{
  "best_metric": 0.8272572159767151,
  "best_model_checkpoint": "experts/mistralic-expert-22/checkpoint-200",
  "epoch": 0.07699340743948799,
  "eval_steps": 200,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9917,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8752,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8807,
      "step": 3
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7598,
      "step": 4
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6723,
      "step": 5
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.1704,
      "step": 6
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.7503,
      "step": 7
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.624,
      "step": 8
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8103,
      "step": 9
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9605,
      "step": 10
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.905,
      "step": 11
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.759,
      "step": 12
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8095,
      "step": 13
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.1747,
      "step": 14
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8053,
      "step": 15
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6902,
      "step": 16
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.878,
      "step": 17
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.89,
      "step": 18
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.899,
      "step": 19
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.0926,
      "step": 20
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9986,
      "step": 21
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.8805,
      "step": 22
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.6184,
      "step": 23
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 0.9538,
      "step": 24
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.0293,
      "step": 25
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.4661,
      "step": 26
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8878,
      "step": 27
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8821,
      "step": 28
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.85,
      "step": 29
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9784,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7107,
      "step": 31
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7971,
      "step": 32
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8528,
      "step": 33
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8603,
      "step": 34
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7917,
      "step": 35
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6604,
      "step": 36
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8287,
      "step": 37
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.1047,
      "step": 38
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6991,
      "step": 39
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9996,
      "step": 40
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8583,
      "step": 41
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7114,
      "step": 42
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7781,
      "step": 43
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9923,
      "step": 44
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8158,
      "step": 45
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7188,
      "step": 46
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8207,
      "step": 47
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7159,
      "step": 48
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8143,
      "step": 49
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9955,
      "step": 50
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6577,
      "step": 51
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7208,
      "step": 52
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9416,
      "step": 53
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7941,
      "step": 54
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7234,
      "step": 55
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0252,
      "step": 56
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7091,
      "step": 57
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7707,
      "step": 58
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.157,
      "step": 59
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.5682,
      "step": 60
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.807,
      "step": 61
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6721,
      "step": 62
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8324,
      "step": 63
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8405,
      "step": 64
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7994,
      "step": 65
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7385,
      "step": 66
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.4077,
      "step": 67
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8142,
      "step": 68
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6694,
      "step": 69
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.8789,
      "step": 70
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.058,
      "step": 71
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.1531,
      "step": 72
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9339,
      "step": 73
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7771,
      "step": 74
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 1.0121,
      "step": 75
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.7184,
      "step": 76
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.6577,
      "step": 77
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9318,
      "step": 78
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.105,
      "step": 79
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6653,
      "step": 80
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.631,
      "step": 81
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9929,
      "step": 82
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.03,
      "step": 83
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7572,
      "step": 84
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8269,
      "step": 85
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6807,
      "step": 86
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9773,
      "step": 87
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.866,
      "step": 88
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9673,
      "step": 89
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8735,
      "step": 90
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8992,
      "step": 91
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8388,
      "step": 92
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8848,
      "step": 93
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6855,
      "step": 94
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7848,
      "step": 95
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.1573,
      "step": 96
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8753,
      "step": 97
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.016,
      "step": 98
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8481,
      "step": 99
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8461,
