mistralic-expert-25 / checkpoint-200 /trainer_state.json
pharaouk's picture
Training in progress, step 200, checkpoint
13de902
raw
history blame
25.3 kB
{
"best_metric": 0.8657823204994202,
"best_model_checkpoint": "experts/mistralic-expert-25/checkpoint-200",
"epoch": 0.029117379435850774,
"eval_steps": 200,
"global_step": 200,
"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
}
],
"logging_steps": 1,
"max_steps": 20604,
"num_train_epochs": 3,
"save_steps": 200,
"total_flos": 4.663782708058522e+16,
"trial_name": null,
"trial_params": null
}