mistralic-expert-28 / checkpoint-600 /trainer_state.json
pharaouk's picture
Training in progress, step 600, checkpoint
e07efb4
raw
history blame
75.2 kB
{
"best_metric": 0.8959276676177979,
"best_model_checkpoint": "experts/mistralic-expert-28/checkpoint-600",
"epoch": 0.10399965333448888,
"eval_steps": 200,
"global_step": 600,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9153,
"step": 1
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.1126,
"step": 2
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9623,
"step": 3
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.0142,
"step": 4
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.0401,
"step": 5
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7844,
"step": 6
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.014,
"step": 7
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8062,
"step": 8
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7818,
"step": 9
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9638,
"step": 10
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7674,
"step": 11
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.1185,
"step": 12
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8292,
"step": 13
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6469,
"step": 14
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9843,
"step": 15
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.1966,
"step": 16
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.0369,
"step": 17
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.0368,
"step": 18
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9693,
"step": 19
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8299,
"step": 20
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9227,
"step": 21
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7564,
"step": 22
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8868,
"step": 23
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.1065,
"step": 24
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8981,
"step": 25
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9463,
"step": 26
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.756,
"step": 27
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8951,
"step": 28
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8101,
"step": 29
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8278,
"step": 30
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9454,
"step": 31
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9271,
"step": 32
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6077,
"step": 33
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0646,
"step": 34
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6949,
"step": 35
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9631,
"step": 36
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9045,
"step": 37
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9562,
"step": 38
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9692,
"step": 39
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9586,
"step": 40
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9513,
"step": 41
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0374,
"step": 42
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9099,
"step": 43
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9617,
"step": 44
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0033,
"step": 45
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0036,
"step": 46
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9478,
"step": 47
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6875,
"step": 48
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.893,
"step": 49
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9539,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8106,
"step": 51
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8526,
"step": 52
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0052,
"step": 53
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9359,
"step": 54
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8657,
"step": 55
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9159,
"step": 56
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7594,
"step": 57
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9969,
"step": 58
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9412,
"step": 59
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8695,
"step": 60
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9306,
"step": 61
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.1756,
"step": 62
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8805,
"step": 63
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0183,
"step": 64
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9008,
"step": 65
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.787,
"step": 66
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8999,
"step": 67
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8642,
"step": 68
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8307,
"step": 69
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0866,
"step": 70
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8396,
"step": 71
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.3022,
"step": 72
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9115,
"step": 73
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0308,
"step": 74
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9449,
"step": 75
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.835,
"step": 76
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0277,
"step": 77
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0807,
"step": 78
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9567,
"step": 79
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9698,
"step": 80
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9644,
"step": 81
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7428,
"step": 82
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9928,
"step": 83
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0731,
"step": 84
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8638,
"step": 85
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9244,
"step": 86
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6978,
"step": 87
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0289,
"step": 88
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7879,
"step": 89
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0854,
"step": 90
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8177,
"step": 91
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0794,
"step": 92
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.3084,
"step": 93
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8073,
"step": 94
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9544,
"step": 95
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8831,
"step": 96
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8451,
"step": 97
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9125,
"step": 98
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7928,
"step": 99
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8143,
"step": 100
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0305,
"step": 101
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.1199,
"step": 102
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9476,
"step": 103
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9556,
"step": 104
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9603,
"step": 105
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.1322,
"step": 106
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0186,
"step": 107
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0976,
"step": 108
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6639,
"step": 109
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7569,
"step": 110
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8701,
"step": 111
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0863,
"step": 112
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9724,
"step": 113
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8466,
"step": 114
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7405,
"step": 115
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0396,
"step": 116
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0059,
"step": 117
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8913,
"step": 118
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.1489,
"step": 119
