
      "loss": 0.1635,
      "step": 935
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004,
      "loss": 0.4921,
      "step": 936
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004,
      "loss": 0.3392,
      "step": 937
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004,
      "loss": 0.5501,
      "step": 938
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004,
      "loss": 0.7825,
      "step": 939
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004,
      "loss": 0.5922,
      "step": 940
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004,
      "loss": 1.0295,
      "step": 941
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004,
      "loss": 0.4026,
      "step": 942
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004,
      "loss": 0.5983,
      "step": 943
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004,
      "loss": 1.1622,
      "step": 944
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004,
      "loss": 0.58,
      "step": 945
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004,
      "loss": 0.3097,
      "step": 946
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0004,
      "loss": 1.2314,
      "step": 947
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004,
      "loss": 0.404,
      "step": 948
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004,
      "loss": 0.3989,
      "step": 949
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004,
      "loss": 0.2975,
      "step": 950
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004,
      "loss": 0.3987,
      "step": 951
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004,
      "loss": 0.1746,
      "step": 952
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004,
      "loss": 1.0329,
      "step": 953
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004,
      "loss": 0.5963,
      "step": 954
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004,
      "loss": 0.8642,
      "step": 955
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004,
      "loss": 0.7199,
      "step": 956
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004,
      "loss": 0.7995,
      "step": 957
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004,
      "loss": 0.9483,
      "step": 958
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004,
      "loss": 0.8768,
      "step": 959
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004,
      "loss": 0.7032,
      "step": 960
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004,
      "loss": 0.773,
      "step": 961
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004,
      "loss": 0.7956,
      "step": 962
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004,
      "loss": 0.7093,
      "step": 963
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004,
      "loss": 0.7432,
      "step": 964
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004,
      "loss": 0.8717,
      "step": 965
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004,
      "loss": 0.472,
      "step": 966
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004,
      "loss": 0.7872,
      "step": 967
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004,
      "loss": 0.8238,
      "step": 968
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004,
      "loss": 0.4689,
      "step": 969
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004,
      "loss": 0.5136,
      "step": 970
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004,
      "loss": 0.4863,
      "step": 971
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004,
      "loss": 0.6755,
      "step": 972
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004,
      "loss": 0.463,
      "step": 973
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004,
      "loss": 0.5894,
      "step": 974
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0004,
      "loss": 0.4749,
      "step": 975
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004,
      "loss": 0.644,
      "step": 976
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004,
      "loss": 0.8176,
      "step": 977
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004,
      "loss": 0.6255,
      "step": 978
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004,
      "loss": 0.5182,
      "step": 979
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004,
      "loss": 0.4552,
      "step": 980
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004,
      "loss": 0.4505,
      "step": 981
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004,
      "loss": 0.8373,
      "step": 982
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004,
      "loss": 0.5285,
      "step": 983
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004,
      "loss": 1.0748,
      "step": 984
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004,
      "loss": 0.5467,
      "step": 985
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004,
      "loss": 0.7294,
      "step": 986
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004,
      "loss": 0.6341,
      "step": 987
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004,
      "loss": 0.2867,
      "step": 988
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004,
      "loss": 1.2924,
      "step": 989
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004,
      "loss": 0.4195,
      "step": 990
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004,
      "loss": 0.5777,
      "step": 991
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004,
      "loss": 1.6128,
      "step": 992
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004,
      "loss": 0.187,
      "step": 993
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004,
      "loss": 0.3625,
      "step": 994
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004,
      "loss": 0.2464,
      "step": 995
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004,
      "loss": 0.6021,
      "step": 996
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004,
      "loss": 0.5822,
      "step": 997
