pharaouk's picture
Training in progress, step 1200, checkpoint
f0d0fd9
raw
history blame
150 kB
{
"best_metric": 0.5329148173332214,
"best_model_checkpoint": "experts/mistralic-expert-24/checkpoint-1200",
"epoch": 0.2160799495813451,
"eval_steps": 200,
"global_step": 1200,
"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
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4642,
"step": 601
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5805,
"step": 602
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4317,
"step": 603
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4954,
"step": 604
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5975,
"step": 605
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5586,
"step": 606
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6258,
"step": 607
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6829,
"step": 608
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.581,
"step": 609
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3792,
"step": 610
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6389,
"step": 611
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4764,
"step": 612
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4157,
"step": 613
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4616,
"step": 614
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4526,
"step": 615
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3869,
"step": 616
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5001,
"step": 617
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6661,
"step": 618
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3069,
"step": 619
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.532,
"step": 620
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4815,
"step": 621
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5963,
"step": 622
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4567,
"step": 623
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5038,
"step": 624
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.535,
"step": 625
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6171,
"step": 626
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4042,
"step": 627
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.527,
"step": 628
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6249,
"step": 629
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5914,
"step": 630
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5425,
"step": 631
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5134,
"step": 632
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4873,
"step": 633
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6505,
"step": 634
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5515,
"step": 635
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4668,
"step": 636
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4767,
"step": 637
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7176,
"step": 638
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4866,
"step": 639
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.405,
"step": 640
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5024,
"step": 641
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8296,
"step": 642
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4577,
"step": 643
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5099,
"step": 644
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5029,
"step": 645
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6126,
"step": 646
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.465,
"step": 647
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3062,
"step": 648
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5562,
"step": 649
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7371,
"step": 650
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.609,
"step": 651
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5509,
"step": 652
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.751,
"step": 653
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.9078,
"step": 654
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6472,
"step": 655
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5766,
"step": 656
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3951,
"step": 657
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4733,
"step": 658
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6431,
"step": 659
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.513,
"step": 660
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5163,
"step": 661
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5054,
"step": 662
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4747,
"step": 663
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4521,
"step": 664
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4414,
"step": 665
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4021,
"step": 666
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5453,
"step": 667
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6636,
"step": 668
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5122,
"step": 669
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6873,
"step": 670
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4264,
"step": 671
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4749,
"step": 672
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5014,
"step": 673
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6289,
"step": 674
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5373,
"step": 675
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5185,
"step": 676
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.3556,
"step": 677
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5715,
"step": 678
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4015,
"step": 679
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4127,
"step": 680
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5251,
"step": 681
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5674,
"step": 682
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4494,
"step": 683
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5874,
"step": 684
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4337,
"step": 685
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4706,
"step": 686
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6157,
"step": 687
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5256,
"step": 688
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4893,
"step": 689
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6024,
"step": 690
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7282,
"step": 691
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5943,
"step": 692
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5004,
"step": 693
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5015,
"step": 694
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4632,
"step": 695
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5742,
"step": 696
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5978,
"step": 697
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6051,
"step": 698
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5645,