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8515,
"step": 120
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9792,
"step": 121
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.1725,
"step": 122
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8935,
"step": 123
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9956,
"step": 124
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9844,
"step": 125
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9799,
"step": 126
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6775,
"step": 127
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7439,
"step": 128
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9888,
"step": 129
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9094,
"step": 130
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9001,
"step": 131
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6952,
"step": 132
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8925,
"step": 133
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9453,
"step": 134
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0495,
"step": 135
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5741,
"step": 136
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.1232,
"step": 137
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9066,
"step": 138
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0007,
"step": 139
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7621,
"step": 140
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8324,
"step": 141
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9299,
"step": 142
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8145,
"step": 143
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6945,
"step": 144
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9305,
"step": 145
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0048,
"step": 146
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7729,
"step": 147
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0078,
"step": 148
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7944,
"step": 149
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7829,
"step": 150
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0636,
"step": 151
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.02,
"step": 152
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8744,
"step": 153
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9944,
"step": 154
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0413,
"step": 155
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9772,
"step": 156
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0797,
"step": 157
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8422,
"step": 158
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9591,
"step": 159
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8085,
"step": 160
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6908,
"step": 161
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0377,
"step": 162
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9779,
"step": 163
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8234,
"step": 164
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8714,
"step": 165
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.1599,
"step": 166
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8281,
"step": 167
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9024,
"step": 168
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9552,
"step": 169
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9298,
"step": 170
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8903,
"step": 171
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9371,
"step": 172
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8271,
"step": 173
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0959,
"step": 174
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9753,
"step": 175
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9881,
"step": 176
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.955,
"step": 177
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8817,
"step": 178
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8931,
"step": 179
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8116,
"step": 180
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8405,
"step": 181
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0196,
"step": 182
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0874,
"step": 183
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8415,
"step": 184
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8357,
"step": 185
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8063,
"step": 186
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9091,
"step": 187
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8249,
"step": 188
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9477,
"step": 189
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9192,
"step": 190
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.1802,
"step": 191
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.1188,
"step": 192
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7995,
"step": 193
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.2284,
"step": 194
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9915,
"step": 195
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0516,
"step": 196
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8023,
"step": 197
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8398,
"step": 198
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0297,
"step": 199
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9599,
"step": 200
},
{
"epoch": 0.03,
"eval_loss": 0.918748676776886,
"eval_runtime": 110.4336,
"eval_samples_per_second": 9.055,
"eval_steps_per_second": 4.528,
"step": 200
},
{
"epoch": 0.03,
"mmlu_eval_accuracy": 0.5936350541565827,
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
"mmlu_eval_accuracy_anatomy": 0.5714285714285714,
"mmlu_eval_accuracy_astronomy": 0.8125,
"mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
"mmlu_eval_accuracy_clinical_knowledge": 0.5862068965517241,
"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": 0.6363636363636364,
"mmlu_eval_accuracy_college_medicine": 0.6363636363636364,
"mmlu_eval_accuracy_college_physics": 0.45454545454545453,
"mmlu_eval_accuracy_computer_security": 0.5454545454545454,
"mmlu_eval_accuracy_conceptual_physics": 0.5384615384615384,
"mmlu_eval_accuracy_econometrics": 0.5833333333333334,
"mmlu_eval_accuracy_electrical_engineering": 0.625,
"mmlu_eval_accuracy_elementary_mathematics": 0.4146341463414634,
"mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
"mmlu_eval_accuracy_global_facts": 0.3,
"mmlu_eval_accuracy_high_school_biology": 0.59375,
"mmlu_eval_accuracy_high_school_chemistry": 0.3181818181818182,
"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": 0.627906976744186,
"mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
"mmlu_eval_accuracy_high_school_microeconomics": 0.5769230769230769,
"mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
"mmlu_eval_accuracy_high_school_psychology": 0.8333333333333334,
"mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
"mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
"mmlu_eval_accuracy_high_school_world_history": 0.6153846153846154,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.5,
"mmlu_eval_accuracy_international_law": 0.8461538461538461,
"mmlu_eval_accuracy_jurisprudence": 0.5454545454545454,
"mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
"mmlu_eval_accuracy_machine_learning": 0.5454545454545454,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.92,
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
"mmlu_eval_accuracy_miscellaneous": 0.7441860465116279,
"mmlu_eval_accuracy_moral_disputes": 0.5789473684210527,
"mmlu_eval_accuracy_moral_scenarios": 0.36,
"mmlu_eval_accuracy_nutrition": 0.696969696969697,
"mmlu_eval_accuracy_philosophy": 0.7352941176470589,
"mmlu_eval_accuracy_prehistory": 0.5428571428571428,
"mmlu_eval_accuracy_professional_accounting": 0.4838709677419355,
"mmlu_eval_accuracy_professional_law": 0.4117647058823529,
"mmlu_eval_accuracy_professional_medicine": 0.6129032258064516,
"mmlu_eval_accuracy_professional_psychology": 0.6086956521739131,
"mmlu_eval_accuracy_public_relations": 0.4166666666666667,
"mmlu_eval_accuracy_security_studies": 0.6296296296296297,
"mmlu_eval_accuracy_sociology": 0.9090909090909091,
"mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
"mmlu_eval_accuracy_virology": 0.5555555555555556,