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004,
      "loss": 1.3364,
      "step": 998
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004,
      "loss": 0.2851,
      "step": 999
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004,
      "loss": 0.5591,
      "step": 1000
    },
    {
      "epoch": 0.7,
      "eval_loss": 0.6062986850738525,
      "eval_runtime": 261.9741,
      "eval_samples_per_second": 1.909,
      "eval_steps_per_second": 0.954,
      "step": 1000
    },
    {
      "epoch": 0.7,
      "mmlu_eval_accuracy": 0.2696631261532825,
      "mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
      "mmlu_eval_accuracy_anatomy": 0.42857142857142855,
      "mmlu_eval_accuracy_astronomy": 0.25,
      "mmlu_eval_accuracy_business_ethics": 0.36363636363636365,
      "mmlu_eval_accuracy_clinical_knowledge": 0.3793103448275862,
      "mmlu_eval_accuracy_college_biology": 0.25,
      "mmlu_eval_accuracy_college_chemistry": 0.125,
      "mmlu_eval_accuracy_college_computer_science": 0.09090909090909091,
      "mmlu_eval_accuracy_college_mathematics": 0.2727272727272727,
      "mmlu_eval_accuracy_college_medicine": 0.22727272727272727,
      "mmlu_eval_accuracy_college_physics": 0.18181818181818182,
      "mmlu_eval_accuracy_computer_security": 0.2727272727272727,
      "mmlu_eval_accuracy_conceptual_physics": 0.23076923076923078,
      "mmlu_eval_accuracy_econometrics": 0.25,
      "mmlu_eval_accuracy_electrical_engineering": 0.1875,
      "mmlu_eval_accuracy_elementary_mathematics": 0.24390243902439024,
      "mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
      "mmlu_eval_accuracy_global_facts": 0.2,
      "mmlu_eval_accuracy_high_school_biology": 0.1875,
      "mmlu_eval_accuracy_high_school_chemistry": 0.13636363636363635,
      "mmlu_eval_accuracy_high_school_computer_science": 0.4444444444444444,
      "mmlu_eval_accuracy_high_school_european_history": 0.3333333333333333,
      "mmlu_eval_accuracy_high_school_geography": 0.5909090909090909,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.09523809523809523,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.32558139534883723,
      "mmlu_eval_accuracy_high_school_mathematics": 0.20689655172413793,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.35714285714285715,
      "mmlu_loss": 0.8371138982176781,
      "step": 1000
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004,
      "loss": 0.2568,
      "step": 1001
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004,
      "loss": 0.2256,
      "step": 1002
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004,
      "loss": 0.1923,
      "step": 1003
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0004,
      "loss": 0.8292,
      "step": 1004
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004,
      "loss": 0.7634,
      "step": 1005
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004,
      "loss": 0.8455,
      "step": 1006
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004,
      "loss": 0.9398,
      "step": 1007
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004,
      "loss": 0.8622,
      "step": 1008
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004,
      "loss": 0.6987,
      "step": 1009
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004,
      "loss": 0.5708,
      "step": 1010
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004,
      "loss": 0.889,
      "step": 1011
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004,
      "loss": 0.754,
      "step": 1012
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004,
      "loss": 0.7802,
      "step": 1013
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004,
      "loss": 0.8057,
      "step": 1014
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004,
      "loss": 0.7057,
      "step": 1015
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004,
      "loss": 0.5939,
      "step": 1016
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004,
      "loss": 0.9458,
      "step": 1017
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0004,
      "loss": 0.4199,
      "step": 1018
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004,
      "loss": 0.6285,
      "step": 1019
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004,
      "loss": 0.5437,
      "step": 1020
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004,
      "loss": 0.7155,
      "step": 1021
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004,
      "loss": 0.5762,
      "step": 1022
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004,
      "loss": 0.6708,
      "step": 1023
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004,
      "loss": 0.6386,
      "step": 1024
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004,
      "loss": 0.6075,
      "step": 1025
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004,
      "loss": 0.6806,
      "step": 1026
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004,
      "loss": 0.5212,
      "step": 1027
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004,
      "loss": 0.4784,
      "step": 1028
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004,
      "loss": 0.2564,
      "step": 1029
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004,
      "loss": 0.6152,
      "step": 1030
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004,
      "loss": 0.3457,
      "step": 1031
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0004,
      "loss": 0.5407,
      "step": 1032
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004,
      "loss": 0.4891,
      "step": 1033
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004,
      "loss": 0.5085,
      "step": 1034
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004,
      "loss": 0.2077,
      "step": 1035
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004,
      "loss": 0.4229,
      "step": 1036
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004,
      "loss": 0.6036,
      "step": 1037
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004,
      "loss": 0.5033,
