mistralic-expert-25 / checkpoint-800 /trainer_state.json
pharaouk's picture
Training in progress, step 800, checkpoint
6400fb2
raw
history blame
100 kB
{
"best_metric": 0.8657823204994202,
"best_model_checkpoint": "experts/mistralic-expert-25/checkpoint-200",
"epoch": 0.1164695177434031,
"eval_steps": 200,
"global_step": 800,
"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": 0.6586,
"step": 1
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7382,
"step": 2
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9771,
"step": 3
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.239,
"step": 4
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.1367,
"step": 5
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9121,
"step": 6
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8723,
"step": 7
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7729,
"step": 8
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9516,
"step": 9
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7457,
"step": 10
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9671,
"step": 11
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8585,
"step": 12
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8002,
"step": 13
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.0357,
"step": 14
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6908,
"step": 15
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9481,
"step": 16
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.0791,
"step": 17
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9394,
"step": 18
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8856,
"step": 19
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7797,
"step": 20
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.1097,
"step": 21
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8909,
"step": 22
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.0102,
"step": 23
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.4585,
"step": 24
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8925,
"step": 25
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9697,
"step": 26
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.0254,
"step": 27
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.0075,
"step": 28
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8735,
"step": 29
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.0522,
"step": 30
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.973,
"step": 31
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.559,
"step": 32
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.4682,
"step": 33
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.0415,
"step": 34
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6446,
"step": 35
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8066,
"step": 36
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5727,
"step": 37
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0717,
"step": 38
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9121,
"step": 39
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8141,
"step": 40
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6281,
"step": 41
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0932,
"step": 42
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.971,
"step": 43
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9128,
"step": 44
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8409,
"step": 45
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0842,
"step": 46
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8916,
"step": 47
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7288,
"step": 48
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.889,
"step": 49
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9508,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.854,
"step": 51
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6206,
"step": 52
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8621,
"step": 53
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4238,
"step": 54
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.2063,
"step": 55
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7552,
"step": 56
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8874,
"step": 57
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8812,
"step": 58
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7006,
"step": 59
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8047,
"step": 60
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.1526,
"step": 61
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5679,
"step": 62
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.794,
"step": 63
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9858,
"step": 64
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9189,
"step": 65
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0108,
"step": 66
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.1895,
"step": 67
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5368,
"step": 68
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7866,
"step": 69
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.868,
"step": 70
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9624,
"step": 71
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.656,
"step": 72
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7712,
"step": 73
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8823,
"step": 74
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6565,
"step": 75
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0115,
"step": 76
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.2936,
"step": 77
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8947,
"step": 78
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6353,
"step": 79
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5608,
"step": 80
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0677,
"step": 81
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8707,
"step": 82
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0677,
"step": 83
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7594,
"step": 84
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6097,
"step": 85
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.1323,
"step": 86
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6166,
"step": 87
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9671,
"step": 88
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9475,
"step": 89
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5055,
"step": 90
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9368,
"step": 91
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0028,
"step": 92
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7944,
"step": 93
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9518,
"step": 94
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6992,
"step": 95
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7948,
"step": 96
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.5601,
"step": 97
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8346,
"step": 98
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9327,
"step": 99
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0734,
"step": 100
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7085,
"step": 101
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8345,
"step": 102
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8653,
"step": 103
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0132,
"step": 104
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6714,
"step": 105
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.1224,
"step": 106
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9094,
"step": 107
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7111,
"step": 108
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8389,
"step": 109
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.994,
"step": 110
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7135,
"step": 111
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6824,
"step": 112
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8392,
"step": 113
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6321,
"step": 114
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5622,
"step": 115
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0738,
"step": 116
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9969,
"step": 117
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.1434,
"step": 118
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7824,
"step": 119
