mistralic-expert-24 / checkpoint-600 /trainer_state.json
pharaouk's picture
Training in progress, step 600, checkpoint
815a8a3
raw
history blame
75.1 kB
{
"best_metric": 0.5414947867393494,
"best_model_checkpoint": "experts/mistralic-expert-24/checkpoint-600",
"epoch": 0.10803997479067255,
"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": 0.5117,
"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
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4771,
"step": 201
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6302,
"step": 202
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7167,
"step": 203
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5508,
"step": 204
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5985,
"step": 205
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6444,
"step": 206
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5587,
"step": 207
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4161,
"step": 208
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4295,
"step": 209
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4249,
"step": 210
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3347,
"step": 211
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5797,
"step": 212
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.472,
"step": 213
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4323,
"step": 214
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3656,
"step": 215
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5046,
"step": 216
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7236,
"step": 217
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3978,
"step": 218
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.724,
"step": 219
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6719,
"step": 220
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4198,
"step": 221
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4589,
"step": 222
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4209,
"step": 223
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4574,
"step": 224
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6178,
"step": 225
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4134,
"step": 226
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3615,
"step": 227
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4253,
"step": 228
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6592,
"step": 229
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4886,
"step": 230
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.603,
"step": 231
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4899,
"step": 232
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5666,
"step": 233
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6954,
"step": 234
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6629,
"step": 235
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5663,
"step": 236
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.413,
"step": 237
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4868,
"step": 238
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5815,
"step": 239
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6621,
"step": 240
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6301,
"step": 241
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3807,
"step": 242
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4793,
"step": 243
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4946,
"step": 244
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5446,
"step": 245
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6933,
"step": 246
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0135,
"step": 247
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.556,
"step": 248
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5444,
"step": 249
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6291,
"step": 250
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5632,
"step": 251
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.508,
"step": 252
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3671,
"step": 253
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6577,
"step": 254
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.607,
"step": 255
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.392,
"step": 256
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.563,
"step": 257
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.443,
"step": 258
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3944,
"step": 259
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4961,
"step": 260
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5628,
"step": 261
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4118,
"step": 262
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4639,
"step": 263
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5075,
"step": 264
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4799,
"step": 265
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4849,
"step": 266
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6582,
"step": 267
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5181,
"step": 268
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.539,
"step": 269
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.487,
"step": 270
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6015,
"step": 271
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6475,
"step": 272
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5932,
"step": 273
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5924,
"step": 274
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3837,
"step": 275
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8193,
"step": 276
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5695,
"step": 277
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7658,
"step": 278
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4919,
"step": 279
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.46,
"step": 280
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4583,
"step": 281
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3968,
"step": 282
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3873,
"step": 283
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5765,
"step": 284
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6067,
"step": 285
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2832,
"step": 286
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4241,
"step": 287
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6125,
"step": 288
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4966,
"step": 289
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6386,
"step": 290
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5911,
"step": 291
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3884,
"step": 292
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.53,
"step": 293
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4554,
"step": 294
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4077,
"step": 295
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3473,
"step": 296
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.2661,
"step": 297
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6654,
"step": 298
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5642,
"step": 299
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.553,
"step": 300
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4969,
"step": 301
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5998,
"step": 302
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5133,
"step": 303
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6338,
"step": 304
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.3908,
"step": 305
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4956,
"step": 306
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4665,
"step": 307
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.562,
"step": 308
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8452,
"step": 309
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6849,
"step": 310
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5413,
"step": 311
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3369,
"step": 312