"step": 699
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4623,
"step": 700
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.2784,
"step": 701
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.458,
"step": 702
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5949,
"step": 703
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6304,
"step": 704
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5583,
"step": 705
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4257,
"step": 706
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4948,
"step": 707
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5785,
"step": 708
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4898,
"step": 709
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5131,
"step": 710
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6441,
"step": 711
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5909,
"step": 712
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.478,
"step": 713
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4122,
"step": 714
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.539,
"step": 715
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4476,
"step": 716
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4447,
"step": 717
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.425,
"step": 718
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.3987,
"step": 719
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.524,
"step": 720
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8292,
"step": 721
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.3988,
"step": 722
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5389,
"step": 723
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.496,
"step": 724
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5422,
"step": 725
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6515,
"step": 726
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8199,
"step": 727
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.391,
"step": 728
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.3745,
"step": 729
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6721,
"step": 730
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6148,
"step": 731
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6323,
"step": 732
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6396,
"step": 733
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5892,
"step": 734
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4324,
"step": 735
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4461,
"step": 736
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.457,
"step": 737
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4041,
"step": 738
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6304,
"step": 739
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5858,
"step": 740
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4544,
"step": 741
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5471,
"step": 742
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4115,
"step": 743
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.3517,
"step": 744
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4389,
"step": 745
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5832,
"step": 746
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6529,
"step": 747
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7869,
"step": 748
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5561,
"step": 749
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5794,
"step": 750
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4393,
"step": 751
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.488,
"step": 752
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.419,
"step": 753
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5628,
"step": 754
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7296,
"step": 755
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6726,
"step": 756
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6309,
"step": 757
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5297,
"step": 758
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4851,
"step": 759
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.2835,
"step": 760
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6063,
"step": 761
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5745,
"step": 762
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4849,
"step": 763
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.3366,
"step": 764
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4994,
"step": 765
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.431,
"step": 766
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5118,
"step": 767
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5479,
"step": 768
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6533,
"step": 769
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.596,
"step": 770
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.3035,
"step": 771
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.3941,
"step": 772
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7275,
"step": 773
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.648,
"step": 774
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4947,
"step": 775
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5208,
"step": 776
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5287,
"step": 777
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4572,
"step": 778
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.712,
"step": 779
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6872,
"step": 780
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5623,
"step": 781
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4497,
"step": 782
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.8345,
"step": 783
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.3833,
"step": 784
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5827,
"step": 785
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4106,
"step": 786
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5307,
"step": 787
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4739,
"step": 788
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4821,
"step": 789
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5066,
"step": 790
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7109,
"step": 791
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5116,
"step": 792
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6195,
"step": 793
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4742,
"step": 794
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5385,
"step": 795
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5441,
"step": 796
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4244,
"step": 797
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4749,
"step": 798
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5881,
"step": 799
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5864,
"step": 800
},
{
"epoch": 0.14,
"eval_loss": 0.5366095304489136,
"eval_runtime": 144.6064,
"eval_samples_per_second": 6.915,