"mmlu_eval_accuracy_world_religions": 0.8421052631578947,
"mmlu_loss": 1.3518375517805314,
"step": 200
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9508,
"step": 201
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6532,
"step": 202
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.934,
"step": 203
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8147,
"step": 204
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.483,
"step": 205
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9612,
"step": 206
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9069,
"step": 207
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.1141,
"step": 208
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.04,
"step": 209
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8768,
"step": 210
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0499,
"step": 211
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8034,
"step": 212
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.1387,
"step": 213
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0499,
"step": 214
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9957,
"step": 215
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8957,
"step": 216
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8687,
"step": 217
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8224,
"step": 218
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0095,
"step": 219
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.225,
"step": 220
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8774,
"step": 221
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8663,
"step": 222
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7407,
"step": 223
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8002,
"step": 224
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9033,
"step": 225
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.2643,
"step": 226
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9764,
"step": 227
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0823,
"step": 228
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8546,
"step": 229
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0453,
"step": 230
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8317,
"step": 231
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0105,
"step": 232
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9766,
"step": 233
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8363,
"step": 234
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8998,
"step": 235
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9636,
"step": 236
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7943,
"step": 237
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0646,
"step": 238
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9673,
"step": 239
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9132,
"step": 240
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9139,
"step": 241
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9085,
"step": 242
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9753,
"step": 243
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9792,
"step": 244
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9407,
"step": 245
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7692,
"step": 246
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.908,
"step": 247
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0076,
"step": 248
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.1506,
"step": 249
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9199,
"step": 250
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.1417,
"step": 251
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0081,
"step": 252
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0554,
"step": 253
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9647,
"step": 254
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9484,
"step": 255
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9182,
"step": 256
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9055,
"step": 257
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9412,
"step": 258
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8184,
"step": 259
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0315,
"step": 260
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8093,
"step": 261
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8895,
"step": 262
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9222,
"step": 263
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0261,
"step": 264
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9542,
"step": 265
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7613,
"step": 266
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.862,
"step": 267
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8565,
"step": 268
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9046,
"step": 269
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.2807,
"step": 270
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0658,
"step": 271
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8939,
"step": 272
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9156,
"step": 273
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9697,
"step": 274
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7813,
"step": 275
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8824,
"step": 276
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9141,
"step": 277
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9661,
"step": 278
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.008,
"step": 279
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0581,
"step": 280
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9922,
"step": 281
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9843,
"step": 282
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.02,
"step": 283
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8112,
"step": 284
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7316,
"step": 285
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5477,
"step": 286
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8083,
"step": 287
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9947,
"step": 288
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0805,
"step": 289
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8961,
"step": 290
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0162,
"step": 291
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7729,
"step": 292
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.2236,
"step": 293
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6903,
"step": 294
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9266,
"step": 295
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8389,
"step": 296
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.1089,
"step": 297
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8078,
"step": 298
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0642,
"step": 299
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9602,
"step": 300
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0887,
"step": 301
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9027,
"step": 302
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0864,
"step": 303
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0036,
"step": 304
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5798,
"step": 305
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6762,
"step": 306
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9747,
"step": 307
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.775,
"step": 308
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8708,
"step": 309
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.008,
"step": 310
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7898,
"step": 311
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9316,
"step": 312
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0352,
"step": 313
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.958,
"step": 314
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9278,
"step": 315
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.861,
"step": 316
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9259,
"step": 317
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0211,
"step": 318
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.1027,
"step": 319
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.909,
"step": 320
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.696,
"step": 321