      "step": 1038
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004,
      "loss": 0.3658,
      "step": 1039
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004,
      "loss": 0.7938,
      "step": 1040
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004,
      "loss": 0.4784,
      "step": 1041
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004,
      "loss": 0.475,
      "step": 1042
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004,
      "loss": 0.3961,
      "step": 1043
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004,
      "loss": 0.5404,
      "step": 1044
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004,
      "loss": 0.3528,
      "step": 1045
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004,
      "loss": 0.3864,
      "step": 1046
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0004,
      "loss": 0.5713,
      "step": 1047
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004,
      "loss": 0.9218,
      "step": 1048
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004,
      "loss": 0.9482,
      "step": 1049
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004,
      "loss": 0.8584,
      "step": 1050
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004,
      "loss": 0.2274,
      "step": 1051
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004,
      "loss": 0.9419,
      "step": 1052
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004,
      "loss": 0.6805,
      "step": 1053
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004,
      "loss": 0.8443,
      "step": 1054
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004,
      "loss": 0.8746,
      "step": 1055
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004,
      "loss": 0.8156,
      "step": 1056
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004,
      "loss": 0.7459,
      "step": 1057
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004,
      "loss": 0.8752,
      "step": 1058
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004,
      "loss": 0.7534,
      "step": 1059
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004,
      "loss": 0.8391,
      "step": 1060
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004,
      "loss": 0.8513,
      "step": 1061
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004,
      "loss": 0.8419,
      "step": 1062
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004,
      "loss": 0.491,
      "step": 1063
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004,
      "loss": 0.7671,
      "step": 1064
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004,
      "loss": 0.912,
      "step": 1065
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004,
      "loss": 0.3856,
      "step": 1066
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004,
      "loss": 0.5201,
      "step": 1067
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004,
      "loss": 0.54,
      "step": 1068
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004,
      "loss": 0.4883,
      "step": 1069
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004,
      "loss": 0.6604,
      "step": 1070
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004,
      "loss": 0.8023,
      "step": 1071
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004,
      "loss": 0.8695,
      "step": 1072
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004,
      "loss": 0.7843,
      "step": 1073
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004,
      "loss": 0.4502,
      "step": 1074
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0004,
      "loss": 0.6346,
      "step": 1075
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004,
      "loss": 0.9306,
      "step": 1076
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004,
      "loss": 0.5691,
      "step": 1077
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004,
      "loss": 0.6568,
      "step": 1078
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004,
      "loss": 0.7103,
      "step": 1079
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004,
      "loss": 0.9179,
      "step": 1080
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004,
      "loss": 0.3694,
      "step": 1081
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004,
      "loss": 0.4531,
      "step": 1082
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004,
      "loss": 1.2981,
      "step": 1083
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004,
      "loss": 1.3878,
      "step": 1084
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004,
      "loss": 0.5209,
      "step": 1085
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004,
      "loss": 1.1522,
      "step": 1086
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004,
      "loss": 0.2509,
      "step": 1087
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004,
      "loss": 1.4895,
      "step": 1088
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0004,
      "loss": 0.8326,
      "step": 1089
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004,
      "loss": 0.53,
      "step": 1090
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004,
      "loss": 0.4127,
      "step": 1091
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004,
      "loss": 1.2986,
      "step": 1092
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004,
      "loss": 0.4335,
      "step": 1093
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004,
      "loss": 0.7598,
      "step": 1094
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004,
      "loss": 0.9294,
      "step": 1095
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004,
      "loss": 0.7546,
      "step": 1096
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004,
      "loss": 0.5667,
      "step": 1097
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004,
      "loss": 0.444,
      "step": 1098
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004,
      "loss": 0.4992,
      "step": 1099
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004,
