mistralic-expert-24 / checkpoint-200 /trainer_state.json
pharaouk's picture
Training in progress, step 200, checkpoint
7b4c91a
raw
history blame
25.3 kB
{
"best_metric": 0.554664134979248,
"best_model_checkpoint": "experts/mistralic-expert-24/checkpoint-200",
"epoch": 0.03601332493022418,
"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": 0.5117,
"step": 1
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.3111,
"step": 2
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.716,
"step": 3
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6607,
"step": 4
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.4214,
"step": 5
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.4492,
"step": 6
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7175,
"step": 7
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5274,
"step": 8
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5826,
"step": 9
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5033,
"step": 10
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5042,
"step": 11
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5294,
"step": 12
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.624,
"step": 13
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5486,
"step": 14
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.4771,
"step": 15
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5612,
"step": 16
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5457,
"step": 17
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6896,
"step": 18
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6553,
"step": 19
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6381,
"step": 20
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.3828,
"step": 21
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.4695,
"step": 22
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.487,
"step": 23
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.3967,
"step": 24
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5751,
"step": 25
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6579,
"step": 26
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5842,
"step": 27
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3797,
"step": 28
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4901,
"step": 29
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5481,
"step": 30
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6913,
"step": 31
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3539,
"step": 32
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5571,
"step": 33
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5536,
"step": 34
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6269,
"step": 35
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7078,
"step": 36
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.45,
"step": 37
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5255,
"step": 38
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3599,
"step": 39
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5937,
"step": 40
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6467,
"step": 41
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.714,
"step": 42
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5231,
"step": 43
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6227,
"step": 44
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7712,
"step": 45
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6226,
"step": 46
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5692,
"step": 47
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4322,
"step": 48
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6904,
"step": 49
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.56,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5573,
"step": 51
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7992,
"step": 52
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5063,
"step": 53
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7108,
"step": 54
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4377,
"step": 55
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5721,
"step": 56
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3159,
"step": 57
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4994,
"step": 58
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4391,
"step": 59
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7907,
"step": 60
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4058,
"step": 61
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7346,
"step": 62
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.61,
"step": 63
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4692,
"step": 64
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4497,
"step": 65
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6778,
"step": 66
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4733,
"step": 67
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5037,
"step": 68
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4756,
"step": 69
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.666,
"step": 70
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.571,
"step": 71
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6768,
"step": 72
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5702,
"step": 73
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5459,
"step": 74
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5258,
"step": 75
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5933,
"step": 76
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3874,
"step": 77
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4945,
"step": 78
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.3502,
"step": 79
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5587,
"step": 80
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4846,
"step": 81
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4432,
"step": 82
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4406,
"step": 83
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4158,
"step": 84
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5382,
"step": 85
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4711,
"step": 86
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4975,
"step": 87
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4515,
"step": 88
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4235,
"step": 89
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4553,
"step": 90
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.451,
"step": 91
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5366,
"step": 92
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5172,
"step": 93
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4486,
"step": 94
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4729,
"step": 95
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.467,
"step": 96
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5829,
"step": 97
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6049,
"step": 98
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8424,
"step": 99
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4827,
"step": 100
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5126,
"step": 101
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6989,
"step": 102
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6008,
"step": 103
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.464,
"step": 104
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5899,
"step": 105
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5255,
"step": 106
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5189,
"step": 107
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5066,
"step": 108
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4434,
"step": 109
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5153,
"step": 110
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4537,
"step": 111
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5494,
"step": 112
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6164,
"step": 113
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5129,
"step": 114
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7047,
"step": 115
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6549,
"step": 116
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5573,
"step": 117
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4964,