"eval_steps_per_second": 3.458,
"step": 800
},
{
"epoch": 0.14,
"mmlu_eval_accuracy": 0.6021210906273604,
"mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
"mmlu_eval_accuracy_anatomy": 0.5,
"mmlu_eval_accuracy_astronomy": 0.8125,
"mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
"mmlu_eval_accuracy_clinical_knowledge": 0.5517241379310345,
"mmlu_eval_accuracy_college_biology": 0.625,
"mmlu_eval_accuracy_college_chemistry": 0.375,
"mmlu_eval_accuracy_college_computer_science": 0.6363636363636364,
"mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
"mmlu_eval_accuracy_college_medicine": 0.6363636363636364,
"mmlu_eval_accuracy_college_physics": 0.36363636363636365,
"mmlu_eval_accuracy_computer_security": 0.5454545454545454,
"mmlu_eval_accuracy_conceptual_physics": 0.46153846153846156,
"mmlu_eval_accuracy_econometrics": 0.75,
"mmlu_eval_accuracy_electrical_engineering": 0.6875,
"mmlu_eval_accuracy_elementary_mathematics": 0.36585365853658536,
"mmlu_eval_accuracy_formal_logic": 0.07142857142857142,
"mmlu_eval_accuracy_global_facts": 0.3,
"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": 0.8636363636363636,
"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": 0.7727272727272727,
"mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
"mmlu_eval_accuracy_human_aging": 0.782608695652174,
"mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
"mmlu_eval_accuracy_international_law": 0.8461538461538461,
"mmlu_eval_accuracy_jurisprudence": 0.6363636363636364,
"mmlu_eval_accuracy_logical_fallacies": 0.7222222222222222,
"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": 0.7441860465116279,
"mmlu_eval_accuracy_moral_disputes": 0.5789473684210527,
"mmlu_eval_accuracy_moral_scenarios": 0.31,
"mmlu_eval_accuracy_nutrition": 0.8181818181818182,
"mmlu_eval_accuracy_philosophy": 0.7352941176470589,
"mmlu_eval_accuracy_prehistory": 0.5428571428571428,
"mmlu_eval_accuracy_professional_accounting": 0.5483870967741935,
"mmlu_eval_accuracy_professional_law": 0.4,
"mmlu_eval_accuracy_professional_medicine": 0.6451612903225806,
"mmlu_eval_accuracy_professional_psychology": 0.6086956521739131,
"mmlu_eval_accuracy_public_relations": 0.4166666666666667,
"mmlu_eval_accuracy_security_studies": 0.7037037037037037,
"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": 0.8947368421052632,
"mmlu_loss": 1.290061207998826,
"step": 800
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4491,
"step": 801
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4534,
"step": 802
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4491,
"step": 803
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4132,
"step": 804
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.3319,
"step": 805
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4763,
"step": 806
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4999,
"step": 807
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4367,
"step": 808
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4247,
"step": 809
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5459,
"step": 810
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.511,
"step": 811
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5284,
"step": 812
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4084,
"step": 813
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4515,
"step": 814
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4664,
"step": 815
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.3393,
"step": 816
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4565,
"step": 817
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5084,
"step": 818
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6686,
"step": 819
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4817,
"step": 820
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5601,
"step": 821
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.3879,
"step": 822
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6512,
"step": 823
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6336,
"step": 824
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.3862,
"step": 825
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5604,
"step": 826
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.7159,
"step": 827
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.568,
"step": 828
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5482,
"step": 829
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.3759,
"step": 830
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5443,
"step": 831
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.3403,
"step": 832
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.566,
"step": 833
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.612,
"step": 834
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5171,
"step": 835
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4977,
"step": 836
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.496,
"step": 837
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4437,
"step": 838
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.469,
"step": 839
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.7219,
"step": 840
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4777,
"step": 841
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.562,
"step": 842
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4256,
"step": 843
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.7645,
"step": 844
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4831,
"step": 845
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6506,
"step": 846
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4757,
"step": 847
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4734,
"step": 848
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5405,
"step": 849
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4276,
"step": 850
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5237,
"step": 851
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4339,
"step": 852
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.449,
"step": 853
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.3583,
"step": 854
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6436,
"step": 855
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5282,
"step": 856
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5722,
"step": 857
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.4473,
"step": 858
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5471,
"step": 859
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5913,
"step": 860
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.4373,
"step": 861
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.8387,
"step": 862
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.3249,
"step": 863
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.4583,
"step": 864
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5322,
"step": 865
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.7051,
"step": 866
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5374,
"step": 867
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.4996,
"step": 868
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.7896,
"step": 869
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.473,