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9981,
"step": 322
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9565,
"step": 323
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9335,
"step": 324
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0028,
"step": 325
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8264,
"step": 326
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8432,
"step": 327
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0344,
"step": 328
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9314,
"step": 329
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8693,
"step": 330
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9567,
"step": 331
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.1389,
"step": 332
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.1522,
"step": 333
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9452,
"step": 334
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9818,
"step": 335
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8578,
"step": 336
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.2169,
"step": 337
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8826,
"step": 338
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6723,
"step": 339
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.111,
"step": 340
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8057,
"step": 341
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4511,
"step": 342
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7871,
"step": 343
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9753,
"step": 344
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7468,
"step": 345
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9199,
"step": 346
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.95,
"step": 347
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.3679,
"step": 348
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.78,
"step": 349
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8936,
"step": 350
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8851,
"step": 351
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8275,
"step": 352
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9469,
"step": 353
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9664,
"step": 354
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7857,
"step": 355
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9312,
"step": 356
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.1484,
"step": 357
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7847,
"step": 358
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0041,
"step": 359
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0418,
"step": 360
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0907,
"step": 361
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.012,
"step": 362
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8481,
"step": 363
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7312,
"step": 364
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9435,
"step": 365
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0647,
"step": 366
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0256,
"step": 367
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8531,
"step": 368
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8118,
"step": 369
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9456,
"step": 370
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7462,
"step": 371
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7342,
"step": 372
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9343,
"step": 373
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8028,
"step": 374
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.1676,
"step": 375
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0576,
"step": 376
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0029,
"step": 377
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0235,
"step": 378
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7557,
"step": 379
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7831,
"step": 380
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.023,
"step": 381
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7636,
"step": 382
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.868,
"step": 383
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.039,
"step": 384
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0509,
"step": 385
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0969,
"step": 386
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0429,
"step": 387
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.995,
"step": 388
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.1526,
"step": 389
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.2525,
"step": 390
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7519,
"step": 391
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9599,
"step": 392
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7064,
"step": 393
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0201,
"step": 394
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0668,
"step": 395
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.857,
"step": 396
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.1075,
"step": 397
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.878,
"step": 398
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9995,
"step": 399
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8536,
"step": 400
},
{
"epoch": 0.07,
"eval_loss": 0.9089547395706177,
"eval_runtime": 110.2762,
"eval_samples_per_second": 9.068,
"eval_steps_per_second": 4.534,
"step": 400
},
{
"epoch": 0.07,
"mmlu_eval_accuracy": 0.5882033985113294,
"mmlu_eval_accuracy_abstract_algebra": 0.18181818181818182,
"mmlu_eval_accuracy_anatomy": 0.42857142857142855,
"mmlu_eval_accuracy_astronomy": 0.625,
"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": 0.45454545454545453,
"mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
"mmlu_eval_accuracy_college_medicine": 0.5909090909090909,
"mmlu_eval_accuracy_college_physics": 0.45454545454545453,
"mmlu_eval_accuracy_computer_security": 0.5454545454545454,
"mmlu_eval_accuracy_conceptual_physics": 0.5,
"mmlu_eval_accuracy_econometrics": 0.6666666666666666,
"mmlu_eval_accuracy_electrical_engineering": 0.6875,
"mmlu_eval_accuracy_elementary_mathematics": 0.43902439024390244,
"mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
"mmlu_eval_accuracy_global_facts": 0.4,
"mmlu_eval_accuracy_high_school_biology": 0.625,
"mmlu_eval_accuracy_high_school_chemistry": 0.5,
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
"mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
"mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.5813953488372093,
"mmlu_eval_accuracy_high_school_mathematics": 0.41379310344827586,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6153846153846154,
"mmlu_eval_accuracy_high_school_physics": 0.058823529411764705,
"mmlu_eval_accuracy_high_school_psychology": 0.8833333333333333,
"mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
"mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
"mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.5833333333333334,
"mmlu_eval_accuracy_international_law": 0.8461538461538461,
"mmlu_eval_accuracy_jurisprudence": 0.6363636363636364,
"mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
"mmlu_eval_accuracy_machine_learning": 0.2727272727272727,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.88,
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
"mmlu_eval_accuracy_miscellaneous": 0.7674418604651163,
"mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
"mmlu_eval_accuracy_moral_scenarios": 0.36,
"mmlu_eval_accuracy_nutrition": 0.6666666666666666,
"mmlu_eval_accuracy_philosophy": 0.7058823529411765,
"mmlu_eval_accuracy_prehistory": 0.5428571428571428,
"mmlu_eval_accuracy_professional_accounting": 0.5161290322580645,
"mmlu_eval_accuracy_professional_law": 0.38235294117647056,
"mmlu_eval_accuracy_professional_medicine": 0.6774193548387096,
"mmlu_eval_accuracy_professional_psychology": 0.5797101449275363,
"mmlu_eval_accuracy_public_relations": 0.4166666666666667,
"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": 0.6111111111111112,
"mmlu_eval_accuracy_world_religions": 0.8421052631578947,
"mmlu_loss": 1.3722924169532313,
"step": 400
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8379,
"step": 401
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8581,
"step": 402
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.1749,
