Hugging Face's logo

HydraLM
/
mistralic-expert-26

Model card Files Community
mistralic-expert-26 / checkpoint-800 /trainer_state.json
pharaouk's picture
pharaouk
Training in progress, step 800, checkpoint
18ec1e9
raw
history blame
100 kB
{
"best_metric": 0.743760883808136,
"best_model_checkpoint": "experts/mistralic-expert-26/checkpoint-600",
"epoch": 0.10711297071129706,
"eval_steps": 200,
"global_step": 800,
"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": 0.8469,
"step": 1
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8006,
"step": 2
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.793,
"step": 3
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8309,
"step": 4
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8613,
"step": 5
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7609,
"step": 6
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7405,
"step": 7
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6562,
"step": 8
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8598,
"step": 9
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.8061,
"step": 10
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6911,
"step": 11
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7309,
"step": 12
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6352,
"step": 13
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7019,
"step": 14
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.601,
"step": 15
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7527,
"step": 16
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9178,
"step": 17
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8868,
"step": 18
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9064,
"step": 19
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.646,
"step": 20
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8691,
"step": 21
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9415,
"step": 22
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7216,
"step": 23
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.765,
"step": 24
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7059,
"step": 25
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7747,
"step": 26
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6623,
"step": 27
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7587,
"step": 28
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9956,
"step": 29
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7459,
"step": 30
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9004,
"step": 31
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6704,
"step": 32
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6312,
"step": 33
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6526,
"step": 34
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.632,
"step": 35
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7005,
"step": 36
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.847,
"step": 37
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7952,
"step": 38
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7577,
"step": 39
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7908,
"step": 40
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9948,
"step": 41
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5511,
"step": 42
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6293,
"step": 43
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8038,
"step": 44
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6822,
"step": 45
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8786,
"step": 46
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7836,
"step": 47
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7824,
"step": 48
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7181,
"step": 49
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8472,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7194,
"step": 51
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6755,
"step": 52
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9392,
"step": 53
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7776,
"step": 54
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7357,
"step": 55
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7,
"step": 56
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.801,
"step": 57
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8782,
"step": 58
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8814,
"step": 59
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7144,
"step": 60
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5467,
"step": 61
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.879,
"step": 62
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7626,
"step": 63
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6348,
"step": 64
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7493,
"step": 65
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.703,
"step": 66
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6604,
"step": 67
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7255,
"step": 68
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9477,
"step": 69
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7842,
"step": 70
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7174,
"step": 71
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6468,
"step": 72
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9514,
"step": 73
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6043,
"step": 74
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7093,
"step": 75
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6341,
"step": 76
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6099,
"step": 77
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6215,
"step": 78
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8154,
"step": 79
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6901,
"step": 80
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7121,
"step": 81
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7695,
"step": 82
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7517,
"step": 83
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8143,
"step": 84
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5674,
"step": 85
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6716,
"step": 86
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7034,
"step": 87
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.665,
"step": 88
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7358,
"step": 89
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.65,
"step": 90
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5839,
"step": 91
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7898,
"step": 92
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8046,
"step": 93
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9556,
"step": 94
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6294,
"step": 95
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7129,
"step": 96
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9083,
"step": 97
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6225,
"step": 98
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9371,
"step": 99
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9608,
"step": 100
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7615,
"step": 101
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.85,
"step": 102
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8309,
"step": 103
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6994,
"step": 104
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7748,
"step": 105
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6932,
"step": 106
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7905,
"step": 107
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7735,
"step": 108
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6634,
"step": 109
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.844,
"step": 110
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9069,
"step": 111
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.894,
"step": 112
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7108,
"step": 113
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.797,
"step": 114
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.712,
"step": 115
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7004,
"step": 116
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8133,