"step": 870
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5607,
"step": 871
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6683,
"step": 872
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6066,
"step": 873
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.4384,
"step": 874
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.4536,
"step": 875
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.3649,
"step": 876
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5699,
"step": 877
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.491,
"step": 878
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.3947,
"step": 879
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5217,
"step": 880
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.4693,
"step": 881
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6592,
"step": 882
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5211,
"step": 883
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5648,
"step": 884
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.4001,
"step": 885
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5667,
"step": 886
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.7097,
"step": 887
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.624,
"step": 888
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.64,
"step": 889
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5947,
"step": 890
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6003,
"step": 891
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.4572,
"step": 892
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.3911,
"step": 893
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5089,
"step": 894
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.4634,
"step": 895
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.3578,
"step": 896
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5048,
"step": 897
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6365,
"step": 898
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.398,
"step": 899
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6568,
"step": 900
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.4722,
"step": 901
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5459,
"step": 902
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.3676,
"step": 903
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.4934,
"step": 904
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5609,
"step": 905
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.4205,
"step": 906
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6175,
"step": 907
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.602,
"step": 908
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.4668,
"step": 909
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6476,
"step": 910
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6779,
"step": 911
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5873,
"step": 912
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.4787,
"step": 913
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.3995,
"step": 914
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5447,
"step": 915
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.3917,
"step": 916
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4081,
"step": 917
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.3205,
"step": 918
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6527,
"step": 919
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4622,
"step": 920
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.3732,
"step": 921
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6698,
"step": 922
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.7144,
"step": 923
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6206,
"step": 924
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4928,
"step": 925
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5387,
"step": 926
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.71,
"step": 927
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5363,
"step": 928
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5932,
"step": 929
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5517,
"step": 930
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4709,
"step": 931
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5738,
"step": 932
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4077,
"step": 933
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4895,
"step": 934
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4473,
"step": 935
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5404,
"step": 936
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4606,
"step": 937
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5567,
"step": 938
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.7089,
"step": 939
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5,
"step": 940
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.3836,
"step": 941
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4717,
"step": 942
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5981,
"step": 943
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5045,
"step": 944
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.387,
"step": 945
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4753,
"step": 946
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4837,
"step": 947
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.2862,
"step": 948
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6503,
"step": 949
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5148,
"step": 950
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.449,
"step": 951
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4091,
"step": 952
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5257,
"step": 953
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5068,
"step": 954
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5868,
"step": 955
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5605,
"step": 956
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4437,
"step": 957
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4086,
"step": 958
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6482,
"step": 959
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4213,
"step": 960
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5614,
"step": 961
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4184,
"step": 962
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.3369,
"step": 963
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4805,
"step": 964
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.3853,
"step": 965
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5556,
"step": 966
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5291,
"step": 967
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.3799,
"step": 968
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.553,
"step": 969
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5187,
"step": 970
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5202,
"step": 971
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5255,
"step": 972