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4665,
"step": 313
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3569,
"step": 314
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7633,
"step": 315
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3404,
"step": 316
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4228,
"step": 317
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.48,
"step": 318
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3888,
"step": 319
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6661,
"step": 320
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9178,
"step": 321
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5153,
"step": 322
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5238,
"step": 323
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5639,
"step": 324
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6523,
"step": 325
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4562,
"step": 326
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4474,
"step": 327
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7445,
"step": 328
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6328,
"step": 329
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5792,
"step": 330
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4711,
"step": 331
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.623,
"step": 332
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4799,
"step": 333
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5311,
"step": 334
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6722,
"step": 335
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6247,
"step": 336
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6586,
"step": 337
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4482,
"step": 338
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4584,
"step": 339
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5161,
"step": 340
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7856,
"step": 341
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5149,
"step": 342
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5313,
"step": 343
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4932,
"step": 344
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.3697,
"step": 345
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4181,
"step": 346
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6921,
"step": 347
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4734,
"step": 348
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5266,
"step": 349
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5752,
"step": 350
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6372,
"step": 351
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.354,
"step": 352
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5545,
"step": 353
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5169,
"step": 354
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6304,
"step": 355
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5059,
"step": 356
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4003,
"step": 357
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5806,
"step": 358
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5514,
"step": 359
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4679,
"step": 360
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3588,
"step": 361
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5031,
"step": 362
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5722,
"step": 363
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.558,
"step": 364
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4432,
"step": 365
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3995,
"step": 366
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6473,
"step": 367
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4249,
"step": 368
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5646,
"step": 369
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6557,
"step": 370
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6462,
"step": 371
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5351,
"step": 372
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.47,
"step": 373
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4184,
"step": 374
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4879,
"step": 375
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5237,
"step": 376
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4797,
"step": 377
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5242,
"step": 378
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4848,
"step": 379
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5238,
"step": 380
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6566,
"step": 381
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6197,
"step": 382
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4861,
"step": 383
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5668,
"step": 384
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4406,
"step": 385
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4544,
"step": 386
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6223,
"step": 387
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5145,
"step": 388
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3692,
"step": 389
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5162,
"step": 390
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5107,
"step": 391
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6502,
"step": 392
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4908,
"step": 393
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5245,
"step": 394
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4745,
"step": 395
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6319,
"step": 396
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4801,
"step": 397
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5535,
"step": 398
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5974,
"step": 399
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5451,
"step": 400
},
{
"epoch": 0.07,
"eval_loss": 0.5446667075157166,
"eval_runtime": 144.7879,
"eval_samples_per_second": 6.907,
"eval_steps_per_second": 3.453,
"step": 400
},
{
"epoch": 0.07,
"mmlu_eval_accuracy": 0.59563842789031,
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
"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": 0.6818181818181818,
"mmlu_eval_accuracy_college_physics": 0.36363636363636365,
"mmlu_eval_accuracy_computer_security": 0.6363636363636364,
"mmlu_eval_accuracy_conceptual_physics": 0.5384615384615384,
"mmlu_eval_accuracy_econometrics": 0.6666666666666666,
"mmlu_eval_accuracy_electrical_engineering": 0.75,
"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": 0.59375,
"mmlu_eval_accuracy_high_school_chemistry": 0.45454545454545453,
"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": 0.3793103448275862,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6153846153846154,
"mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
"mmlu_eval_accuracy_high_school_psychology": 0.85,
"mmlu_eval_accuracy_high_school_statistics": 0.30434782608695654,
"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": 0.5833333333333334,
"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": 0.36363636363636365,
"mmlu_eval_accuracy_management": 0.8181818181818182,
"mmlu_eval_accuracy_marketing": 0.92,
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
"mmlu_eval_accuracy_miscellaneous": 0.7558139534883721,
"mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
"mmlu_eval_accuracy_moral_scenarios": 0.33,
"mmlu_eval_accuracy_nutrition": 0.6666666666666666,
"mmlu_eval_accuracy_philosophy": 0.7352941176470589,
"mmlu_eval_accuracy_prehistory": 0.5142857142857142,
"mmlu_eval_accuracy_professional_accounting": 0.5161290322580645,
"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": 0.6296296296296297,
"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": 0.8947368421052632,
"mmlu_loss": 1.3502731320907801,
"step": 400
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3468,
"step": 401
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.472,
"step": 402
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4599,