"step": 117
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7058,
"step": 118
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5617,
"step": 119
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6603,
"step": 120
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7919,
"step": 121
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8013,
"step": 122
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.543,
"step": 123
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7799,
"step": 124
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7825,
"step": 125
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7331,
"step": 126
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5461,
"step": 127
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.2894,
"step": 128
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5776,
"step": 129
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4728,
"step": 130
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8037,
"step": 131
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8163,
"step": 132
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9718,
"step": 133
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7634,
"step": 134
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7856,
"step": 135
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6623,
"step": 136
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6417,
"step": 137
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8078,
"step": 138
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6473,
"step": 139
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7786,
"step": 140
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.822,
"step": 141
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8149,
"step": 142
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6739,
"step": 143
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.681,
"step": 144
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8158,
"step": 145
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6431,
"step": 146
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8103,
"step": 147
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5391,
"step": 148
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5818,
"step": 149
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9752,
"step": 150
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7014,
"step": 151
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8224,
"step": 152
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7193,
"step": 153
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5579,
"step": 154
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5877,
"step": 155
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8186,
"step": 156
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5337,
"step": 157
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7355,
"step": 158
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8316,
"step": 159
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.795,
"step": 160
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8623,
"step": 161
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7517,
"step": 162
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7901,
"step": 163
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6109,
"step": 164
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8499,
"step": 165
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6832,
"step": 166
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.563,
"step": 167
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7447,
"step": 168
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5976,
"step": 169
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7516,
"step": 170
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5842,
"step": 171
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5788,
"step": 172
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0914,
"step": 173
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7697,
"step": 174
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7694,
"step": 175
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8007,
"step": 176
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8465,
"step": 177
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8914,
"step": 178
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7133,
"step": 179
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8848,
"step": 180
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7373,
"step": 181
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6826,
"step": 182
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7722,
"step": 183
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5657,
"step": 184
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7213,
"step": 185
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6559,
"step": 186
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6655,
"step": 187
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7012,
"step": 188
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8487,
"step": 189
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7951,
"step": 190
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6477,
"step": 191
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7438,
"step": 192
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8507,
"step": 193
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6214,
"step": 194
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6879,
"step": 195
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7684,
"step": 196
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7626,
"step": 197
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6479,
"step": 198
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7303,
"step": 199
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7522,
"step": 200
},
{
"epoch": 0.03,
"eval_loss": 0.7530023455619812,
"eval_runtime": 89.5744,
"eval_samples_per_second": 11.164,
"eval_steps_per_second": 5.582,
"step": 200
},
{
"epoch": 0.03,
"mmlu_eval_accuracy": 0.5961120128104086,
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
"mmlu_eval_accuracy_anatomy": 0.42857142857142855,
"mmlu_eval_accuracy_astronomy": 0.8125,
"mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
"mmlu_eval_accuracy_clinical_knowledge": 0.6206896551724138,
"mmlu_eval_accuracy_college_biology": 0.5625,
"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": 0.5,
"mmlu_eval_accuracy_college_physics": 0.45454545454545453,
"mmlu_eval_accuracy_computer_security": 0.7272727272727273,
"mmlu_eval_accuracy_conceptual_physics": 0.5769230769230769,
"mmlu_eval_accuracy_econometrics": 0.5833333333333334,
"mmlu_eval_accuracy_electrical_engineering": 0.75,
"mmlu_eval_accuracy_elementary_mathematics": 0.34146341463414637,
"mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
"mmlu_eval_accuracy_global_facts": 0.4,
"mmlu_eval_accuracy_high_school_biology": 0.65625,
"mmlu_eval_accuracy_high_school_chemistry": 0.36363636363636365,
"mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
"mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.6046511627906976,
"mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
"mmlu_eval_accuracy_high_school_microeconomics": 0.7307692307692307,
"mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
"mmlu_eval_accuracy_high_school_psychology": 0.8333333333333334,
"mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
"mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
"mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.5833333333333334,
"mmlu_eval_accuracy_international_law": 0.8461538461538461,
"mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
"mmlu_eval_accuracy_logical_fallacies": 0.7222222222222222,
"mmlu_eval_accuracy_machine_learning": 0.36363636363636365,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.84,
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
"mmlu_eval_accuracy_miscellaneous": 0.7325581395348837,
"mmlu_eval_accuracy_moral_disputes": 0.5789473684210527,
"mmlu_eval_accuracy_moral_scenarios": 0.32,
"mmlu_eval_accuracy_nutrition": 0.7575757575757576,
"mmlu_eval_accuracy_philosophy": 0.7058823529411765,
"mmlu_eval_accuracy_prehistory": 0.6285714285714286,
"mmlu_eval_accuracy_professional_accounting": 0.5483870967741935,
"mmlu_eval_accuracy_professional_law": 0.3941176470588235,
"mmlu_eval_accuracy_professional_medicine": 0.5806451612903226,
"mmlu_eval_accuracy_professional_psychology": 0.5362318840579711,