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6799,
"step": 120
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6665,
"step": 121
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.2051,
"step": 122
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7486,
"step": 123
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8522,
"step": 124
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.07,
"step": 125
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9415,
"step": 126
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.622,
"step": 127
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.1949,
"step": 128
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9893,
"step": 129
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6507,
"step": 130
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6626,
"step": 131
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6973,
"step": 132
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0162,
"step": 133
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8232,
"step": 134
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0537,
"step": 135
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9964,
"step": 136
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0375,
"step": 137
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9369,
"step": 138
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7409,
"step": 139
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7771,
"step": 140
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7179,
"step": 141
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0834,
"step": 142
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6802,
"step": 143
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0296,
"step": 144
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9016,
"step": 145
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6938,
"step": 146
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.739,
"step": 147
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7792,
"step": 148
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3007,
"step": 149
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3336,
"step": 150
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8723,
"step": 151
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.1111,
"step": 152
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.1674,
"step": 153
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8059,
"step": 154
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7778,
"step": 155
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7531,
"step": 156
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6723,
"step": 157
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8483,
"step": 158
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0657,
"step": 159
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.1322,
"step": 160
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9167,
"step": 161
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9756,
"step": 162
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7663,
"step": 163
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.135,
"step": 164
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7567,
"step": 165
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0317,
"step": 166
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.1242,
"step": 167
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.977,
"step": 168
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7498,
"step": 169
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8805,
"step": 170
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7917,
"step": 171
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5351,
"step": 172
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.869,
"step": 173
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.2608,
"step": 174
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8666,
"step": 175
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7938,
"step": 176
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8882,
"step": 177
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0029,
"step": 178
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7313,
"step": 179
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8234,
"step": 180
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.882,
"step": 181
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8096,
"step": 182
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8343,
"step": 183
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6169,
"step": 184
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0987,
"step": 185
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8363,
"step": 186
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0218,
"step": 187
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.838,
"step": 188
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9084,
"step": 189
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5031,
"step": 190
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8404,
"step": 191
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8532,
"step": 192
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9322,
"step": 193
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0271,
"step": 194
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9383,
"step": 195
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8817,
"step": 196
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8641,
"step": 197
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6634,
"step": 198
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8597,
"step": 199
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.765,
"step": 200
},
{
"epoch": 0.03,
"eval_loss": 0.8657823204994202,
"eval_runtime": 150.7329,
"eval_samples_per_second": 6.634,
"eval_steps_per_second": 3.317,
"step": 200
},
{
"epoch": 0.03,
"mmlu_eval_accuracy": 0.5909590703251556,
"mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
"mmlu_eval_accuracy_anatomy": 0.5714285714285714,
"mmlu_eval_accuracy_astronomy": 0.75,
"mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
"mmlu_eval_accuracy_clinical_knowledge": 0.5517241379310345,
"mmlu_eval_accuracy_college_biology": 0.625,
"mmlu_eval_accuracy_college_chemistry": 0.25,
"mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
"mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
"mmlu_eval_accuracy_college_medicine": 0.5909090909090909,
"mmlu_eval_accuracy_college_physics": 0.5454545454545454,
"mmlu_eval_accuracy_computer_security": 0.6363636363636364,
"mmlu_eval_accuracy_conceptual_physics": 0.5384615384615384,
"mmlu_eval_accuracy_econometrics": 0.5833333333333334,
"mmlu_eval_accuracy_electrical_engineering": 0.5625,
"mmlu_eval_accuracy_elementary_mathematics": 0.43902439024390244,
"mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
"mmlu_eval_accuracy_global_facts": 0.3,
"mmlu_eval_accuracy_high_school_biology": 0.5625,
"mmlu_eval_accuracy_high_school_chemistry": 0.36363636363636365,
"mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_european_history": 0.8333333333333334,
"mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.5581395348837209,
"mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
"mmlu_eval_accuracy_high_school_microeconomics": 0.5769230769230769,
"mmlu_eval_accuracy_high_school_physics": 0.23529411764705882,
"mmlu_eval_accuracy_high_school_psychology": 0.8833333333333333,
"mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
"mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
"mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.5833333333333334,
"mmlu_eval_accuracy_international_law": 0.9230769230769231,
"mmlu_eval_accuracy_jurisprudence": 0.6363636363636364,
"mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
"mmlu_eval_accuracy_machine_learning": 0.45454545454545453,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.88,
"mmlu_eval_accuracy_medical_genetics": 1.0,
"mmlu_eval_accuracy_miscellaneous": 0.7558139534883721,
"mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
"mmlu_eval_accuracy_moral_scenarios": 0.35,
"mmlu_eval_accuracy_nutrition": 0.6363636363636364,
"mmlu_eval_accuracy_philosophy": 0.7647058823529411,
"mmlu_eval_accuracy_prehistory": 0.5428571428571428,
"mmlu_eval_accuracy_professional_accounting": 0.5161290322580645,
"mmlu_eval_accuracy_professional_law": 0.3941176470588235,
"mmlu_eval_accuracy_professional_medicine": 0.5806451612903226,
"mmlu_eval_accuracy_professional_psychology": 0.5507246376811594,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.6296296296296297,
"mmlu_eval_accuracy_sociology": 0.8181818181818182,
"mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
"mmlu_eval_accuracy_virology": 0.4444444444444444,
"mmlu_eval_accuracy_world_religions": 0.8421052631578947,
"mmlu_loss": 1.5003863334188885,
"step": 200
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8749,
"step": 201
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8911,
"step": 202
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8477,
"step": 203
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.59,
"step": 204
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.68,
"step": 205
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.1289,
"step": 206
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6094,
"step": 207
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7295,
"step": 208
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9877,
"step": 209
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4836,
"step": 210
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.902,
"step": 211
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0939,
"step": 212
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8077,
"step": 213
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.1104,
"step": 214
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0361,
"step": 215
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7082,
"step": 216
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8439,
"step": 217
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.76,
"step": 218
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9476,
"step": 219
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9157,
"step": 220
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.116,
"step": 221
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8784,
"step": 222
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0402,
"step": 223
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0433,
"step": 224
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7033,
"step": 225
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5797,
"step": 226
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8211,
"step": 227
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9549,
"step": 228
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.1082,
"step": 229
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7167,
"step": 230
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.72,
"step": 231
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.3027,
"step": 232
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9015,
"step": 233
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8538,
"step": 234
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.1021,
"step": 235
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8727,
"step": 236
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.14,
"step": 237
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0598,
"step": 238
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9005,
"step": 239
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7569,
"step": 240
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7344,
"step": 241
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5442,
"step": 242
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5014,
"step": 243
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0667,
"step": 244
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6199,
"step": 245
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8805,
"step": 246
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.943,
"step": 247
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7799,
"step": 248
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6903,
"step": 249
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.078,
"step": 250
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8822,
"step": 251
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8223,
"step": 252
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7609,
"step": 253
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.66,
"step": 254
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8807,
"step": 255
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9438,
"step": 256
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7693,
"step": 257
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8977,
"step": 258
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0147,
"step": 259
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7311,
"step": 260
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.1428,
"step": 261
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9863,
"step": 262
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.1905,
"step": 263
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0178,
"step": 264
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0601,
"step": 265
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8216,
"step": 266
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9855,
"step": 267
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9618,
"step": 268
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9469,
"step": 269
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.1452,
"step": 270
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6944,
"step": 271
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0098,
"step": 272
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7943,
"step": 273
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6379,
"step": 274
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0303,
"step": 275
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0032,
"step": 276
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0306,
"step": 277
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8565,
"step": 278
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.264,
"step": 279
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8809,
"step": 280
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9519,
"step": 281
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6438,
"step": 282
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.734,
"step": 283
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7444,
"step": 284
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6762,
"step": 285
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6035,
"step": 286
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9632,
"step": 287
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.78,
"step": 288
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0164,
"step": 289
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.1476,
"step": 290
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8958,
"step": 291
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.1306,
"step": 292
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.1924,
"step": 293
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5887,
"step": 294
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7278,
"step": 295
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.1009,
"step": 296
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9431,
"step": 297
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0868,
"step": 298
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.962,
"step": 299
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.944,
"step": 300
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.1359,
"step": 301
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.1752,
"step": 302
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.2978,
"step": 303
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.1673,
"step": 304
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6636,
"step": 305
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6604,
"step": 306
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9764,
"step": 307
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9834,
"step": 308
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6069,
"step": 309
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8305,
"step": 310
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8291,
"step": 311
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4372,
"step": 312
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6138,
"step": 313
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4996,
"step": 314
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9616,
"step": 315
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.1558,
"step": 316
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5072,
"step": 317
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8821,
"step": 318
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9128,
"step": 319
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.134,