      "step": 100
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8009,
      "step": 101
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8147,
      "step": 102
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.818,
      "step": 103
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7188,
      "step": 104
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8583,
      "step": 105
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8898,
      "step": 106
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.5784,
      "step": 107
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6973,
      "step": 108
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8995,
      "step": 109
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8895,
      "step": 110
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7302,
      "step": 111
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.975,
      "step": 112
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7004,
      "step": 113
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.6645,
      "step": 114
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7651,
      "step": 115
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8179,
      "step": 116
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8355,
      "step": 117
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8815,
      "step": 118
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8535,
      "step": 119
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7703,
      "step": 120
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8666,
      "step": 121
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9532,
      "step": 122
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7817,
      "step": 123
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.0935,
      "step": 124
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7434,
      "step": 125
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.8263,
      "step": 126
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9015,
      "step": 127
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 1.5507,
      "step": 128
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.923,
      "step": 129
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0015,
      "step": 130
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8959,
      "step": 131
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.996,
      "step": 132
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9214,
      "step": 133
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6441,
      "step": 134
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8035,
      "step": 135
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8738,
      "step": 136
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6934,
      "step": 137
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7723,
      "step": 138
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8101,
      "step": 139
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6368,
      "step": 140
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7078,
      "step": 141
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8225,
      "step": 142
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7866,
      "step": 143
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8628,
      "step": 144
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6387,
      "step": 145
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7479,
      "step": 146
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8173,
      "step": 147
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8456,
      "step": 148
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7523,
      "step": 149
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6802,
      "step": 150
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7845,
      "step": 151
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.622,
      "step": 152
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.842,
      "step": 153
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0277,
      "step": 154
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0264,
      "step": 155
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9195,
      "step": 156
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.797,
      "step": 157
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.87,
      "step": 158
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.6444,
      "step": 159
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8486,
      "step": 160
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9643,
      "step": 161
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6787,
      "step": 162
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7781,
      "step": 163
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6973,
      "step": 164
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7924,
      "step": 165
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9079,
      "step": 166
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7007,
      "step": 167
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8613,
      "step": 168
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.608,
      "step": 169
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9465,
      "step": 170
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9435,
      "step": 171
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7896,
      "step": 172
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6892,
      "step": 173
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.8605,
      "step": 174
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7067,
      "step": 175
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.6283,
      "step": 176
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9131,
      "step": 177
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0426,
      "step": 178
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7652,
      "step": 179
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.0208,
      "step": 180
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7484,
      "step": 181
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8419,
      "step": 182
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9176,
      "step": 183
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.861,
      "step": 184
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9,
      "step": 185
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.907,
      "step": 186
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.651,
      "step": 187
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6038,
      "step": 188
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7511,
      "step": 189
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6037,
      "step": 190
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.592,
      "step": 191
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5935,
      "step": 192
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9313,
      "step": 193
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8181,
      "step": 194
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.792,
      "step": 195
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7445,
      "step": 196
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6853,
      "step": 197
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9783,
      "step": 198
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9599,
      "step": 199
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9033,