"mmlu_eval_accuracy_public_relations": 0.4166666666666667,
"mmlu_eval_accuracy_security_studies": 0.5925925925925926,
"mmlu_eval_accuracy_sociology": 0.9090909090909091,
"mmlu_eval_accuracy_us_foreign_policy": 0.9090909090909091,
"mmlu_eval_accuracy_virology": 0.4444444444444444,
"mmlu_eval_accuracy_world_religions": 0.8421052631578947,
"mmlu_loss": 1.0777005341205521,
"step": 200
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6374,
"step": 201
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7244,
"step": 202
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6978,
"step": 203
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7097,
"step": 204
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5994,
"step": 205
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7382,
"step": 206
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6402,
"step": 207
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6145,
"step": 208
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7074,
"step": 209
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6967,
"step": 210
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8425,
"step": 211
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7417,
"step": 212
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.2699,
"step": 213
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7487,
"step": 214
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6402,
"step": 215
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7093,
"step": 216
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6423,
"step": 217
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0206,
"step": 218
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7523,
"step": 219
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6754,
"step": 220
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.807,
"step": 221
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9181,
"step": 222
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7908,
"step": 223
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6954,
"step": 224
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7227,
"step": 225
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7656,
"step": 226
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6452,
"step": 227
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7213,
"step": 228
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6493,
"step": 229
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7346,
"step": 230
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7853,
"step": 231
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6369,
"step": 232
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7253,
"step": 233
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7517,
"step": 234
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7005,
"step": 235
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7019,
"step": 236
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6547,
"step": 237
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7818,
"step": 238
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8312,
"step": 239
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5597,
"step": 240
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6868,
"step": 241
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7939,
"step": 242
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5629,
"step": 243
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8805,
"step": 244
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7323,
"step": 245
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6907,
"step": 246
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8249,
"step": 247
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6834,
"step": 248
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6812,
"step": 249
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8043,
"step": 250
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8576,
"step": 251
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7742,
"step": 252
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6906,
"step": 253
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6783,
"step": 254
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6227,
"step": 255
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8105,
"step": 256
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6703,
"step": 257
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.1037,
"step": 258
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6402,
"step": 259
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9586,
"step": 260
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7577,
"step": 261
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8077,
"step": 262
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8953,
"step": 263
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7339,
"step": 264
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6343,
"step": 265
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9431,
"step": 266
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8053,
"step": 267
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7228,
"step": 268
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7385,
"step": 269
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5731,
"step": 270
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8134,
"step": 271
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7456,
"step": 272
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8167,
"step": 273
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6061,
"step": 274
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6861,
"step": 275
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5313,
"step": 276
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6491,
"step": 277
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7365,
"step": 278
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7581,
"step": 279
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7466,
"step": 280
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.116,
"step": 281
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7806,
"step": 282
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8139,
"step": 283
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7923,
"step": 284
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7013,
"step": 285
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6228,
"step": 286
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6846,
"step": 287
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7022,
"step": 288
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6284,
"step": 289
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5942,
"step": 290
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.712,
"step": 291
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6987,
"step": 292
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6721,
"step": 293
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6781,
"step": 294
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6876,
"step": 295
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6842,
"step": 296
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7728,
"step": 297
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7394,
"step": 298
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6089,
"step": 299
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7674,
"step": 300
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6131,
"step": 301
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7942,
"step": 302
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6117,
"step": 303
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5835,
"step": 304
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7201,
"step": 305
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0504,
"step": 306
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6432,
"step": 307
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7559,
"step": 308
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8892,
"step": 309
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0394,
"step": 310
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7912,
"step": 311
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7815,
"step": 312
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7433,
"step": 313
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6478,
"step": 314
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8275,
"step": 315