"step": 320
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.842,
"step": 321
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0481,
"step": 322
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0161,
"step": 323
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0036,
"step": 324
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6041,
"step": 325
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.1767,
"step": 326
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6586,
"step": 327
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6043,
"step": 328
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7591,
"step": 329
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7701,
"step": 330
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8548,
"step": 331
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9349,
"step": 332
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7302,
"step": 333
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.295,
"step": 334
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7391,
"step": 335
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5129,
"step": 336
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.837,
"step": 337
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0422,
"step": 338
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9438,
"step": 339
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7376,
"step": 340
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5119,
"step": 341
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6394,
"step": 342
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.1076,
"step": 343
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0006,
"step": 344
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7598,
"step": 345
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8456,
"step": 346
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4541,
"step": 347
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7358,
"step": 348
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8682,
"step": 349
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9197,
"step": 350
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7392,
"step": 351
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.96,
"step": 352
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8633,
"step": 353
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7089,
"step": 354
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7632,
"step": 355
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8844,
"step": 356
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.1651,
"step": 357
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8405,
"step": 358
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9419,
"step": 359
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9078,
"step": 360
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5853,
"step": 361
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3986,
"step": 362
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8585,
"step": 363
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6812,
"step": 364
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9836,
"step": 365
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8018,
"step": 366
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7313,
"step": 367
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.261,
"step": 368
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9301,
"step": 369
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7423,
"step": 370
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4561,
"step": 371
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8535,
"step": 372
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7937,
"step": 373
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0679,
"step": 374
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6891,
"step": 375
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9629,
"step": 376
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5769,
"step": 377
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.2939,
"step": 378
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4939,
"step": 379
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4871,
"step": 380
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9562,
"step": 381
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0184,
"step": 382
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6472,
"step": 383
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.91,
"step": 384
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8464,
"step": 385
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4652,
"step": 386
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9297,
"step": 387
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.1218,
"step": 388
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7709,
"step": 389
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7838,
"step": 390
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0776,
"step": 391
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9132,
"step": 392
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.028,
"step": 393
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.1235,
"step": 394
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8227,
"step": 395
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0908,
"step": 396
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5499,
"step": 397
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9925,
"step": 398
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9207,
"step": 399
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8515,
"step": 400
},
{
"epoch": 0.06,
"eval_loss": 0.876522958278656,
"eval_runtime": 150.8131,
"eval_samples_per_second": 6.631,
"eval_steps_per_second": 3.315,
"step": 400
},
{
"epoch": 0.06,
"mmlu_eval_accuracy": 0.5887418611751553,
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
"mmlu_eval_accuracy_anatomy": 0.42857142857142855,
"mmlu_eval_accuracy_astronomy": 0.75,
"mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
"mmlu_eval_accuracy_clinical_knowledge": 0.6206896551724138,
"mmlu_eval_accuracy_college_biology": 0.625,
"mmlu_eval_accuracy_college_chemistry": 0.25,
"mmlu_eval_accuracy_college_computer_science": 0.5454545454545454,
"mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
"mmlu_eval_accuracy_college_medicine": 0.5,
"mmlu_eval_accuracy_college_physics": 0.5454545454545454,
"mmlu_eval_accuracy_computer_security": 0.7272727272727273,
"mmlu_eval_accuracy_conceptual_physics": 0.46153846153846156,
"mmlu_eval_accuracy_econometrics": 0.5,
"mmlu_eval_accuracy_electrical_engineering": 0.5625,
"mmlu_eval_accuracy_elementary_mathematics": 0.4634146341463415,
"mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
"mmlu_eval_accuracy_global_facts": 0.3,
"mmlu_eval_accuracy_high_school_biology": 0.59375,
"mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
"mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_european_history": 0.8333333333333334,
"mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.6744186046511628,
"mmlu_eval_accuracy_high_school_mathematics": 0.1724137931034483,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6923076923076923,
"mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
"mmlu_eval_accuracy_high_school_psychology": 0.8333333333333334,
"mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
"mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
"mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.5,
"mmlu_eval_accuracy_international_law": 1.0,
"mmlu_eval_accuracy_jurisprudence": 0.5454545454545454,
"mmlu_eval_accuracy_logical_fallacies": 0.6111111111111112,
"mmlu_eval_accuracy_machine_learning": 0.36363636363636365,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.92,
"mmlu_eval_accuracy_medical_genetics": 1.0,
"mmlu_eval_accuracy_miscellaneous": 0.7441860465116279,
"mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
"mmlu_eval_accuracy_moral_scenarios": 0.29,
"mmlu_eval_accuracy_nutrition": 0.6666666666666666,
"mmlu_eval_accuracy_philosophy": 0.6764705882352942,
"mmlu_eval_accuracy_prehistory": 0.4857142857142857,
"mmlu_eval_accuracy_professional_accounting": 0.5483870967741935,
"mmlu_eval_accuracy_professional_law": 0.4,
"mmlu_eval_accuracy_professional_medicine": 0.6129032258064516,
"mmlu_eval_accuracy_professional_psychology": 0.6521739130434783,
"mmlu_eval_accuracy_public_relations": 0.5833333333333334,
"mmlu_eval_accuracy_security_studies": 0.5555555555555556,
"mmlu_eval_accuracy_sociology": 0.8181818181818182,
"mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