"step": 403
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0031,
"step": 404
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8899,
"step": 405
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7814,
"step": 406
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8567,
"step": 407
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.109,
"step": 408
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9617,
"step": 409
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9198,
"step": 410
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7959,
"step": 411
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9849,
"step": 412
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8212,
"step": 413
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9083,
"step": 414
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7546,
"step": 415
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7893,
"step": 416
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5201,
"step": 417
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.142,
"step": 418
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.969,
"step": 419
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6761,
"step": 420
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9495,
"step": 421
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.1075,
"step": 422
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.1908,
"step": 423
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7321,
"step": 424
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9699,
"step": 425
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9044,
"step": 426
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.92,
"step": 427
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6132,
"step": 428
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0147,
"step": 429
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7865,
"step": 430
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0387,
"step": 431
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0642,
"step": 432
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9613,
"step": 433
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9591,
"step": 434
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9993,
"step": 435
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9081,
"step": 436
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.1343,
"step": 437
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7894,
"step": 438
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0983,
"step": 439
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9792,
"step": 440
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9052,
"step": 441
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0188,
"step": 442
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8964,
"step": 443
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9913,
"step": 444
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7435,
"step": 445
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.943,
"step": 446
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9065,
"step": 447
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.901,
"step": 448
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7939,
"step": 449
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.2298,
"step": 450
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0091,
"step": 451
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8213,
"step": 452
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.756,
"step": 453
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8005,
"step": 454
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0129,
"step": 455
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9463,
"step": 456
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7104,
"step": 457
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8058,
"step": 458
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.1198,
"step": 459
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9775,
"step": 460
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8958,
"step": 461
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8772,
"step": 462
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0402,
"step": 463
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.4543,
"step": 464
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.919,
"step": 465
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6586,
"step": 466
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8691,
"step": 467
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7884,
"step": 468
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7535,
"step": 469
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0748,
"step": 470
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7725,
"step": 471
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9248,
"step": 472
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5976,
"step": 473
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.079,
"step": 474
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8693,
"step": 475
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5991,
"step": 476
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6166,
"step": 477
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9733,
"step": 478
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9233,
"step": 479
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8304,
"step": 480
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9823,
"step": 481
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8229,
"step": 482
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0206,
"step": 483
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9035,
"step": 484
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9331,
"step": 485
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7454,
"step": 486
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7747,
"step": 487
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8711,
"step": 488
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8918,
"step": 489
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8776,
"step": 490
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0387,
"step": 491
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6836,
"step": 492
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8048,
"step": 493
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9795,
"step": 494
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7852,
"step": 495
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0038,
"step": 496
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.1319,
"step": 497
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9535,
"step": 498
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9704,
"step": 499
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8074,
"step": 500
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9755,
"step": 501
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9822,
"step": 502
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9083,
"step": 503
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.901,
"step": 504
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0439,
"step": 505
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9061,
"step": 506
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9485,
"step": 507
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7206,
"step": 508
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9206,
"step": 509
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8356,
"step": 510
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.2195,
"step": 511
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6338,
"step": 512
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8957,
"step": 513
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9846,
"step": 514
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9582,
"step": 515
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0594,
"step": 516
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8856,
"step": 517
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0264,
"step": 518
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9973,
"step": 519
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7677,
"step": 520
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0392,
"step": 521
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0031,
"step": 522
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9803,
"step": 523
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.1092,
"step": 524
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9817,