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8695,
"step": 316
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8166,
"step": 317
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6874,
"step": 318
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7705,
"step": 319
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.603,
"step": 320
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8942,
"step": 321
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8217,
"step": 322
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6167,
"step": 323
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8269,
"step": 324
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7044,
"step": 325
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7432,
"step": 326
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6309,
"step": 327
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7465,
"step": 328
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.793,
"step": 329
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7845,
"step": 330
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7565,
"step": 331
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6906,
"step": 332
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6994,
"step": 333
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7468,
"step": 334
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8354,
"step": 335
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8465,
"step": 336
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7144,
"step": 337
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8756,
"step": 338
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7708,
"step": 339
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7718,
"step": 340
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9794,
"step": 341
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7535,
"step": 342
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8074,
"step": 343
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7776,
"step": 344
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7466,
"step": 345
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8341,
"step": 346
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.688,
"step": 347
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.819,
"step": 348
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7021,
"step": 349
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7799,
"step": 350
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8469,
"step": 351
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7183,
"step": 352
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7751,
"step": 353
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5917,
"step": 354
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9829,
"step": 355
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6522,
"step": 356
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9012,
"step": 357
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0506,
"step": 358
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7775,
"step": 359
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8593,
"step": 360
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7605,
"step": 361
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6358,
"step": 362
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7548,
"step": 363
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7528,
"step": 364
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0458,
"step": 365
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9207,
"step": 366
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6877,
"step": 367
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7077,
"step": 368
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7992,
"step": 369
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.954,
"step": 370
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8487,
"step": 371
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6797,
"step": 372
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7036,
"step": 373
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8241,
"step": 374
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6894,
"step": 375
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8354,
"step": 376
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7885,
"step": 377
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7283,
"step": 378
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7293,
"step": 379
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8178,
"step": 380
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6726,
"step": 381
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7794,
"step": 382
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8613,
"step": 383
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7072,
"step": 384
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7113,
"step": 385
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6039,
"step": 386
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9879,
"step": 387
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6862,
"step": 388
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7019,
"step": 389
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7679,
"step": 390
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8131,
"step": 391
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6174,
"step": 392
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6729,
"step": 393
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6222,
"step": 394
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6997,
"step": 395
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7967,
"step": 396
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6941,
"step": 397
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8199,
"step": 398
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6833,
"step": 399
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7067,
"step": 400
},
{
"epoch": 0.05,
"eval_loss": 0.7461190223693848,
"eval_runtime": 89.365,
"eval_samples_per_second": 11.19,
"eval_steps_per_second": 5.595,
"step": 400
},
{
"epoch": 0.05,
"mmlu_eval_accuracy": 0.5797172913046011,
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
"mmlu_eval_accuracy_anatomy": 0.42857142857142855,
"mmlu_eval_accuracy_astronomy": 0.6875,
"mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
"mmlu_eval_accuracy_clinical_knowledge": 0.6206896551724138,
"mmlu_eval_accuracy_college_biology": 0.4375,
"mmlu_eval_accuracy_college_chemistry": 0.375,
"mmlu_eval_accuracy_college_computer_science": 0.6363636363636364,
"mmlu_eval_accuracy_college_mathematics": 0.36363636363636365,
"mmlu_eval_accuracy_college_medicine": 0.5909090909090909,
"mmlu_eval_accuracy_college_physics": 0.45454545454545453,
"mmlu_eval_accuracy_computer_security": 0.5454545454545454,
"mmlu_eval_accuracy_conceptual_physics": 0.46153846153846156,
"mmlu_eval_accuracy_econometrics": 0.5,
"mmlu_eval_accuracy_electrical_engineering": 0.6875,
"mmlu_eval_accuracy_elementary_mathematics": 0.36585365853658536,
"mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
"mmlu_eval_accuracy_global_facts": 0.5,
"mmlu_eval_accuracy_high_school_biology": 0.5,
"mmlu_eval_accuracy_high_school_chemistry": 0.2727272727272727,
"mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
"mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.5581395348837209,
"mmlu_eval_accuracy_high_school_mathematics": 0.4827586206896552,
"mmlu_eval_accuracy_high_school_microeconomics": 0.5769230769230769,
"mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
"mmlu_eval_accuracy_high_school_psychology": 0.7833333333333333,
"mmlu_eval_accuracy_high_school_statistics": 0.43478260869565216,
"mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
"mmlu_eval_accuracy_high_school_world_history": 0.7307692307692307,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.5,
"mmlu_eval_accuracy_international_law": 0.8461538461538461,
"mmlu_eval_accuracy_jurisprudence": 0.6363636363636364,
"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": 0.5789473684210527,
"mmlu_eval_accuracy_moral_scenarios": 0.3,
"mmlu_eval_accuracy_nutrition": 0.5757575757575758,
"mmlu_eval_accuracy_philosophy": 0.6764705882352942,
"mmlu_eval_accuracy_prehistory": 0.45714285714285713,
"mmlu_eval_accuracy_professional_accounting": 0.6129032258064516,
"mmlu_eval_accuracy_professional_law": 0.38235294117647056,
"mmlu_eval_accuracy_professional_medicine": 0.5483870967741935,