"mmlu_eval_accuracy_virology": 0.5555555555555556,
"mmlu_eval_accuracy_world_religions": 0.8421052631578947,
"mmlu_loss": 1.1819022774851975,
"step": 400
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3854,
"step": 401
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9,
"step": 402
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5635,
"step": 403
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7944,
"step": 404
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8498,
"step": 405
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.825,
"step": 406
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.027,
"step": 407
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8133,
"step": 408
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8375,
"step": 409
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.867,
"step": 410
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6564,
"step": 411
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0908,
"step": 412
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8063,
"step": 413
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0551,
"step": 414
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7884,
"step": 415
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.196,
"step": 416
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9368,
"step": 417
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7889,
"step": 418
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.92,
"step": 419
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.923,
"step": 420
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.505,
"step": 421
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6246,
"step": 422
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5029,
"step": 423
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6821,
"step": 424
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0521,
"step": 425
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.042,
"step": 426
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.1292,
"step": 427
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6349,
"step": 428
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.1455,
"step": 429
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.1963,
"step": 430
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.2138,
"step": 431
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8752,
"step": 432
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0621,
"step": 433
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6651,
"step": 434
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.859,
"step": 435
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.716,
"step": 436
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.1293,
"step": 437
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.839,
"step": 438
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7725,
"step": 439
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7605,
"step": 440
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6043,
"step": 441
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7387,
"step": 442
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7862,
"step": 443
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6836,
"step": 444
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7421,
"step": 445
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6429,
"step": 446
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8383,
"step": 447
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9174,
"step": 448
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0507,
"step": 449
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8531,
"step": 450
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9912,
"step": 451
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3324,
"step": 452
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9734,
"step": 453
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8301,
"step": 454
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8535,
"step": 455
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0383,
"step": 456
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7763,
"step": 457
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9024,
"step": 458
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9093,
"step": 459
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0652,
"step": 460
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8565,
"step": 461
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.1394,
"step": 462
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8269,
"step": 463
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8476,
"step": 464
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9835,
"step": 465
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.1196,
"step": 466
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.1287,
"step": 467
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.2022,
"step": 468
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.128,
"step": 469
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.907,
"step": 470
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6075,
"step": 471
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5395,
"step": 472
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8201,
"step": 473
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7534,
"step": 474
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5769,
"step": 475
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8533,
"step": 476
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.811,
"step": 477
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9281,
"step": 478
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7688,
"step": 479
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0322,
"step": 480
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9395,
"step": 481
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.1276,
"step": 482
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7285,
"step": 483
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.3029,
"step": 484
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0498,
"step": 485
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8145,
"step": 486
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0636,
"step": 487
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7355,
"step": 488
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.3619,
"step": 489
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0884,
"step": 490
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4766,
"step": 491
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.1152,
"step": 492
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6886,
"step": 493
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0537,
"step": 494
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8822,
"step": 495
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7957,
"step": 496
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.796,
"step": 497
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8098,
"step": 498
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9354,
"step": 499
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8849,
"step": 500
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9698,
"step": 501
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9888,
"step": 502
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5669,
"step": 503
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.3472,
"step": 504
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0173,
"step": 505
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8667,
"step": 506
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0662,
"step": 507
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9613,
"step": 508
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0865,
"step": 509
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.595,
"step": 510
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0889,
"step": 511
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8027,
"step": 512
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.731,
"step": 513
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6451,
"step": 514
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7651,
"step": 515
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7503,
"step": 516
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5418,
"step": 517
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7422,
"step": 518
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.1012,
"step": 519
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9694,
"step": 520