      "step": 200
    },
    {
      "epoch": 0.04,
      "eval_loss": 0.8272572159767151,
      "eval_runtime": 108.0162,
      "eval_samples_per_second": 9.258,
      "eval_steps_per_second": 4.629,
      "step": 200
    },
    {
      "epoch": 0.04,
      "mmlu_eval_accuracy": 0.599579385714013,
      "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
      "mmlu_eval_accuracy_anatomy": 0.42857142857142855,
      "mmlu_eval_accuracy_astronomy": 0.8125,
      "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
      "mmlu_eval_accuracy_clinical_knowledge": 0.6206896551724138,
      "mmlu_eval_accuracy_college_biology": 0.5625,
      "mmlu_eval_accuracy_college_chemistry": 0.25,
      "mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
      "mmlu_eval_accuracy_college_mathematics": 0.6363636363636364,
      "mmlu_eval_accuracy_college_medicine": 0.5909090909090909,
      "mmlu_eval_accuracy_college_physics": 0.5454545454545454,
      "mmlu_eval_accuracy_computer_security": 0.7272727272727273,
      "mmlu_eval_accuracy_conceptual_physics": 0.5384615384615384,
      "mmlu_eval_accuracy_econometrics": 0.5,
      "mmlu_eval_accuracy_electrical_engineering": 0.5,
      "mmlu_eval_accuracy_elementary_mathematics": 0.4146341463414634,
      "mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
      "mmlu_eval_accuracy_global_facts": 0.4,
      "mmlu_eval_accuracy_high_school_biology": 0.625,
      "mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
      "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
      "mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.6511627906976745,
      "mmlu_eval_accuracy_high_school_mathematics": 0.3103448275862069,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.6538461538461539,
      "mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
      "mmlu_eval_accuracy_high_school_psychology": 0.85,
      "mmlu_eval_accuracy_high_school_statistics": 0.43478260869565216,
      "mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
      "mmlu_eval_accuracy_high_school_world_history": 0.7307692307692307,
      "mmlu_eval_accuracy_human_aging": 0.7391304347826086,
      "mmlu_eval_accuracy_human_sexuality": 0.5,
      "mmlu_eval_accuracy_international_law": 0.9230769230769231,
      "mmlu_eval_accuracy_jurisprudence": 0.6363636363636364,
      "mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
      "mmlu_eval_accuracy_machine_learning": 0.36363636363636365,
      "mmlu_eval_accuracy_management": 0.9090909090909091,
      "mmlu_eval_accuracy_marketing": 0.88,
      "mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
      "mmlu_eval_accuracy_miscellaneous": 0.7558139534883721,
      "mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
      "mmlu_eval_accuracy_moral_scenarios": 0.33,
      "mmlu_eval_accuracy_nutrition": 0.6666666666666666,
      "mmlu_eval_accuracy_philosophy": 0.7352941176470589,
      "mmlu_eval_accuracy_prehistory": 0.6,
      "mmlu_eval_accuracy_professional_accounting": 0.5806451612903226,
      "mmlu_eval_accuracy_professional_law": 0.38823529411764707,
      "mmlu_eval_accuracy_professional_medicine": 0.6451612903225806,
      "mmlu_eval_accuracy_professional_psychology": 0.6231884057971014,
      "mmlu_eval_accuracy_public_relations": 0.5,
      "mmlu_eval_accuracy_security_studies": 0.6296296296296297,
      "mmlu_eval_accuracy_sociology": 0.9090909090909091,
      "mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
      "mmlu_eval_accuracy_virology": 0.5,
      "mmlu_eval_accuracy_world_religions": 0.7894736842105263,
      "mmlu_loss": 1.1056992180898357,
      "step": 200
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8147,
      "step": 201
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.2314,
      "step": 202
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9407,
      "step": 203
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7344,
      "step": 204
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7506,
      "step": 205
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9563,
      "step": 206
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9926,
      "step": 207
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8277,
      "step": 208
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7435,
      "step": 209
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9858,
      "step": 210
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8615,
      "step": 211
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7798,
      "step": 212
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8055,
      "step": 213
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8161,
      "step": 214
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8922,
      "step": 215
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7699,
      "step": 216
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9098,
      "step": 217
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8848,
      "step": 218
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7967,
      "step": 219
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7214,
      "step": 220
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.6714,
      "step": 221
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8672,
      "step": 222
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7304,
      "step": 223
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7675,
      "step": 224
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.5663,
      "step": 225
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9466,
      "step": 226
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 1.0255,
      "step": 227
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.9056,
      "step": 228
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8921,
      "step": 229
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.7836,
      "step": 230
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8398,
      "step": 231
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8849,
      "step": 232
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8133,
      "step": 233
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9293,
      "step": 234
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6496,
      "step": 235
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5932,
      "step": 236
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9838,
      "step": 237
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8546,
      "step": 238
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9101,
      "step": 239
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.859,
      "step": 240
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9602,
      "step": 241
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9327,
      "step": 242
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5277,
      "step": 243
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.748,
      "step": 244
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7524,
      "step": 245
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9312,
      "step": 246
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.2737,
      "step": 247
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0512,
      "step": 248
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7334,
      "step": 249
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7493,
      "step": 250
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5577,
      "step": 251
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 1.0016,
      "step": 252
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9124,