"mmlu_eval_accuracy_professional_psychology": 0.5942028985507246,
"mmlu_eval_accuracy_public_relations": 0.4166666666666667,
"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": 0.5,
"mmlu_eval_accuracy_world_religions": 0.8947368421052632,
"mmlu_loss": 1.0156579838175686,
"step": 400
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6591,
"step": 401
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0622,
"step": 402
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6899,
"step": 403
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9544,
"step": 404
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8056,
"step": 405
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.851,
"step": 406
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7279,
"step": 407
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.722,
"step": 408
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.663,
"step": 409
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8089,
"step": 410
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7354,
"step": 411
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6057,
"step": 412
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7454,
"step": 413
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7661,
"step": 414
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.694,
"step": 415
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7702,
"step": 416
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.125,
"step": 417
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8887,
"step": 418
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8579,
"step": 419
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9822,
"step": 420
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6885,
"step": 421
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8592,
"step": 422
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7949,
"step": 423
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6083,
"step": 424
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8611,
"step": 425
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5553,
"step": 426
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6633,
"step": 427
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5986,
"step": 428
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.757,
"step": 429
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6426,
"step": 430
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6319,
"step": 431
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6958,
"step": 432
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8013,
"step": 433
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8315,
"step": 434
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7421,
"step": 435
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6711,
"step": 436
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7642,
"step": 437
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7059,
"step": 438
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8151,
"step": 439
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.909,
"step": 440
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6869,
"step": 441
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8638,
"step": 442
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6705,
"step": 443
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7413,
"step": 444
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.806,
"step": 445
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8236,
"step": 446
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8451,
"step": 447
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6083,
"step": 448
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8368,
"step": 449
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8221,
"step": 450
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8461,
"step": 451
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7972,
"step": 452
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.729,
"step": 453
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5537,
"step": 454
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6895,
"step": 455
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0495,
"step": 456
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.1062,
"step": 457
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5772,
"step": 458
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6685,
"step": 459
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7265,
"step": 460
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.791,
"step": 461
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6042,
"step": 462
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7798,
"step": 463
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6451,
"step": 464
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.956,
"step": 465
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6902,
"step": 466
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6969,
"step": 467
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6515,
"step": 468
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.61,
"step": 469
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6979,
"step": 470
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7441,
"step": 471
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7141,
"step": 472
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.65,
"step": 473
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.633,
"step": 474
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.758,
"step": 475
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6618,
"step": 476
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8628,
"step": 477
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7577,
"step": 478
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9919,
"step": 479
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7109,
"step": 480
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8242,
"step": 481
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8499,
"step": 482
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8678,
"step": 483
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8809,
"step": 484
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8299,
"step": 485
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.786,
"step": 486
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7843,
"step": 487
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.829,
"step": 488
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7041,
"step": 489
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6071,
"step": 490
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7835,
"step": 491
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6423,
"step": 492
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6988,
"step": 493
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7137,
"step": 494
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5409,
"step": 495
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7214,
"step": 496
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.562,
"step": 497
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.768,
"step": 498
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9842,
"step": 499
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7726,
"step": 500
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6769,
"step": 501
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7563,
"step": 502
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7367,
"step": 503
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5904,
"step": 504
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6535,
"step": 505
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7636,
"step": 506
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7446,
"step": 507
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7755,
"step": 508
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.803,
"step": 509
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.696,
"step": 510
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8669,
"step": 511
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7929,
"step": 512
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.776,
"step": 513
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0267,
"step": 514
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6952,
"step": 515
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7596,
"step": 516
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6995,
"step": 517