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6278,
"step": 521
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8709,
"step": 522
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0756,
"step": 523
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.1526,
"step": 524
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7537,
"step": 525
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0647,
"step": 526
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0072,
"step": 527
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8717,
"step": 528
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7804,
"step": 529
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0055,
"step": 530
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5341,
"step": 531
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8038,
"step": 532
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.654,
"step": 533
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8137,
"step": 534
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7152,
"step": 535
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7722,
"step": 536
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6854,
"step": 537
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6428,
"step": 538
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9156,
"step": 539
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9922,
"step": 540
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7666,
"step": 541
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6726,
"step": 542
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7685,
"step": 543
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7058,
"step": 544
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0806,
"step": 545
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8948,
"step": 546
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5322,
"step": 547
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8362,
"step": 548
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8556,
"step": 549
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.777,
"step": 550
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6575,
"step": 551
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.1689,
"step": 552
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9121,
"step": 553
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.515,
"step": 554
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.1138,
"step": 555
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0415,
"step": 556
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9963,
"step": 557
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9907,
"step": 558
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8903,
"step": 559
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9432,
"step": 560
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8081,
"step": 561
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0427,
"step": 562
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.819,
"step": 563
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8198,
"step": 564
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8071,
"step": 565
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6164,
"step": 566
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9555,
"step": 567
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9625,
"step": 568
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8593,
"step": 569
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.1361,
"step": 570
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8922,
"step": 571
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.481,
"step": 572
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.937,
"step": 573
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6731,
"step": 574
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.863,
"step": 575
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4361,
"step": 576
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.1296,
"step": 577
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7308,
"step": 578
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8204,
"step": 579
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7462,
"step": 580
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9368,
"step": 581
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6612,
"step": 582
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.3348,
"step": 583
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7409,
"step": 584
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8789,
"step": 585
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8893,
"step": 586
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.952,
"step": 587
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.579,
"step": 588
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0583,
"step": 589
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.2175,
"step": 590
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6296,
"step": 591
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6771,
"step": 592
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0436,
"step": 593
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9851,
"step": 594
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8748,
"step": 595
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.1432,
"step": 596
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8468,
"step": 597
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5878,
"step": 598
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6763,
"step": 599
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7246,
"step": 600
},
{
"epoch": 0.09,
"eval_loss": 0.8749128580093384,
"eval_runtime": 150.7177,
"eval_samples_per_second": 6.635,
"eval_steps_per_second": 3.317,
"step": 600
},
{
"epoch": 0.09,
"mmlu_eval_accuracy": 0.5945356668386175,
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
"mmlu_eval_accuracy_anatomy": 0.5,
"mmlu_eval_accuracy_astronomy": 0.8125,
"mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
"mmlu_eval_accuracy_clinical_knowledge": 0.5862068965517241,
"mmlu_eval_accuracy_college_biology": 0.6875,
"mmlu_eval_accuracy_college_chemistry": 0.25,
"mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
"mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
"mmlu_eval_accuracy_college_medicine": 0.5454545454545454,
"mmlu_eval_accuracy_college_physics": 0.36363636363636365,
"mmlu_eval_accuracy_computer_security": 0.6363636363636364,
"mmlu_eval_accuracy_conceptual_physics": 0.46153846153846156,
"mmlu_eval_accuracy_econometrics": 0.5,
"mmlu_eval_accuracy_electrical_engineering": 0.6875,
"mmlu_eval_accuracy_elementary_mathematics": 0.43902439024390244,
"mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
"mmlu_eval_accuracy_global_facts": 0.5,
"mmlu_eval_accuracy_high_school_biology": 0.5625,
"mmlu_eval_accuracy_high_school_chemistry": 0.36363636363636365,
"mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_european_history": 0.8333333333333334,
"mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.627906976744186,
"mmlu_eval_accuracy_high_school_mathematics": 0.27586206896551724,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6153846153846154,
"mmlu_eval_accuracy_high_school_physics": 0.23529411764705882,
"mmlu_eval_accuracy_high_school_psychology": 0.8833333333333333,
"mmlu_eval_accuracy_high_school_statistics": 0.43478260869565216,
"mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
"mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.5,
"mmlu_eval_accuracy_international_law": 0.9230769230769231,
"mmlu_eval_accuracy_jurisprudence": 0.5454545454545454,
"mmlu_eval_accuracy_logical_fallacies": 0.6111111111111112,
"mmlu_eval_accuracy_machine_learning": 0.5454545454545454,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.92,
"mmlu_eval_accuracy_medical_genetics": 1.0,
"mmlu_eval_accuracy_miscellaneous": 0.7093023255813954,
"mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
"mmlu_eval_accuracy_moral_scenarios": 0.32,
"mmlu_eval_accuracy_nutrition": 0.696969696969697,
"mmlu_eval_accuracy_philosophy": 0.7352941176470589,
"mmlu_eval_accuracy_prehistory": 0.5142857142857142,
"mmlu_eval_accuracy_professional_accounting": 0.5806451612903226,
"mmlu_eval_accuracy_professional_law": 0.36470588235294116,
"mmlu_eval_accuracy_professional_medicine": 0.5161290322580645,
"mmlu_eval_accuracy_professional_psychology": 0.6376811594202898,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.6296296296296297,
"mmlu_eval_accuracy_sociology": 0.8636363636363636,
"mmlu_eval_accuracy_us_foreign_policy": 0.9090909090909091,
"mmlu_eval_accuracy_virology": 0.4444444444444444,