"step": 403
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7189,
"step": 404
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4516,
"step": 405
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3887,
"step": 406
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.2939,
"step": 407
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4829,
"step": 408
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5295,
"step": 409
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4269,
"step": 410
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3893,
"step": 411
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4653,
"step": 412
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.2917,
"step": 413
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.574,
"step": 414
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5537,
"step": 415
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3899,
"step": 416
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4671,
"step": 417
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5915,
"step": 418
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4878,
"step": 419
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5554,
"step": 420
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3838,
"step": 421
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4592,
"step": 422
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5172,
"step": 423
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5647,
"step": 424
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7022,
"step": 425
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.636,
"step": 426
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3475,
"step": 427
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4519,
"step": 428
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6054,
"step": 429
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4783,
"step": 430
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5805,
"step": 431
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5028,
"step": 432
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5735,
"step": 433
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5809,
"step": 434
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5188,
"step": 435
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5045,
"step": 436
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5549,
"step": 437
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4818,
"step": 438
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4809,
"step": 439
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5145,
"step": 440
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4645,
"step": 441
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5796,
"step": 442
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4327,
"step": 443
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4538,
"step": 444
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5876,
"step": 445
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5775,
"step": 446
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3551,
"step": 447
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6221,
"step": 448
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5273,
"step": 449
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8619,
"step": 450
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6748,
"step": 451
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5478,
"step": 452
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6225,
"step": 453
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4775,
"step": 454
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7795,
"step": 455
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4995,
"step": 456
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3697,
"step": 457
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6555,
"step": 458
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4833,
"step": 459
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4076,
"step": 460
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5413,
"step": 461
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7217,
"step": 462
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4547,
"step": 463
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5387,
"step": 464
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5941,
"step": 465
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5118,
"step": 466
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3596,
"step": 467
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3598,
"step": 468
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5262,
"step": 469
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5125,
"step": 470
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.3624,
"step": 471
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5775,
"step": 472
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7362,
"step": 473
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3322,
"step": 474
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4117,
"step": 475
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.457,
"step": 476
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5103,
"step": 477
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3227,
"step": 478
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6907,
"step": 479
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3241,
"step": 480
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6381,
"step": 481
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5874,
"step": 482
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4597,
"step": 483
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5769,
"step": 484
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5261,
"step": 485
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6467,
"step": 486
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3657,
"step": 487
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6098,
"step": 488
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3525,
"step": 489
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4836,
"step": 490
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4528,
"step": 491
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5486,
"step": 492
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.48,
"step": 493
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4883,
"step": 494
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6014,
"step": 495
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6339,
"step": 496
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6686,
"step": 497
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5518,
"step": 498
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4066,
"step": 499
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6109,
"step": 500
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3591,
"step": 501
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5504,
"step": 502
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4453,
"step": 503
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4683,
"step": 504
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.2399,
"step": 505
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6047,
"step": 506
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6541,
"step": 507
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.509,
"step": 508
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.2881,
"step": 509
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4289,
"step": 510
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5731,
"step": 511
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5484,
"step": 512
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5527,
"step": 513
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3615,
"step": 514
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.63,
"step": 515
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5487,
"step": 516
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7353,
"step": 517
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4918,
"step": 518
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6171,
"step": 519
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5706,
"step": 520
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.3239,