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4394,
"step": 973
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.838,
"step": 974
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 1.1685,
"step": 975
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5258,
"step": 976
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4625,
"step": 977
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6283,
"step": 978
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4213,
"step": 979
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.7434,
"step": 980
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4355,
"step": 981
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.3839,
"step": 982
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.631,
"step": 983
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6164,
"step": 984
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6118,
"step": 985
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.7349,
"step": 986
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.2735,
"step": 987
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5808,
"step": 988
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4553,
"step": 989
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4814,
"step": 990
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5072,
"step": 991
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6989,
"step": 992
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6842,
"step": 993
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.7487,
"step": 994
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4894,
"step": 995
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4236,
"step": 996
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6057,
"step": 997
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5865,
"step": 998
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5325,
"step": 999
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.504,
"step": 1000
},
{
"epoch": 0.18,
"eval_loss": 0.5353824496269226,
"eval_runtime": 144.5663,
"eval_samples_per_second": 6.917,
"eval_steps_per_second": 3.459,
"step": 1000
},
{
"epoch": 0.18,
"mmlu_eval_accuracy": 0.6019563348168743,
"mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
"mmlu_eval_accuracy_anatomy": 0.5,
"mmlu_eval_accuracy_astronomy": 0.6875,
"mmlu_eval_accuracy_business_ethics": 0.7272727272727273,
"mmlu_eval_accuracy_clinical_knowledge": 0.5517241379310345,
"mmlu_eval_accuracy_college_biology": 0.625,
"mmlu_eval_accuracy_college_chemistry": 0.375,
"mmlu_eval_accuracy_college_computer_science": 0.5454545454545454,
"mmlu_eval_accuracy_college_mathematics": 0.6363636363636364,
"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": 0.5833333333333334,
"mmlu_eval_accuracy_electrical_engineering": 0.5625,
"mmlu_eval_accuracy_elementary_mathematics": 0.4146341463414634,
"mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
"mmlu_eval_accuracy_global_facts": 0.3,
"mmlu_eval_accuracy_high_school_biology": 0.5625,
"mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
"mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_european_history": 0.8333333333333334,
"mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.6046511627906976,
"mmlu_eval_accuracy_high_school_mathematics": 0.3448275862068966,
"mmlu_eval_accuracy_high_school_microeconomics": 0.5769230769230769,
"mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
"mmlu_eval_accuracy_high_school_psychology": 0.8166666666666667,
"mmlu_eval_accuracy_high_school_statistics": 0.43478260869565216,
"mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
"mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
"mmlu_eval_accuracy_international_law": 0.8461538461538461,
"mmlu_eval_accuracy_jurisprudence": 0.6363636363636364,
"mmlu_eval_accuracy_logical_fallacies": 0.7222222222222222,
"mmlu_eval_accuracy_machine_learning": 0.45454545454545453,
"mmlu_eval_accuracy_management": 0.8181818181818182,
"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": 0.8181818181818182,
"mmlu_eval_accuracy_philosophy": 0.7647058823529411,
"mmlu_eval_accuracy_prehistory": 0.5428571428571428,
"mmlu_eval_accuracy_professional_accounting": 0.6129032258064516,
"mmlu_eval_accuracy_professional_law": 0.3941176470588235,
"mmlu_eval_accuracy_professional_medicine": 0.6129032258064516,
"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": 0.9090909090909091,
"mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
"mmlu_eval_accuracy_virology": 0.5555555555555556,
"mmlu_eval_accuracy_world_religions": 0.8947368421052632,
"mmlu_loss": 1.3596020666566282,
"step": 1000
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6388,
"step": 1001
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4921,
"step": 1002
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4972,
"step": 1003
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.406,
"step": 1004
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4079,
"step": 1005
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.3058,
"step": 1006
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4314,
"step": 1007
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.3648,
"step": 1008
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4541,
"step": 1009
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4812,
"step": 1010
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.3208,
"step": 1011
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4731,
"step": 1012
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4434,
"step": 1013
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6104,
"step": 1014
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4914,
"step": 1015
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5401,
"step": 1016
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6036,
"step": 1017
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5923,
"step": 1018
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5459,
"step": 1019
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4531,
"step": 1020
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5606,
"step": 1021
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5014,
"step": 1022
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.4772,
"step": 1023
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.466,
"step": 1024
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5857,
"step": 1025
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5883,
"step": 1026
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.3402,
"step": 1027
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5107,
"step": 1028
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.7831,
"step": 1029
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.4148,
"step": 1030
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5812,
"step": 1031
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5807,
"step": 1032
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6771,
"step": 1033
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.4808,
"step": 1034
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.4706,
"step": 1035
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.3822,
"step": 1036
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.457,
"step": 1037
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5543,