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6184,
"step": 518
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.876,
"step": 519
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7309,
"step": 520
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6508,
"step": 521
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4789,
"step": 522
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.688,
"step": 523
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9034,
"step": 524
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6676,
"step": 525
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7026,
"step": 526
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5841,
"step": 527
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6495,
"step": 528
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6895,
"step": 529
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7517,
"step": 530
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8024,
"step": 531
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6866,
"step": 532
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8517,
"step": 533
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8033,
"step": 534
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8277,
"step": 535
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7518,
"step": 536
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7194,
"step": 537
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7939,
"step": 538
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6058,
"step": 539
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7423,
"step": 540
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5718,
"step": 541
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6836,
"step": 542
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7325,
"step": 543
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0166,
"step": 544
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8838,
"step": 545
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8832,
"step": 546
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7268,
"step": 547
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.697,
"step": 548
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7966,
"step": 549
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6128,
"step": 550
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5957,
"step": 551
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8312,
"step": 552
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7346,
"step": 553
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7053,
"step": 554
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7295,
"step": 555
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6156,
"step": 556
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7919,
"step": 557
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5621,
"step": 558
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.88,
"step": 559
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7268,
"step": 560
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6599,
"step": 561
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7941,
"step": 562
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6645,
"step": 563
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7271,
"step": 564
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8323,
"step": 565
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6958,
"step": 566
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.891,
"step": 567
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7096,
"step": 568
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7868,
"step": 569
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7476,
"step": 570
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6828,
"step": 571
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7018,
"step": 572
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6881,
"step": 573
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6763,
"step": 574
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9433,
"step": 575
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0527,
"step": 576
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6141,
"step": 577
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9813,
"step": 578
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7673,
"step": 579
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8586,
"step": 580
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7057,
"step": 581
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6039,
"step": 582
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7464,
"step": 583
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7213,
"step": 584
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7082,
"step": 585
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8468,
"step": 586
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8355,
"step": 587
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0266,
"step": 588
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7383,
"step": 589
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6909,
"step": 590
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6144,
"step": 591
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0297,
"step": 592
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6855,
"step": 593
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7546,
"step": 594
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.595,
"step": 595
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7289,
"step": 596
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8997,
"step": 597
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7764,
"step": 598
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9256,
"step": 599
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8644,
"step": 600
},
{
"epoch": 0.08,
"eval_loss": 0.743760883808136,
"eval_runtime": 89.3521,
"eval_samples_per_second": 11.192,
"eval_steps_per_second": 5.596,
"step": 600
},
{
"epoch": 0.08,
"mmlu_eval_accuracy": 0.565071697525313,
"mmlu_eval_accuracy_abstract_algebra": 0.18181818181818182,
"mmlu_eval_accuracy_anatomy": 0.5714285714285714,
"mmlu_eval_accuracy_astronomy": 0.625,
"mmlu_eval_accuracy_business_ethics": 0.45454545454545453,
"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": 0.36363636363636365,
"mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
"mmlu_eval_accuracy_college_medicine": 0.5909090909090909,
"mmlu_eval_accuracy_college_physics": 0.45454545454545453,
"mmlu_eval_accuracy_computer_security": 0.5454545454545454,
"mmlu_eval_accuracy_conceptual_physics": 0.5,
"mmlu_eval_accuracy_econometrics": 0.5833333333333334,
"mmlu_eval_accuracy_electrical_engineering": 0.4375,
"mmlu_eval_accuracy_elementary_mathematics": 0.36585365853658536,
"mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
"mmlu_eval_accuracy_global_facts": 0.4,
"mmlu_eval_accuracy_high_school_biology": 0.5,
"mmlu_eval_accuracy_high_school_chemistry": 0.36363636363636365,
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
"mmlu_eval_accuracy_high_school_european_history": 0.7222222222222222,
"mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.5581395348837209,
"mmlu_eval_accuracy_high_school_mathematics": 0.3448275862068966,
"mmlu_eval_accuracy_high_school_microeconomics": 0.5384615384615384,
"mmlu_eval_accuracy_high_school_physics": 0.058823529411764705,
"mmlu_eval_accuracy_high_school_psychology": 0.7833333333333333,
"mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
"mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
"mmlu_eval_accuracy_high_school_world_history": 0.7307692307692307,
"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": 0.5454545454545454,
"mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
"mmlu_eval_accuracy_machine_learning": 0.45454545454545453,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.92,
"mmlu_eval_accuracy_medical_genetics": 1.0,
"mmlu_eval_accuracy_miscellaneous": 0.7674418604651163,
"mmlu_eval_accuracy_moral_disputes": 0.5,
"mmlu_eval_accuracy_moral_scenarios": 0.3,
"mmlu_eval_accuracy_nutrition": 0.7272727272727273,
"mmlu_eval_accuracy_philosophy": 0.6470588235294118,
"mmlu_eval_accuracy_prehistory": 0.5142857142857142,
"mmlu_eval_accuracy_professional_accounting": 0.6129032258064516,
"mmlu_eval_accuracy_professional_law": 0.4,
"mmlu_eval_accuracy_professional_medicine": 0.5806451612903226,
"mmlu_eval_accuracy_professional_psychology": 0.5942028985507246,
"mmlu_eval_accuracy_public_relations": 0.5833333333333334,
