mistralic-expert-29 / checkpoint-800 /trainer_state.json
pharaouk's picture
Training in progress, step 800, checkpoint
35f73f8
raw
history blame
99.9 kB
{
"best_metric": 0.681043803691864,
"best_model_checkpoint": "experts/mistralic-expert-29/checkpoint-600",
"epoch": 0.20199469763918698,
"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": 0.8281,
"step": 1
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6834,
"step": 2
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6795,
"step": 3
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5884,
"step": 4
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.836,
"step": 5
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7988,
"step": 6
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.531,
"step": 7
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6883,
"step": 8
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7717,
"step": 9
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7416,
"step": 10
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5991,
"step": 11
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6144,
"step": 12
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5785,
"step": 13
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7473,
"step": 14
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6916,
"step": 15
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9093,
"step": 16
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9202,
"step": 17
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5683,
"step": 18
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6056,
"step": 19
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8613,
"step": 20
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7247,
"step": 21
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5612,
"step": 22
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6764,
"step": 23
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8115,
"step": 24
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5642,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5383,
"step": 26
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5503,
"step": 27
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6636,
"step": 28
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7039,
"step": 29
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7322,
"step": 30
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6291,
"step": 31
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6892,
"step": 32
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8208,
"step": 33
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6776,
"step": 34
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5404,
"step": 35
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6605,
"step": 36
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8808,
"step": 37
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7159,
"step": 38
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6436,
"step": 39
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5478,
"step": 40
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.879,
"step": 41
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8325,
"step": 42
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7474,
"step": 43
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6632,
"step": 44
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6072,
"step": 45
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7562,
"step": 46
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6429,
"step": 47
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6926,
"step": 48
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7064,
"step": 49
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7125,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6166,
"step": 51
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5655,
"step": 52
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7084,
"step": 53
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6504,
"step": 54
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6086,
"step": 55
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8146,
"step": 56
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.706,
"step": 57
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6723,
"step": 58
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4581,
"step": 59
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7367,
"step": 60
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7673,
"step": 61
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.513,
"step": 62
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7892,
"step": 63
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.662,
"step": 64
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.656,
"step": 65
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.559,
"step": 66
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6614,
"step": 67
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7893,
"step": 68
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7972,
"step": 69
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8093,
"step": 70
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8171,
"step": 71
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6491,
"step": 72
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6915,
"step": 73
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5994,
"step": 74
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5994,
"step": 75
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.608,
"step": 76
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6353,
"step": 77
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6389,
"step": 78
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.539,
"step": 79
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6736,
"step": 80
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7245,
"step": 81
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7079,
"step": 82
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7158,
"step": 83
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5509,
"step": 84
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.688,
"step": 85
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9429,
"step": 86
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7074,
"step": 87
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7848,
"step": 88
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7461,
"step": 89
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7644,
"step": 90
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.893,
"step": 91
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9111,
"step": 92
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.729,
"step": 93
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6529,
"step": 94
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9489,
"step": 95
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6918,
"step": 96
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5863,
"step": 97
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6953,
"step": 98
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6269,
"step": 99
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5336,
"step": 100
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6729,
"step": 101
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.57,
"step": 102
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7521,
"step": 103
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6486,
"step": 104
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6212,
"step": 105
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8345,
"step": 106
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6985,
"step": 107
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6532,
"step": 108
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7224,
"step": 109
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6173,
"step": 110
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8925,
"step": 111
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6129,
"step": 112
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6461,
"step": 113
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5614,
"step": 114
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7075,
"step": 115
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5555,
"step": 116