"step": 1038
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5114,
"step": 1039
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5179,
"step": 1040
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6044,
"step": 1041
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.4552,
"step": 1042
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6384,
"step": 1043
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.48,
"step": 1044
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5248,
"step": 1045
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.3735,
"step": 1046
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.4716,
"step": 1047
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5486,
"step": 1048
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5562,
"step": 1049
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.4019,
"step": 1050
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6075,
"step": 1051
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5703,
"step": 1052
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5469,
"step": 1053
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.4441,
"step": 1054
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6998,
"step": 1055
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5067,
"step": 1056
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.553,
"step": 1057
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6391,
"step": 1058
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5814,
"step": 1059
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.3522,
"step": 1060
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6043,
"step": 1061
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.4488,
"step": 1062
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5095,
"step": 1063
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.4496,
"step": 1064
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.4392,
"step": 1065
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.464,
"step": 1066
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5026,
"step": 1067
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5192,
"step": 1068
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5179,
"step": 1069
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.559,
"step": 1070
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.4476,
"step": 1071
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5546,
"step": 1072
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.4579,
"step": 1073
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.4153,
"step": 1074
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6128,
"step": 1075
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.4041,
"step": 1076
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5751,
"step": 1077
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6639,
"step": 1078
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.8455,
"step": 1079
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.7034,
"step": 1080
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.4609,
"step": 1081
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5666,
"step": 1082
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.3978,
"step": 1083
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6233,
"step": 1084
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.539,
"step": 1085
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5418,
"step": 1086
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5805,
"step": 1087
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.4951,
"step": 1088
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5749,
"step": 1089
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.3894,
"step": 1090
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.7037,
"step": 1091
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.4263,
"step": 1092
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.4605,
"step": 1093
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.4575,
"step": 1094
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6158,
"step": 1095
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.4156,
"step": 1096
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.516,
"step": 1097
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5015,
"step": 1098
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.4069,
"step": 1099
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.8502,
"step": 1100
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.3786,
"step": 1101
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5512,
"step": 1102
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.4587,
"step": 1103
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5744,
"step": 1104
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5341,
"step": 1105
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.539,
"step": 1106
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5067,
"step": 1107
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6217,
"step": 1108
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.3259,
"step": 1109
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.3536,
"step": 1110
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.43,
"step": 1111
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.749,
"step": 1112
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.9327,
"step": 1113
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5477,
"step": 1114
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5311,
"step": 1115
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.3894,
"step": 1116
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6437,
"step": 1117
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.3963,
"step": 1118
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5927,
"step": 1119
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6567,
"step": 1120
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5766,
"step": 1121
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.461,
"step": 1122
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.3639,
"step": 1123
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.4215,
"step": 1124
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5917,
"step": 1125
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5007,
"step": 1126
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.4698,
"step": 1127
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5557,
"step": 1128
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.3694,
"step": 1129
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.4994,
"step": 1130
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5844,
"step": 1131
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.2683,
"step": 1132
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.595,
"step": 1133
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.3744,
"step": 1134
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.7134,
"step": 1135
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6082,
"step": 1136
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6729,
"step": 1137
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.4565,
"step": 1138
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.43,