"mmlu_eval_accuracy_security_studies": 0.6296296296296297,
"mmlu_eval_accuracy_sociology": 0.8636363636363636,
"mmlu_eval_accuracy_us_foreign_policy": 0.7272727272727273,
"mmlu_eval_accuracy_virology": 0.5,
"mmlu_eval_accuracy_world_religions": 0.8421052631578947,
"mmlu_loss": 1.2181325036614121,
"step": 600
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7094,
"step": 601
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7145,
"step": 602
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5794,
"step": 603
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7724,
"step": 604
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5489,
"step": 605
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.715,
"step": 606
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7017,
"step": 607
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8113,
"step": 608
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8428,
"step": 609
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8791,
"step": 610
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.624,
"step": 611
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8173,
"step": 612
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7313,
"step": 613
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6631,
"step": 614
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6923,
"step": 615
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6735,
"step": 616
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6709,
"step": 617
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6724,
"step": 618
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8058,
"step": 619
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6962,
"step": 620
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6929,
"step": 621
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6734,
"step": 622
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.881,
"step": 623
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8587,
"step": 624
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7214,
"step": 625
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8604,
"step": 626
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9142,
"step": 627
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6497,
"step": 628
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8066,
"step": 629
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.608,
"step": 630
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7822,
"step": 631
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6428,
"step": 632
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.2876,
"step": 633
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6049,
"step": 634
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5934,
"step": 635
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8383,
"step": 636
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7019,
"step": 637
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8661,
"step": 638
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7072,
"step": 639
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6863,
"step": 640
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8906,
"step": 641
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7297,
"step": 642
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8378,
"step": 643
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7096,
"step": 644
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7277,
"step": 645
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9819,
"step": 646
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6086,
"step": 647
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8803,
"step": 648
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5869,
"step": 649
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7977,
"step": 650
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7327,
"step": 651
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6776,
"step": 652
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6132,
"step": 653
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8032,
"step": 654
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5353,
"step": 655
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6708,
"step": 656
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9608,
"step": 657
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.022,
"step": 658
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6821,
"step": 659
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9153,
"step": 660
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8644,
"step": 661
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8628,
"step": 662
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8642,
"step": 663
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9579,
"step": 664
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8579,
"step": 665
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7265,
"step": 666
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7212,
"step": 667
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6839,
"step": 668
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5838,
"step": 669
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7268,
"step": 670
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6842,
"step": 671
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6154,
"step": 672
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8246,
"step": 673
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6787,
"step": 674
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7345,
"step": 675
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6068,
"step": 676
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8426,
"step": 677
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8661,
"step": 678
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7505,
"step": 679
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8469,
"step": 680
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7919,
"step": 681
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7789,
"step": 682
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6588,
"step": 683
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0344,
"step": 684
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7221,
"step": 685
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7382,
"step": 686
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0287,
"step": 687
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6518,
"step": 688
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8762,
"step": 689
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6829,
"step": 690
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5576,
"step": 691
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7016,
"step": 692
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.888,
"step": 693
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7887,
"step": 694
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8151,
"step": 695
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8987,
"step": 696
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7141,
"step": 697
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6386,
"step": 698
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8168,
"step": 699
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7109,
"step": 700
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8059,
"step": 701
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4836,
"step": 702
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6875,
"step": 703
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7136,
"step": 704
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8191,
"step": 705
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6994,
"step": 706
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7418,
"step": 707
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6842,
"step": 708
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0262,
"step": 709
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7728,
"step": 710
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7462,
"step": 711
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7572,
"step": 712
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6278,
"step": 713
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5963,
"step": 714
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9349,
"step": 715
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7012,