"mmlu_eval_accuracy_world_religions": 0.7894736842105263,
"mmlu_loss": 1.300421111344046,
"step": 600
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7933,
"step": 601
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6467,
"step": 602
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7209,
"step": 603
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6045,
"step": 604
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7133,
"step": 605
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7786,
"step": 606
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6171,
"step": 607
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8977,
"step": 608
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6011,
"step": 609
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9039,
"step": 610
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0054,
"step": 611
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.884,
"step": 612
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.103,
"step": 613
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9803,
"step": 614
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8419,
"step": 615
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.2294,
"step": 616
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9067,
"step": 617
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6545,
"step": 618
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7901,
"step": 619
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9912,
"step": 620
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.1933,
"step": 621
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0804,
"step": 622
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.1767,
"step": 623
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.1722,
"step": 624
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0863,
"step": 625
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5893,
"step": 626
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.1878,
"step": 627
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5038,
"step": 628
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0097,
"step": 629
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9198,
"step": 630
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9796,
"step": 631
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7624,
"step": 632
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8333,
"step": 633
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.7832,
"step": 634
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8904,
"step": 635
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0267,
"step": 636
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6716,
"step": 637
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7147,
"step": 638
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8542,
"step": 639
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0915,
"step": 640
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7105,
"step": 641
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9823,
"step": 642
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6649,
"step": 643
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5645,
"step": 644
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.938,
"step": 645
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6768,
"step": 646
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8096,
"step": 647
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9646,
"step": 648
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0146,
"step": 649
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.1331,
"step": 650
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9202,
"step": 651
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.671,
"step": 652
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5388,
"step": 653
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7091,
"step": 654
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.1427,
"step": 655
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9963,
"step": 656
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5947,
"step": 657
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9741,
"step": 658
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4016,
"step": 659
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7136,
"step": 660
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0186,
"step": 661
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8837,
"step": 662
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6063,
"step": 663
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0557,
"step": 664
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6535,
"step": 665
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9654,
"step": 666
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8981,
"step": 667
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8701,
"step": 668
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.888,
"step": 669
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6904,
"step": 670
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8483,
"step": 671
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.731,
"step": 672
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0177,
"step": 673
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9986,
"step": 674
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6163,
"step": 675
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7424,
"step": 676
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.175,
"step": 677
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0164,
"step": 678
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6473,
"step": 679
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9453,
"step": 680
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6732,
"step": 681
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.1588,
"step": 682
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8268,
"step": 683
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.2667,
"step": 684
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0797,
"step": 685
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8778,
"step": 686
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0952,
"step": 687
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.686,
"step": 688
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8517,
"step": 689
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9482,
"step": 690
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0309,
"step": 691
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9224,
"step": 692
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6838,
"step": 693
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5536,
"step": 694
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8209,
"step": 695
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.567,
"step": 696
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.1088,
"step": 697
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8476,
"step": 698
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.1508,
"step": 699
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6776,
"step": 700
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.1719,
"step": 701
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7328,
"step": 702
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.1598,
"step": 703
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6818,
"step": 704
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8995,
"step": 705
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9277,
"step": 706
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.3337,
"step": 707
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8181,
"step": 708
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8226,
"step": 709
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.2111,
"step": 710
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9376,
"step": 711
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8854,
"step": 712
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.999,
"step": 713
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9933,
"step": 714
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3753,
"step": 715
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.85,
"step": 716
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9298,
"step": 717
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.936,
"step": 718
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7745,
"step": 719
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5277,
"step": 720
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.1924,