"step": 1139
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4705,
"step": 1140
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.7478,
"step": 1141
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.6237,
"step": 1142
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.5548,
"step": 1143
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.3497,
"step": 1144
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4131,
"step": 1145
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.6769,
"step": 1146
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.2853,
"step": 1147
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.413,
"step": 1148
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.5703,
"step": 1149
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.6202,
"step": 1150
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.6101,
"step": 1151
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.864,
"step": 1152
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.6377,
"step": 1153
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4582,
"step": 1154
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.3955,
"step": 1155
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.3744,
"step": 1156
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.3419,
"step": 1157
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.3798,
"step": 1158
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.6266,
"step": 1159
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4682,
"step": 1160
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4106,
"step": 1161
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.5606,
"step": 1162
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.5343,
"step": 1163
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.5325,
"step": 1164
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.6973,
"step": 1165
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.5762,
"step": 1166
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.2874,
"step": 1167
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.6796,
"step": 1168
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.5683,
"step": 1169
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4818,
"step": 1170
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4742,
"step": 1171
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4653,
"step": 1172
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4132,
"step": 1173
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4785,
"step": 1174
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.7125,
"step": 1175
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4336,
"step": 1176
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.3961,
"step": 1177
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4037,
"step": 1178
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4204,
"step": 1179
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.6358,
"step": 1180
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.688,
"step": 1181
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4181,
"step": 1182
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.3185,
"step": 1183
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.464,
"step": 1184
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.617,
"step": 1185
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.3663,
"step": 1186
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.423,
"step": 1187
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.3835,
"step": 1188
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.8015,
"step": 1189
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.5,
"step": 1190
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4824,
"step": 1191
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.3845,
"step": 1192
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.5623,
"step": 1193
},
{
"epoch": 0.21,
"learning_rate": 0.0002,
"loss": 0.4671,
"step": 1194
},
{
"epoch": 0.22,
"learning_rate": 0.0002,
"loss": 0.458,
"step": 1195
},
{
"epoch": 0.22,
"learning_rate": 0.0002,
"loss": 0.3773,
"step": 1196
},
{
"epoch": 0.22,
"learning_rate": 0.0002,
"loss": 0.5529,
"step": 1197
},
{
"epoch": 0.22,
"learning_rate": 0.0002,
"loss": 0.5312,
"step": 1198
},
{
"epoch": 0.22,
"learning_rate": 0.0002,
"loss": 0.3734,
"step": 1199
},
{
"epoch": 0.22,
"learning_rate": 0.0002,
"loss": 0.378,
"step": 1200
},
{
"epoch": 0.22,
"eval_loss": 0.5329148173332214,
"eval_runtime": 144.6734,
"eval_samples_per_second": 6.912,
"eval_steps_per_second": 3.456,
"step": 1200
},
{
"epoch": 0.22,
"mmlu_eval_accuracy": 0.5990549539947501,
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
"mmlu_eval_accuracy_anatomy": 0.5714285714285714,
"mmlu_eval_accuracy_astronomy": 0.875,
"mmlu_eval_accuracy_business_ethics": 0.7272727272727273,
"mmlu_eval_accuracy_clinical_knowledge": 0.5517241379310345,
"mmlu_eval_accuracy_college_biology": 0.5,
"mmlu_eval_accuracy_college_chemistry": 0.25,
"mmlu_eval_accuracy_college_computer_science": 0.5454545454545454,
"mmlu_eval_accuracy_college_mathematics": 0.6363636363636364,
"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": 0.5769230769230769,
"mmlu_eval_accuracy_econometrics": 0.6666666666666666,
"mmlu_eval_accuracy_electrical_engineering": 0.625,
"mmlu_eval_accuracy_elementary_mathematics": 0.3902439024390244,
"mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
"mmlu_eval_accuracy_global_facts": 0.3,
"mmlu_eval_accuracy_high_school_biology": 0.5,
"mmlu_eval_accuracy_high_school_chemistry": 0.3181818181818182,
"mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_european_history": 0.8333333333333334,
"mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.6046511627906976,
"mmlu_eval_accuracy_high_school_mathematics": 0.3448275862068966,
"mmlu_eval_accuracy_high_school_microeconomics": 0.5384615384615384,
"mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
"mmlu_eval_accuracy_high_school_psychology": 0.8333333333333334,
"mmlu_eval_accuracy_high_school_statistics": 0.43478260869565216,
"mmlu_eval_accuracy_high_school_us_history": 0.8181818181818182,
"mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
"mmlu_eval_accuracy_human_aging": 0.782608695652174,
"mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
"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": 0.8181818181818182,
"mmlu_eval_accuracy_marketing": 0.84,
"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": 0.32,
"mmlu_eval_accuracy_nutrition": 0.7878787878787878,
"mmlu_eval_accuracy_philosophy": 0.7352941176470589,
"mmlu_eval_accuracy_prehistory": 0.5714285714285714,
"mmlu_eval_accuracy_professional_accounting": 0.5806451612903226,
"mmlu_eval_accuracy_professional_law": 0.3764705882352941,
"mmlu_eval_accuracy_professional_medicine": 0.5483870967741935,
"mmlu_eval_accuracy_professional_psychology": 0.6231884057971014,
"mmlu_eval_accuracy_public_relations": 0.4166666666666667,
"mmlu_eval_accuracy_security_studies": 0.6296296296296297,
"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": 0.9473684210526315,
"mmlu_loss": 1.5290046740449752,
"step": 1200
}
],
"logging_steps": 1,
"max_steps": 16659,
"num_train_epochs": 3,
"save_steps": 200,
"total_flos": 2.6556674279512474e+17,
"trial_name": null,
"trial_params": null
}