"step": 521
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4646,
"step": 522
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.355,
"step": 523
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4947,
"step": 524
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6404,
"step": 525
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7208,
"step": 526
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5575,
"step": 527
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4053,
"step": 528
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3755,
"step": 529
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5134,
"step": 530
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7995,
"step": 531
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6039,
"step": 532
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5288,
"step": 533
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4056,
"step": 534
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4696,
"step": 535
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3226,
"step": 536
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4974,
"step": 537
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.356,
"step": 538
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3916,
"step": 539
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5339,
"step": 540
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5466,
"step": 541
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4411,
"step": 542
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4353,
"step": 543
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6056,
"step": 544
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4129,
"step": 545
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5044,
"step": 546
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5708,
"step": 547
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5067,
"step": 548
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5268,
"step": 549
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.872,
"step": 550
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4811,
"step": 551
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4171,
"step": 552
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3894,
"step": 553
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5298,
"step": 554
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5155,
"step": 555
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3351,
"step": 556
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.3583,
"step": 557
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8062,
"step": 558
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4545,
"step": 559
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7254,
"step": 560
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.521,
"step": 561
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4765,
"step": 562
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5918,
"step": 563
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4671,
"step": 564
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4906,
"step": 565
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5199,
"step": 566
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.619,
"step": 567
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.684,
"step": 568
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5165,
"step": 569
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4393,
"step": 570
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.51,
"step": 571
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5991,
"step": 572
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.2851,
"step": 573
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5848,
"step": 574
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6357,
"step": 575
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4779,
"step": 576
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.541,
"step": 577
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7716,
"step": 578
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5097,
"step": 579
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6648,
"step": 580
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.429,
"step": 581
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7619,
"step": 582
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6017,
"step": 583
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7838,
"step": 584
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5616,
"step": 585
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4064,
"step": 586
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3623,
"step": 587
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6836,
"step": 588
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5234,
"step": 589
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6112,
"step": 590
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.587,
"step": 591
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.682,
"step": 592
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4636,
"step": 593
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5116,
"step": 594
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5997,
"step": 595
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3824,
"step": 596
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3724,
"step": 597
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.526,
"step": 598
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3089,
"step": 599
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8847,
"step": 600
},
{
"epoch": 0.11,
"eval_loss": 0.5414947867393494,
"eval_runtime": 144.7335,
"eval_samples_per_second": 6.909,
"eval_steps_per_second": 3.455,
"step": 600
},
{
"epoch": 0.11,
"mmlu_eval_accuracy": 0.6038475689431909,
"mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
"mmlu_eval_accuracy_anatomy": 0.5714285714285714,
"mmlu_eval_accuracy_astronomy": 0.6875,
"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": 0.7272727272727273,
"mmlu_eval_accuracy_college_mathematics": 0.5454545454545454,
"mmlu_eval_accuracy_college_medicine": 0.6818181818181818,
"mmlu_eval_accuracy_college_physics": 0.36363636363636365,
"mmlu_eval_accuracy_computer_security": 0.5454545454545454,
"mmlu_eval_accuracy_conceptual_physics": 0.5384615384615384,
"mmlu_eval_accuracy_econometrics": 0.6666666666666666,
"mmlu_eval_accuracy_electrical_engineering": 0.6875,
"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": 0.53125,
"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": 0.4482758620689655,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6538461538461539,
"mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
"mmlu_eval_accuracy_high_school_psychology": 0.8333333333333334,
"mmlu_eval_accuracy_high_school_statistics": 0.43478260869565216,
"mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
"mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
"mmlu_eval_accuracy_human_aging": 0.782608695652174,
"mmlu_eval_accuracy_human_sexuality": 0.5,
"mmlu_eval_accuracy_international_law": 0.9230769230769231,
"mmlu_eval_accuracy_jurisprudence": 0.5454545454545454,
"mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
"mmlu_eval_accuracy_machine_learning": 0.36363636363636365,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.88,
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
"mmlu_eval_accuracy_miscellaneous": 0.7441860465116279,
"mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
"mmlu_eval_accuracy_moral_scenarios": 0.33,
"mmlu_eval_accuracy_nutrition": 0.7575757575757576,
"mmlu_eval_accuracy_philosophy": 0.7647058823529411,
"mmlu_eval_accuracy_prehistory": 0.5428571428571428,
"mmlu_eval_accuracy_professional_accounting": 0.5483870967741935,
"mmlu_eval_accuracy_professional_law": 0.38823529411764707,
"mmlu_eval_accuracy_professional_medicine": 0.6129032258064516,
"mmlu_eval_accuracy_professional_psychology": 0.5507246376811594,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.7407407407407407,
"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": 0.8947368421052632,
"mmlu_loss": 1.2973803256105816,
"step": 600
}
],
"logging_steps": 1,
"max_steps": 16659,
"num_train_epochs": 3,
"save_steps": 200,
"total_flos": 1.332030465099694e+17,
"trial_name": null,
"trial_params": null
}