"step": 118
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.394,
"step": 119
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4223,
"step": 120
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4477,
"step": 121
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6967,
"step": 122
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4975,
"step": 123
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4863,
"step": 124
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5701,
"step": 125
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4304,
"step": 126
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5954,
"step": 127
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.486,
"step": 128
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6194,
"step": 129
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4526,
"step": 130
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6007,
"step": 131
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9758,
"step": 132
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4996,
"step": 133
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4912,
"step": 134
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6348,
"step": 135
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5226,
"step": 136
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.48,
"step": 137
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.594,
"step": 138
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.66,
"step": 139
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5426,
"step": 140
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6161,
"step": 141
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3786,
"step": 142
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7152,
"step": 143
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4942,
"step": 144
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4854,
"step": 145
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7047,
"step": 146
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7923,
"step": 147
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5231,
"step": 148
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.438,
"step": 149
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4621,
"step": 150
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6048,
"step": 151
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5903,
"step": 152
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5223,
"step": 153
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7192,
"step": 154
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4741,
"step": 155
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4673,
"step": 156
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5428,
"step": 157
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5027,
"step": 158
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6386,
"step": 159
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5278,
"step": 160
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5083,
"step": 161
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4713,
"step": 162
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4939,
"step": 163
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5385,
"step": 164
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4755,
"step": 165
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6208,
"step": 166
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5808,
"step": 167
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6076,
"step": 168
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5491,
"step": 169
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6313,
"step": 170
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4108,
"step": 171
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7379,
"step": 172
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4843,
"step": 173
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.551,
"step": 174
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4906,
"step": 175
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6413,
"step": 176
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6238,
"step": 177
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4573,
"step": 178
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4895,
"step": 179
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4594,
"step": 180
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4024,
"step": 181
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6236,
"step": 182
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4693,
"step": 183
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.345,
"step": 184
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5059,
"step": 185
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3356,
"step": 186
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.3406,
"step": 187
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6401,
"step": 188
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5648,
"step": 189
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6161,
"step": 190
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4426,
"step": 191
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5505,
"step": 192
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4358,
"step": 193
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5671,
"step": 194
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5038,
"step": 195
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3115,
"step": 196
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7811,
"step": 197
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6175,
"step": 198
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6518,
"step": 199
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4179,
"step": 200
},
{
"epoch": 0.04,
"eval_loss": 0.554664134979248,
"eval_runtime": 144.8673,
"eval_samples_per_second": 6.903,
"eval_steps_per_second": 3.451,
"step": 200
},
{
"epoch": 0.04,
"mmlu_eval_accuracy": 0.6035858878610763,
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
"mmlu_eval_accuracy_anatomy": 0.5,
"mmlu_eval_accuracy_astronomy": 0.8125,
"mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
"mmlu_eval_accuracy_clinical_knowledge": 0.5862068965517241,
"mmlu_eval_accuracy_college_biology": 0.5,
"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": 0.6818181818181818,
"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": 0.5833333333333334,
"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": 0.5625,
"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": 0.8636363636363636,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.6511627906976745,
"mmlu_eval_accuracy_high_school_mathematics": 0.3103448275862069,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6538461538461539,
"mmlu_eval_accuracy_high_school_physics": 0.23529411764705882,
"mmlu_eval_accuracy_high_school_psychology": 0.8333333333333334,
"mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
"mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
"mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
"mmlu_eval_accuracy_human_aging": 0.782608695652174,
"mmlu_eval_accuracy_human_sexuality": 0.5,
"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": 0.5454545454545454,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.88,
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
"mmlu_eval_accuracy_miscellaneous": 0.7325581395348837,
"mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
"mmlu_eval_accuracy_moral_scenarios": 0.37,
"mmlu_eval_accuracy_nutrition": 0.696969696969697,
"mmlu_eval_accuracy_philosophy": 0.7647058823529411,
"mmlu_eval_accuracy_prehistory": 0.5142857142857142,
"mmlu_eval_accuracy_professional_accounting": 0.6129032258064516,
"mmlu_eval_accuracy_professional_law": 0.3941176470588235,
"mmlu_eval_accuracy_professional_medicine": 0.6451612903225806,
"mmlu_eval_accuracy_professional_psychology": 0.6086956521739131,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.6666666666666666,
"mmlu_eval_accuracy_sociology": 0.8636363636363636,
"mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
"mmlu_eval_accuracy_virology": 0.5555555555555556,
"mmlu_eval_accuracy_world_religions": 0.8421052631578947,
"mmlu_loss": 1.272096410860592,
"step": 200
}
],
"logging_steps": 1,
"max_steps": 16659,
"num_train_epochs": 3,
"save_steps": 200,
"total_flos": 4.494043558536806e+16,
"trial_name": null,
"trial_params": null
}