      "loss": 1.4161,
      "step": 1100
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004,
      "loss": 0.1406,
      "step": 1101
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004,
      "loss": 0.1881,
      "step": 1102
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0004,
      "loss": 0.1375,
      "step": 1103
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004,
      "loss": 0.3047,
      "step": 1104
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004,
      "loss": 0.1846,
      "step": 1105
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004,
      "loss": 0.3801,
      "step": 1106
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004,
      "loss": 0.9691,
      "step": 1107
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004,
      "loss": 0.7683,
      "step": 1108
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004,
      "loss": 0.8054,
      "step": 1109
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004,
      "loss": 0.8484,
      "step": 1110
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004,
      "loss": 0.84,
      "step": 1111
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004,
      "loss": 0.6551,
      "step": 1112
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004,
      "loss": 0.5611,
      "step": 1113
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004,
      "loss": 0.7925,
      "step": 1114
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004,
      "loss": 0.5565,
      "step": 1115
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004,
      "loss": 0.5667,
      "step": 1116
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004,
      "loss": 0.5563,
      "step": 1117
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0004,
      "loss": 0.5991,
      "step": 1118
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0004,
      "loss": 0.8101,
      "step": 1119
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0004,
      "loss": 0.5182,
      "step": 1120
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0004,
      "loss": 0.5591,
      "step": 1121
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0004,
      "loss": 0.8661,
      "step": 1122
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0004,
      "loss": 0.518,
      "step": 1123
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0004,
      "loss": 0.3308,
      "step": 1124
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0004,
      "loss": 0.2169,
      "step": 1125
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0004,
      "loss": 0.7888,
      "step": 1126
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0004,
      "loss": 0.9938,
      "step": 1127
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0004,
      "loss": 0.6998,
      "step": 1128
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0004,
      "loss": 0.3947,
      "step": 1129
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0004,
      "loss": 0.2432,
      "step": 1130
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0004,
      "loss": 1.0518,
      "step": 1131
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0004,
      "loss": 0.4578,
      "step": 1132
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0004,
      "loss": 0.6499,
      "step": 1133
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0004,
      "loss": 0.5507,
      "step": 1134
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0004,
      "loss": 0.5031,
      "step": 1135
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0004,
      "loss": 0.8217,
      "step": 1136
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0004,
      "loss": 0.9918,
      "step": 1137
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0004,
      "loss": 0.6194,
      "step": 1138
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0004,
      "loss": 0.3885,
      "step": 1139
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0004,
      "loss": 0.514,
      "step": 1140
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0004,
      "loss": 0.6231,
      "step": 1141
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0004,
      "loss": 0.3778,
      "step": 1142
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0004,
      "loss": 0.5501,
      "step": 1143
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0004,
      "loss": 1.0975,
      "step": 1144
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0004,
      "loss": 0.3143,
      "step": 1145
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0004,
      "loss": 0.5101,
      "step": 1146
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0004,
      "loss": 0.3943,
      "step": 1147
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0004,
      "loss": 1.0746,
      "step": 1148
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0004,
      "loss": 0.2978,
      "step": 1149
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0004,
      "loss": 0.342,
      "step": 1150
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0004,
      "loss": 0.1872,
      "step": 1151
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0004,
      "loss": 0.2124,
      "step": 1152
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0004,
      "loss": 0.2549,
      "step": 1153
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0004,
      "loss": 0.1347,
      "step": 1154
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0004,
      "loss": 0.8295,
      "step": 1155
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0004,
      "loss": 0.7941,
      "step": 1156
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0004,
      "loss": 0.3214,
      "step": 1157
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0004,
      "loss": 0.8334,
      "step": 1158
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0004,
      "loss": 0.7792,
      "step": 1159
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0004,
      "loss": 0.6914,
      "step": 1160
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004,
      "loss": 0.9085,