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6579,
"step": 117
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7032,
"step": 118
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7109,
"step": 119
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.631,
"step": 120
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5974,
"step": 121
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8038,
"step": 122
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6748,
"step": 123
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8089,
"step": 124
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5831,
"step": 125
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7566,
"step": 126
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6766,
"step": 127
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7793,
"step": 128
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.638,
"step": 129
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6085,
"step": 130
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6366,
"step": 131
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.918,
"step": 132
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6851,
"step": 133
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6124,
"step": 134
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7642,
"step": 135
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.642,
"step": 136
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9157,
"step": 137
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6619,
"step": 138
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8664,
"step": 139
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6881,
"step": 140
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6909,
"step": 141
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7242,
"step": 142
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.635,
"step": 143
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8164,
"step": 144
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6169,
"step": 145
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5874,
"step": 146
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6332,
"step": 147
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8494,
"step": 148
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.834,
"step": 149
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8598,
"step": 150
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5545,
"step": 151
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9171,
"step": 152
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6941,
"step": 153
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8018,
"step": 154
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7894,
"step": 155
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5925,
"step": 156
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6051,
"step": 157
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7172,
"step": 158
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7902,
"step": 159
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8397,
"step": 160
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7131,
"step": 161
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7729,
"step": 162
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.604,
"step": 163
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.682,
"step": 164
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.641,
"step": 165
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7199,
"step": 166
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5337,
"step": 167
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8882,
"step": 168
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7981,
"step": 169
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7438,
"step": 170
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.586,
"step": 171
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8469,
"step": 172
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7154,
"step": 173
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6153,
"step": 174
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5517,
"step": 175
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.951,
"step": 176
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6128,
"step": 177
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6679,
"step": 178
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6546,
"step": 179
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6633,
"step": 180
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7509,
"step": 181
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6667,
"step": 182
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6744,
"step": 183
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8333,
"step": 184
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9516,
"step": 185
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5293,
"step": 186
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5694,
"step": 187
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7939,
"step": 188
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6257,
"step": 189
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6859,
"step": 190
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7746,
"step": 191
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5474,
"step": 192
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6079,
"step": 193
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6545,
"step": 194
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7829,
"step": 195
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6969,
"step": 196
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.713,
"step": 197
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8589,
"step": 198
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6964,
"step": 199
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6997,
"step": 200
},
{
"epoch": 0.05,
"eval_loss": 0.6861502528190613,
"eval_runtime": 95.1356,
"eval_samples_per_second": 10.511,
"eval_steps_per_second": 5.256,
"step": 200
},
{
"epoch": 0.05,
"mmlu_eval_accuracy": 0.5869545490402098,
"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": 0.6206896551724138,
"mmlu_eval_accuracy_college_biology": 0.625,
"mmlu_eval_accuracy_college_chemistry": 0.25,
"mmlu_eval_accuracy_college_computer_science": 0.18181818181818182,
"mmlu_eval_accuracy_college_mathematics": 0.5454545454545454,
"mmlu_eval_accuracy_college_medicine": 0.5909090909090909,
"mmlu_eval_accuracy_college_physics": 0.5454545454545454,
"mmlu_eval_accuracy_computer_security": 0.6363636363636364,
"mmlu_eval_accuracy_conceptual_physics": 0.46153846153846156,
"mmlu_eval_accuracy_econometrics": 0.5,
"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": 0.4,
"mmlu_eval_accuracy_high_school_biology": 0.59375,
"mmlu_eval_accuracy_high_school_chemistry": 0.45454545454545453,
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
"mmlu_eval_accuracy_high_school_european_history": 0.7222222222222222,
"mmlu_eval_accuracy_high_school_geography": 0.9090909090909091,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.6511627906976745,
"mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6153846153846154,
"mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
"mmlu_eval_accuracy_high_school_psychology": 0.8333333333333334,
"mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
"mmlu_eval_accuracy_high_school_us_history": 0.6818181818181818,
"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": 1.0,
"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": 0.92,
"mmlu_eval_accuracy_medical_genetics": 1.0,
"mmlu_eval_accuracy_miscellaneous": 0.7441860465116279,
"mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
"mmlu_eval_accuracy_moral_scenarios": 0.38,
"mmlu_eval_accuracy_nutrition": 0.7272727272727273,
"mmlu_eval_accuracy_philosophy": 0.7058823529411765,
"mmlu_eval_accuracy_prehistory": 0.4857142857142857,
"mmlu_eval_accuracy_professional_accounting": 0.5483870967741935,
"mmlu_eval_accuracy_professional_law": 0.40588235294117647,
"mmlu_eval_accuracy_professional_medicine": 0.6129032258064516,
"mmlu_eval_accuracy_professional_psychology": 0.6231884057971014,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.6296296296296297,
"mmlu_eval_accuracy_sociology": 0.8181818181818182,