      "step": 253
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8657,
      "step": 254
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9144,
      "step": 255
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9532,
      "step": 256
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7411,
      "step": 257
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7931,
      "step": 258
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6414,
      "step": 259
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7083,
      "step": 260
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9409,
      "step": 261
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.856,
      "step": 262
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9955,
      "step": 263
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5929,
      "step": 264
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7547,
      "step": 265
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6542,
      "step": 266
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8503,
      "step": 267
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8429,
      "step": 268
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.857,
      "step": 269
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8845,
      "step": 270
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8411,
      "step": 271
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9812,
      "step": 272
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8003,
      "step": 273
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7384,
      "step": 274
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7263,
      "step": 275
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.78,
      "step": 276
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.9858,
      "step": 277
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7142,
      "step": 278
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7677,
      "step": 279
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.6429,
      "step": 280
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8833,
      "step": 281
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8403,
      "step": 282
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5925,
      "step": 283
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7705,
      "step": 284
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7843,
      "step": 285
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8162,
      "step": 286
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7978,
      "step": 287
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7364,
      "step": 288
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8184,
      "step": 289
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8204,
      "step": 290
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9493,
      "step": 291
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8844,
      "step": 292
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7213,
      "step": 293
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7367,
      "step": 294
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8065,
      "step": 295
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7912,
      "step": 296
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9889,
      "step": 297
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6346,
      "step": 298
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8347,
      "step": 299
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.821,
      "step": 300
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0193,
      "step": 301
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7745,
      "step": 302
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.5443,
      "step": 303
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0296,
      "step": 304
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8065,
      "step": 305
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7023,
      "step": 306
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7963,
      "step": 307
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6676,
      "step": 308
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7119,
      "step": 309
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8342,
      "step": 310
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8657,
      "step": 311
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9189,
      "step": 312
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9195,
      "step": 313
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.737,
      "step": 314
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6512,
      "step": 315
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9037,
      "step": 316
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.942,
      "step": 317
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9569,
      "step": 318
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7858,
      "step": 319
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0143,
      "step": 320
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0598,
      "step": 321
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8547,
      "step": 322
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6792,
      "step": 323
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9757,
      "step": 324
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7605,
      "step": 325
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7029,
      "step": 326
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9174,
      "step": 327
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.908,
      "step": 328
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0953,
      "step": 329
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6913,
      "step": 330
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7406,
      "step": 331
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6988,
      "step": 332
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.4839,
      "step": 333
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.9061,
      "step": 334
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7339,
      "step": 335
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.7077,
      "step": 336
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 1.0788,
      "step": 337
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7634,
      "step": 338
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.652,
      "step": 339
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.802,
      "step": 340
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7895,
      "step": 341
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8487,
      "step": 342
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7397,
      "step": 343
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.857,
      "step": 344
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9782,
      "step": 345
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.688,
      "step": 346
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5103,
      "step": 347
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.762,
      "step": 348
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9286,
      "step": 349
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7606,
      "step": 350
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8981,
      "step": 351
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7572,
      "step": 352
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7552,