"step": 525
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9701,
"step": 526
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8437,
"step": 527
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8808,
"step": 528
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9509,
"step": 529
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9676,
"step": 530
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.2117,
"step": 531
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9391,
"step": 532
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0672,
"step": 533
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8743,
"step": 534
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0313,
"step": 535
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7439,
"step": 536
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9263,
"step": 537
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.1877,
"step": 538
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9968,
"step": 539
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.1528,
"step": 540
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8137,
"step": 541
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6874,
"step": 542
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8553,
"step": 543
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.933,
"step": 544
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.781,
"step": 545
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7481,
"step": 546
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7041,
"step": 547
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9119,
"step": 548
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9267,
"step": 549
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8273,
"step": 550
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9242,
"step": 551
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6097,
"step": 552
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.3431,
"step": 553
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.025,
"step": 554
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.94,
"step": 555
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6028,
"step": 556
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8675,
"step": 557
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.779,
"step": 558
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0838,
"step": 559
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7469,
"step": 560
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7401,
"step": 561
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9701,
"step": 562
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7933,
"step": 563
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7776,
"step": 564
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8945,
"step": 565
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8268,
"step": 566
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8963,
"step": 567
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7957,
"step": 568
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0115,
"step": 569
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8464,
"step": 570
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0025,
"step": 571
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7957,
"step": 572
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.1203,
"step": 573
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9207,
"step": 574
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7012,
"step": 575
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8518,
"step": 576
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9828,
"step": 577
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9933,
"step": 578
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.1687,
"step": 579
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6325,
"step": 580
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6514,
"step": 581
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0105,
"step": 582
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5911,
"step": 583
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.1909,
"step": 584
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7191,
"step": 585
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7781,
"step": 586
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0846,
"step": 587
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8459,
"step": 588
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9575,
"step": 589
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0807,
"step": 590
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8731,
"step": 591
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9409,
"step": 592
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8081,
"step": 593
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.2209,
"step": 594
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.124,
"step": 595
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7926,
"step": 596
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8573,
"step": 597
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8204,
"step": 598
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7193,
"step": 599
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7782,
"step": 600
},
{
"epoch": 0.1,
"eval_loss": 0.8959276676177979,
"eval_runtime": 110.3877,
"eval_samples_per_second": 9.059,
"eval_steps_per_second": 4.529,
"step": 600
},
{
"epoch": 0.1,
"mmlu_eval_accuracy": 0.5554035320736654,
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
"mmlu_eval_accuracy_anatomy": 0.5,
"mmlu_eval_accuracy_astronomy": 0.6875,
"mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
"mmlu_eval_accuracy_clinical_knowledge": 0.5862068965517241,
"mmlu_eval_accuracy_college_biology": 0.625,
"mmlu_eval_accuracy_college_chemistry": 0.25,
"mmlu_eval_accuracy_college_computer_science": 0.18181818181818182,
"mmlu_eval_accuracy_college_mathematics": 0.36363636363636365,
"mmlu_eval_accuracy_college_medicine": 0.5454545454545454,
"mmlu_eval_accuracy_college_physics": 0.45454545454545453,
"mmlu_eval_accuracy_computer_security": 0.45454545454545453,
"mmlu_eval_accuracy_conceptual_physics": 0.38461538461538464,
"mmlu_eval_accuracy_econometrics": 0.4166666666666667,
"mmlu_eval_accuracy_electrical_engineering": 0.375,
"mmlu_eval_accuracy_elementary_mathematics": 0.3902439024390244,
"mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
"mmlu_eval_accuracy_global_facts": 0.4,
"mmlu_eval_accuracy_high_school_biology": 0.46875,
"mmlu_eval_accuracy_high_school_chemistry": 0.3181818181818182,
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
"mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
"mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.5813953488372093,
"mmlu_eval_accuracy_high_school_mathematics": 0.20689655172413793,
"mmlu_eval_accuracy_high_school_microeconomics": 0.5384615384615384,
"mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
"mmlu_eval_accuracy_high_school_psychology": 0.8666666666666667,
"mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
"mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
"mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
"mmlu_eval_accuracy_international_law": 0.9230769230769231,
"mmlu_eval_accuracy_jurisprudence": 0.6363636363636364,
"mmlu_eval_accuracy_logical_fallacies": 0.6111111111111112,
"mmlu_eval_accuracy_machine_learning": 0.45454545454545453,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.96,
"mmlu_eval_accuracy_medical_genetics": 1.0,
"mmlu_eval_accuracy_miscellaneous": 0.7325581395348837,
"mmlu_eval_accuracy_moral_disputes": 0.5,
"mmlu_eval_accuracy_moral_scenarios": 0.39,
"mmlu_eval_accuracy_nutrition": 0.696969696969697,
"mmlu_eval_accuracy_philosophy": 0.7352941176470589,
"mmlu_eval_accuracy_prehistory": 0.5428571428571428,
"mmlu_eval_accuracy_professional_accounting": 0.45161290322580644,
"mmlu_eval_accuracy_professional_law": 0.40588235294117647,
"mmlu_eval_accuracy_professional_medicine": 0.5483870967741935,
"mmlu_eval_accuracy_professional_psychology": 0.6521739130434783,
"mmlu_eval_accuracy_public_relations": 0.4166666666666667,
"mmlu_eval_accuracy_security_studies": 0.5925925925925926,
"mmlu_eval_accuracy_sociology": 0.8181818181818182,
"mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
"mmlu_eval_accuracy_virology": 0.5,
"mmlu_eval_accuracy_world_religions": 0.7368421052631579,
"mmlu_loss": 1.1482681673282742,
"step": 600
}
],
"logging_steps": 1,
"max_steps": 17307,
"num_train_epochs": 3,
"save_steps": 200,
"total_flos": 8.963299646334566e+16,
"trial_name": null,
"trial_params": null
}