"mmlu_eval_accuracy_us_foreign_policy": 0.7272727272727273,
"mmlu_eval_accuracy_virology": 0.4444444444444444,
"mmlu_eval_accuracy_world_religions": 0.7894736842105263,
"mmlu_loss": 1.8230019673977447,
"step": 200
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6604,
"step": 201
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8396,
"step": 202
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7984,
"step": 203
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6176,
"step": 204
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5603,
"step": 205
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6735,
"step": 206
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5549,
"step": 207
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6017,
"step": 208
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7503,
"step": 209
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6149,
"step": 210
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.695,
"step": 211
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6514,
"step": 212
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6449,
"step": 213
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7956,
"step": 214
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5207,
"step": 215
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6516,
"step": 216
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8209,
"step": 217
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6882,
"step": 218
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6878,
"step": 219
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6819,
"step": 220
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5928,
"step": 221
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6976,
"step": 222
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5981,
"step": 223
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8225,
"step": 224
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6529,
"step": 225
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6796,
"step": 226
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5937,
"step": 227
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7849,
"step": 228
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8637,
"step": 229
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7255,
"step": 230
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7026,
"step": 231
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.72,
"step": 232
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7608,
"step": 233
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6111,
"step": 234
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6461,
"step": 235
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7821,
"step": 236
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6617,
"step": 237
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.703,
"step": 238
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6998,
"step": 239
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6882,
"step": 240
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6719,
"step": 241
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7673,
"step": 242
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6299,
"step": 243
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7046,
"step": 244
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.648,
"step": 245
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8277,
"step": 246
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6474,
"step": 247
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6895,
"step": 248
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7114,
"step": 249
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5852,
"step": 250
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.732,
"step": 251
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7306,
"step": 252
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8135,
"step": 253
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6564,
"step": 254
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.837,
"step": 255
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6664,
"step": 256
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6575,
"step": 257
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8251,
"step": 258
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8432,
"step": 259
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.667,
"step": 260
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6629,
"step": 261
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6662,
"step": 262
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7222,
"step": 263
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7911,
"step": 264
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5994,
"step": 265
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7135,
"step": 266
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8315,
"step": 267
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5882,
"step": 268
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7202,
"step": 269
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.561,
"step": 270
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7236,
"step": 271
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6975,
"step": 272
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7764,
"step": 273
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7473,
"step": 274
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6756,
"step": 275
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6694,
"step": 276
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7919,
"step": 277
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6832,
"step": 278
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6489,
"step": 279
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8486,
"step": 280
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.832,
"step": 281
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6387,
"step": 282
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.592,
"step": 283
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6453,
"step": 284
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.751,
"step": 285
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6806,
"step": 286
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7266,
"step": 287
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8536,
"step": 288
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6573,
"step": 289
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6346,
"step": 290
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8819,
"step": 291
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6592,
"step": 292
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6427,
"step": 293
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7104,
"step": 294
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6236,
"step": 295
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5712,
"step": 296
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6399,
"step": 297
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8104,
"step": 298
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7844,
"step": 299
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.887,
"step": 300
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7379,
"step": 301
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5748,
"step": 302
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7696,
"step": 303
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6378,
"step": 304
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5088,
"step": 305
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6444,
"step": 306
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5152,
"step": 307
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7381,
"step": 308
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7065,
"step": 309
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5301,
"step": 310
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6852,
"step": 311
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6646,
"step": 312
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6507,
"step": 313
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7246,
"step": 314
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6344,
"step": 315
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6504,