      "step": 1161
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004,
      "loss": 0.8467,
      "step": 1162
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004,
      "loss": 0.7527,
      "step": 1163
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004,
      "loss": 0.7065,
      "step": 1164
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004,
      "loss": 0.5496,
      "step": 1165
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004,
      "loss": 0.747,
      "step": 1166
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004,
      "loss": 0.9068,
      "step": 1167
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004,
      "loss": 0.5662,
      "step": 1168
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004,
      "loss": 0.8689,
      "step": 1169
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004,
      "loss": 0.8347,
      "step": 1170
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004,
      "loss": 0.3442,
      "step": 1171
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004,
      "loss": 0.375,
      "step": 1172
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004,
      "loss": 0.6623,
      "step": 1173
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004,
      "loss": 0.8362,
      "step": 1174
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004,
      "loss": 0.7407,
      "step": 1175
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0004,
      "loss": 0.4577,
      "step": 1176
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0004,
      "loss": 0.4128,
      "step": 1177
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0004,
      "loss": 0.5529,
      "step": 1178
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0004,
      "loss": 0.385,
      "step": 1179
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0004,
      "loss": 0.7016,
      "step": 1180
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0004,
      "loss": 0.3416,
      "step": 1181
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0004,
      "loss": 0.3132,
      "step": 1182
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0004,
      "loss": 0.4332,
      "step": 1183
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0004,
      "loss": 0.4828,
      "step": 1184
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0004,
      "loss": 0.2299,
      "step": 1185
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0004,
      "loss": 0.8477,
      "step": 1186
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0004,
      "loss": 0.3425,
      "step": 1187
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0004,
      "loss": 1.0418,
      "step": 1188
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0004,
      "loss": 0.7957,
      "step": 1189
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0004,
      "loss": 0.2999,
      "step": 1190
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0004,
      "loss": 0.4034,
      "step": 1191
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0004,
      "loss": 0.6535,
      "step": 1192
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0004,
      "loss": 0.9691,
      "step": 1193
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0004,
      "loss": 0.559,
      "step": 1194
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0004,
      "loss": 1.8883,
      "step": 1195
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0004,
      "loss": 0.1952,
      "step": 1196
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0004,
      "loss": 0.4336,
      "step": 1197
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0004,
      "loss": 0.6651,
      "step": 1198
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0004,
      "loss": 0.7694,
      "step": 1199
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0004,
      "loss": 0.3228,
      "step": 1200
    },
    {
      "epoch": 0.84,
      "eval_loss": 0.6253868341445923,
      "eval_runtime": 261.8289,
      "eval_samples_per_second": 1.91,
      "eval_steps_per_second": 0.955,
      "step": 1200
    },
    {
      "epoch": 0.84,
      "mmlu_eval_accuracy": 0.3628458989343948,
      "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
      "mmlu_eval_accuracy_anatomy": 0.42857142857142855,
      "mmlu_eval_accuracy_astronomy": 0.3125,
      "mmlu_eval_accuracy_business_ethics": 0.36363636363636365,
      "mmlu_eval_accuracy_clinical_knowledge": 0.3793103448275862,
      "mmlu_eval_accuracy_college_biology": 0.3125,
      "mmlu_eval_accuracy_college_chemistry": 0.375,
      "mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
      "mmlu_eval_accuracy_college_mathematics": 0.18181818181818182,
      "mmlu_eval_accuracy_college_medicine": 0.4090909090909091,
      "mmlu_eval_accuracy_college_physics": 0.6363636363636364,
      "mmlu_eval_accuracy_computer_security": 0.18181818181818182,
      "mmlu_eval_accuracy_conceptual_physics": 0.4230769230769231,
      "mmlu_eval_accuracy_econometrics": 0.25,
      "mmlu_eval_accuracy_electrical_engineering": 0.5,
      "mmlu_eval_accuracy_elementary_mathematics": 0.3902439024390244,
      "mmlu_eval_accuracy_formal_logic": 0.35714285714285715,
      "mmlu_eval_accuracy_global_facts": 0.4,
      "mmlu_eval_accuracy_high_school_biology": 0.40625,
      "mmlu_eval_accuracy_high_school_chemistry": 0.22727272727272727,
      "mmlu_eval_accuracy_high_school_computer_science": 0.3333333333333333,
      "mmlu_eval_accuracy_high_school_european_history": 0.2222222222222222,
      "mmlu_eval_accuracy_high_school_geography": 0.6363636363636364,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.42857142857142855,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.4883720930232558,
      "mmlu_eval_accuracy_high_school_mathematics": 0.06896551724137931,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.35714285714285715,
      "mmlu_loss": 0.8975866470783949,
      "step": 1200
    }
  ],
  "max_steps": 30000,
  "num_train_epochs": 22,
  "total_flos": 2.3771033427758285e+17,
  "trial_name": null,
  "trial_params": null
}