"step": 716
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0869,
"step": 717
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7541,
"step": 718
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6106,
"step": 719
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.544,
"step": 720
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7188,
"step": 721
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7198,
"step": 722
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7814,
"step": 723
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.841,
"step": 724
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6318,
"step": 725
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6834,
"step": 726
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7196,
"step": 727
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5788,
"step": 728
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7196,
"step": 729
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5955,
"step": 730
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8073,
"step": 731
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8742,
"step": 732
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6735,
"step": 733
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5954,
"step": 734
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.77,
"step": 735
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7422,
"step": 736
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8056,
"step": 737
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6206,
"step": 738
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6757,
"step": 739
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6844,
"step": 740
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6964,
"step": 741
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7672,
"step": 742
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6398,
"step": 743
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7924,
"step": 744
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6799,
"step": 745
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6193,
"step": 746
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7776,
"step": 747
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7106,
"step": 748
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6651,
"step": 749
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9017,
"step": 750
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.796,
"step": 751
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6757,
"step": 752
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6913,
"step": 753
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0834,
"step": 754
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.633,
"step": 755
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.617,
"step": 756
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7211,
"step": 757
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8172,
"step": 758
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5534,
"step": 759
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8104,
"step": 760
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.101,
"step": 761
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5409,
"step": 762
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7482,
"step": 763
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7524,
"step": 764
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6921,
"step": 765
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.953,
"step": 766
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6443,
"step": 767
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7384,
"step": 768
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.771,
"step": 769
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6294,
"step": 770
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8074,
"step": 771
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6411,
"step": 772
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7485,
"step": 773
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6544,
"step": 774
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.731,
"step": 775
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6775,
"step": 776
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0762,
"step": 777
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6735,
"step": 778
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7733,
"step": 779
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7566,
"step": 780
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8807,
"step": 781
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7086,
"step": 782
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9133,
"step": 783
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.512,
"step": 784
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7931,
"step": 785
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7331,
"step": 786
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5615,
"step": 787
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6168,
"step": 788
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8189,
"step": 789
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8223,
"step": 790
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7372,
"step": 791
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7781,
"step": 792
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7171,
"step": 793
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.648,
"step": 794
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8429,
"step": 795
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6393,
"step": 796
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7588,
"step": 797
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.721,
"step": 798
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7656,
"step": 799
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7229,
"step": 800
},
{
"epoch": 0.11,
"eval_loss": 0.7479018568992615,
"eval_runtime": 89.3728,
"eval_samples_per_second": 11.189,
"eval_steps_per_second": 5.595,
"step": 800
},
{
"epoch": 0.11,
"mmlu_eval_accuracy": 0.5987990559511077,
"mmlu_eval_accuracy_abstract_algebra": 0.45454545454545453,
"mmlu_eval_accuracy_anatomy": 0.5714285714285714,
"mmlu_eval_accuracy_astronomy": 0.75,
"mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
"mmlu_eval_accuracy_clinical_knowledge": 0.5862068965517241,
"mmlu_eval_accuracy_college_biology": 0.625,
"mmlu_eval_accuracy_college_chemistry": 0.375,
"mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
"mmlu_eval_accuracy_college_mathematics": 0.6363636363636364,
"mmlu_eval_accuracy_college_medicine": 0.5909090909090909,
"mmlu_eval_accuracy_college_physics": 0.5454545454545454,
"mmlu_eval_accuracy_computer_security": 0.45454545454545453,
"mmlu_eval_accuracy_conceptual_physics": 0.5,
"mmlu_eval_accuracy_econometrics": 0.5,
"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": 0.53125,
"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": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.627906976744186,
"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": 0.85,
"mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
"mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
"mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.5,
"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": 0.8181818181818182,
"mmlu_eval_accuracy_marketing": 0.84,
"mmlu_eval_accuracy_medical_genetics": 1.0,
"mmlu_eval_accuracy_miscellaneous": 0.7325581395348837,
"mmlu_eval_accuracy_moral_disputes": 0.5,
"mmlu_eval_accuracy_moral_scenarios": 0.39,
"mmlu_eval_accuracy_nutrition": 0.7575757575757576,
"mmlu_eval_accuracy_philosophy": 0.6470588235294118,
"mmlu_eval_accuracy_prehistory": 0.5714285714285714,
"mmlu_eval_accuracy_professional_accounting": 0.6129032258064516,
"mmlu_eval_accuracy_professional_law": 0.3764705882352941,
"mmlu_eval_accuracy_professional_medicine": 0.6451612903225806,
"mmlu_eval_accuracy_professional_psychology": 0.5942028985507246,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.5925925925925926,
"mmlu_eval_accuracy_sociology": 0.8181818181818182,
"mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
"mmlu_eval_accuracy_virology": 0.5555555555555556,
"mmlu_eval_accuracy_world_religions": 0.8421052631578947,
"mmlu_loss": 1.0434870222880075,
"step": 800
}
],
"logging_steps": 1,
"max_steps": 22404,
"num_train_epochs": 3,
"save_steps": 200,
"total_flos": 8.872464658425446e+16,
"trial_name": null,
"trial_params": null
}