      "step": 353
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6557,
      "step": 354
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7197,
      "step": 355
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6654,
      "step": 356
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8249,
      "step": 357
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9703,
      "step": 358
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9126,
      "step": 359
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6395,
      "step": 360
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8437,
      "step": 361
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8885,
      "step": 362
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9231,
      "step": 363
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5622,
      "step": 364
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9886,
      "step": 365
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6807,
      "step": 366
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7713,
      "step": 367
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8205,
      "step": 368
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6163,
      "step": 369
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6523,
      "step": 370
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7218,
      "step": 371
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7817,
      "step": 372
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.1076,
      "step": 373
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9519,
      "step": 374
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9615,
      "step": 375
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8734,
      "step": 376
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7385,
      "step": 377
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6975,
      "step": 378
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6935,
      "step": 379
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6487,
      "step": 380
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8057,
      "step": 381
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7861,
      "step": 382
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6614,
      "step": 383
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6929,
      "step": 384
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.9957,
      "step": 385
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.7739,
      "step": 386
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.953,
      "step": 387
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.5405,
      "step": 388
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8276,
      "step": 389
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6577,
      "step": 390
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.6967,
      "step": 391
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7781,
      "step": 392
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8823,
      "step": 393
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.5715,
      "step": 394
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9113,
      "step": 395
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8301,
      "step": 396
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.9687,
      "step": 397
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.76,
      "step": 398
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7402,
      "step": 399
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.7616,
      "step": 400
    },
    {
      "epoch": 0.08,
      "eval_loss": 0.8319438099861145,
      "eval_runtime": 107.9019,
      "eval_samples_per_second": 9.268,
      "eval_steps_per_second": 4.634,
      "step": 400
    },
    {
      "epoch": 0.08,
      "mmlu_eval_accuracy": 0.5644849203853226,
      "mmlu_eval_accuracy_abstract_algebra": 0.45454545454545453,
      "mmlu_eval_accuracy_anatomy": 0.42857142857142855,
      "mmlu_eval_accuracy_astronomy": 0.6875,
      "mmlu_eval_accuracy_business_ethics": 0.7272727272727273,
      "mmlu_eval_accuracy_clinical_knowledge": 0.5172413793103449,
      "mmlu_eval_accuracy_college_biology": 0.5625,
      "mmlu_eval_accuracy_college_chemistry": 0.375,
      "mmlu_eval_accuracy_college_computer_science": 0.18181818181818182,
      "mmlu_eval_accuracy_college_mathematics": 0.36363636363636365,
      "mmlu_eval_accuracy_college_medicine": 0.5909090909090909,
      "mmlu_eval_accuracy_college_physics": 0.36363636363636365,
      "mmlu_eval_accuracy_computer_security": 0.45454545454545453,
      "mmlu_eval_accuracy_conceptual_physics": 0.38461538461538464,
      "mmlu_eval_accuracy_econometrics": 0.3333333333333333,
      "mmlu_eval_accuracy_electrical_engineering": 0.4375,
      "mmlu_eval_accuracy_elementary_mathematics": 0.4146341463414634,
      "mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
      "mmlu_eval_accuracy_global_facts": 0.4,
      "mmlu_eval_accuracy_high_school_biology": 0.5,
      "mmlu_eval_accuracy_high_school_chemistry": 0.3181818181818182,
      "mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
      "mmlu_eval_accuracy_high_school_european_history": 0.7222222222222222,
      "mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.6190476190476191,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.627906976744186,
      "mmlu_eval_accuracy_high_school_mathematics": 0.27586206896551724,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.46153846153846156,
      "mmlu_eval_accuracy_high_school_physics": 0.23529411764705882,
      "mmlu_eval_accuracy_high_school_psychology": 0.85,
      "mmlu_eval_accuracy_high_school_statistics": 0.30434782608695654,
      "mmlu_eval_accuracy_high_school_us_history": 0.8636363636363636,
      "mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
      "mmlu_eval_accuracy_human_aging": 0.782608695652174,
      "mmlu_eval_accuracy_human_sexuality": 0.5,
      "mmlu_eval_accuracy_international_law": 0.9230769230769231,
      "mmlu_eval_accuracy_jurisprudence": 0.5454545454545454,
      "mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
      "mmlu_eval_accuracy_machine_learning": 0.45454545454545453,
      "mmlu_eval_accuracy_management": 0.9090909090909091,
      "mmlu_eval_accuracy_marketing": 0.92,
      "mmlu_eval_accuracy_medical_genetics": 1.0,
      "mmlu_eval_accuracy_miscellaneous": 0.7209302325581395,
      "mmlu_eval_accuracy_moral_disputes": 0.631578947368421,
      "mmlu_eval_accuracy_moral_scenarios": 0.33,
      "mmlu_eval_accuracy_nutrition": 0.7878787878787878,
      "mmlu_eval_accuracy_philosophy": 0.6176470588235294,
      "mmlu_eval_accuracy_prehistory": 0.5428571428571428,
      "mmlu_eval_accuracy_professional_accounting": 0.5161290322580645,
      "mmlu_eval_accuracy_professional_law": 0.37058823529411766,
      "mmlu_eval_accuracy_professional_medicine": 0.6129032258064516,
      "mmlu_eval_accuracy_professional_psychology": 0.5797101449275363,
      "mmlu_eval_accuracy_public_relations": 0.5,
      "mmlu_eval_accuracy_security_studies": 0.5925925925925926,
      "mmlu_eval_accuracy_sociology": 0.8636363636363636,
      "mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
      "mmlu_eval_accuracy_virology": 0.5555555555555556,
      "mmlu_eval_accuracy_world_religions": 0.7368421052631579,
      "mmlu_loss": 1.3124866759886007,
      "step": 400
    }
  ],
  "logging_steps": 1,
  "max_steps": 15585,
  "num_train_epochs": 3,
  "save_steps": 200,
  "total_flos": 5.84029668655104e+16,
  "trial_name": null,
  "trial_params": null
}