"step": 721
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5876,
"step": 722
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8001,
"step": 723
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7527,
"step": 724
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6356,
"step": 725
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7152,
"step": 726
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.649,
"step": 727
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.219,
"step": 728
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.1549,
"step": 729
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8438,
"step": 730
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.9986,
"step": 731
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8514,
"step": 732
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.0244,
"step": 733
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6488,
"step": 734
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.9765,
"step": 735
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.0243,
"step": 736
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.1347,
"step": 737
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4407,
"step": 738
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8852,
"step": 739
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6979,
"step": 740
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.0013,
"step": 741
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7102,
"step": 742
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5903,
"step": 743
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.655,
"step": 744
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7763,
"step": 745
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.0039,
"step": 746
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.1458,
"step": 747
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.9443,
"step": 748
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8047,
"step": 749
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.1454,
"step": 750
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.0936,
"step": 751
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8144,
"step": 752
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5951,
"step": 753
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.1301,
"step": 754
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8505,
"step": 755
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7856,
"step": 756
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.839,
"step": 757
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.0665,
"step": 758
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5422,
"step": 759
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.1388,
"step": 760
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6734,
"step": 761
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.2507,
"step": 762
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7401,
"step": 763
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.9891,
"step": 764
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8424,
"step": 765
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.749,
"step": 766
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.9126,
"step": 767
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.9094,
"step": 768
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.9753,
"step": 769
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.9845,
"step": 770
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.0428,
"step": 771
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.2379,
"step": 772
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.0308,
"step": 773
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.2342,
"step": 774
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.0322,
"step": 775
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.1706,
"step": 776
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.9846,
"step": 777
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.624,
"step": 778
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.588,
"step": 779
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7001,
"step": 780
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.1015,
"step": 781
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.67,
"step": 782
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7731,
"step": 783
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7011,
"step": 784
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7409,
"step": 785
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.0441,
"step": 786
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.1641,
"step": 787
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7083,
"step": 788
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.0735,
"step": 789
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7574,
"step": 790
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 1.1546,
"step": 791
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8884,
"step": 792
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 1.1335,
"step": 793
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 1.0672,
"step": 794
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6803,
"step": 795
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.9414,
"step": 796
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7552,
"step": 797
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6815,
"step": 798
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7332,
"step": 799
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8239,
"step": 800
},
{
"epoch": 0.12,
"eval_loss": 0.8725178241729736,
"eval_runtime": 150.6633,
"eval_samples_per_second": 6.637,
"eval_steps_per_second": 3.319,
"step": 800
},
{
"epoch": 0.12,
"mmlu_eval_accuracy": 0.5608323161302006,
"mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
"mmlu_eval_accuracy_anatomy": 0.5,
"mmlu_eval_accuracy_astronomy": 0.625,
"mmlu_eval_accuracy_business_ethics": 0.45454545454545453,
"mmlu_eval_accuracy_clinical_knowledge": 0.7241379310344828,
"mmlu_eval_accuracy_college_biology": 0.75,
"mmlu_eval_accuracy_college_chemistry": 0.375,
"mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
"mmlu_eval_accuracy_college_mathematics": 0.09090909090909091,
"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": 0.38461538461538464,
"mmlu_eval_accuracy_econometrics": 0.4166666666666667,
"mmlu_eval_accuracy_electrical_engineering": 0.5625,
"mmlu_eval_accuracy_elementary_mathematics": 0.34146341463414637,
"mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
"mmlu_eval_accuracy_global_facts": 0.5,
"mmlu_eval_accuracy_high_school_biology": 0.46875,
"mmlu_eval_accuracy_high_school_chemistry": 0.45454545454545453,
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
"mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
"mmlu_eval_accuracy_high_school_geography": 0.6818181818181818,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.6046511627906976,
"mmlu_eval_accuracy_high_school_mathematics": 0.20689655172413793,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6538461538461539,
"mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
"mmlu_eval_accuracy_high_school_psychology": 0.85,
"mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
"mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
"mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
"mmlu_eval_accuracy_human_aging": 0.6956521739130435,
"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": 0.6111111111111112,
"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": 0.686046511627907,
"mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
"mmlu_eval_accuracy_moral_scenarios": 0.31,
"mmlu_eval_accuracy_nutrition": 0.6666666666666666,
"mmlu_eval_accuracy_philosophy": 0.6764705882352942,
"mmlu_eval_accuracy_prehistory": 0.4857142857142857,
"mmlu_eval_accuracy_professional_accounting": 0.3870967741935484,
"mmlu_eval_accuracy_professional_law": 0.3764705882352941,
"mmlu_eval_accuracy_professional_medicine": 0.5161290322580645,
"mmlu_eval_accuracy_professional_psychology": 0.6086956521739131,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.6296296296296297,
"mmlu_eval_accuracy_sociology": 0.8636363636363636,
"mmlu_eval_accuracy_us_foreign_policy": 0.7272727272727273,
"mmlu_eval_accuracy_virology": 0.5,
"mmlu_eval_accuracy_world_religions": 0.6842105263157895,
"mmlu_loss": 1.3734735648874825,
"step": 800
}
],
"logging_steps": 1,
"max_steps": 20604,
"num_train_epochs": 3,
"save_steps": 200,
"total_flos": 1.8795497090954035e+17,
"trial_name": null,
"trial_params": null
}