"step": 316
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6365,
"step": 317
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6636,
"step": 318
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6685,
"step": 319
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7406,
"step": 320
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.662,
"step": 321
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6903,
"step": 322
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7694,
"step": 323
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6293,
"step": 324
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7103,
"step": 325
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8745,
"step": 326
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4777,
"step": 327
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6935,
"step": 328
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6813,
"step": 329
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6788,
"step": 330
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6835,
"step": 331
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7498,
"step": 332
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5958,
"step": 333
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5378,
"step": 334
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6314,
"step": 335
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6326,
"step": 336
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7197,
"step": 337
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6465,
"step": 338
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6166,
"step": 339
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6901,
"step": 340
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6614,
"step": 341
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7126,
"step": 342
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7241,
"step": 343
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7211,
"step": 344
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8427,
"step": 345
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8317,
"step": 346
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4921,
"step": 347
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.602,
"step": 348
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7526,
"step": 349
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7541,
"step": 350
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.702,
"step": 351
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6369,
"step": 352
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.71,
"step": 353
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7067,
"step": 354
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7172,
"step": 355
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7921,
"step": 356
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7258,
"step": 357
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8355,
"step": 358
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7401,
"step": 359
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8881,
"step": 360
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7246,
"step": 361
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6681,
"step": 362
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5973,
"step": 363
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5858,
"step": 364
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7266,
"step": 365
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7617,
"step": 366
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8634,
"step": 367
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8171,
"step": 368
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6914,
"step": 369
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6371,
"step": 370
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7257,
"step": 371
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7238,
"step": 372
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6262,
"step": 373
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7033,
"step": 374
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7395,
"step": 375
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7843,
"step": 376
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6959,
"step": 377
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9614,
"step": 378
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7035,
"step": 379
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8038,
"step": 380
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5506,
"step": 381
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8534,
"step": 382
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7633,
"step": 383
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7036,
"step": 384
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7258,
"step": 385
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5646,
"step": 386
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5988,
"step": 387
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6628,
"step": 388
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6735,
"step": 389
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6903,
"step": 390
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7059,
"step": 391
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6156,
"step": 392
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6621,
"step": 393
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5314,
"step": 394
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7155,
"step": 395
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8404,
"step": 396
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7622,
"step": 397
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7688,
"step": 398
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.648,
"step": 399
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.488,
"step": 400
},
{
"epoch": 0.1,
"eval_loss": 0.6824774146080017,
"eval_runtime": 95.1468,
"eval_samples_per_second": 10.51,
"eval_steps_per_second": 5.255,
"step": 400
},
{
"epoch": 0.1,
"mmlu_eval_accuracy": 0.5881224648885633,
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
"mmlu_eval_accuracy_anatomy": 0.5,
"mmlu_eval_accuracy_astronomy": 0.8125,
"mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
"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": 0.2727272727272727,
"mmlu_eval_accuracy_college_mathematics": 0.5454545454545454,
"mmlu_eval_accuracy_college_medicine": 0.6363636363636364,
"mmlu_eval_accuracy_college_physics": 0.45454545454545453,
"mmlu_eval_accuracy_computer_security": 0.7272727272727273,
"mmlu_eval_accuracy_conceptual_physics": 0.5,
"mmlu_eval_accuracy_econometrics": 0.5,
"mmlu_eval_accuracy_electrical_engineering": 0.625,
"mmlu_eval_accuracy_elementary_mathematics": 0.3902439024390244,
"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": 0.4090909090909091,
"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": 0.7142857142857143,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.627906976744186,
"mmlu_eval_accuracy_high_school_mathematics": 0.20689655172413793,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6538461538461539,
"mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
"mmlu_eval_accuracy_high_school_psychology": 0.8833333333333333,
"mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
"mmlu_eval_accuracy_high_school_us_history": 0.6363636363636364,
"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": 1.0,
"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": 1.0,
"mmlu_eval_accuracy_miscellaneous": 0.7441860465116279,
"mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
"mmlu_eval_accuracy_moral_scenarios": 0.37,
"mmlu_eval_accuracy_nutrition": 0.696969696969697,
"mmlu_eval_accuracy_philosophy": 0.7352941176470589,
"mmlu_eval_accuracy_prehistory": 0.5714285714285714,
"mmlu_eval_accuracy_professional_accounting": 0.5483870967741935,
"mmlu_eval_accuracy_professional_law": 0.4117647058823529,
"mmlu_eval_accuracy_professional_medicine": 0.6451612903225806,
"mmlu_eval_accuracy_professional_psychology": 0.6376811594202898,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.6296296296296297,
"mmlu_eval_accuracy_sociology": 0.7727272727272727,
"mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
"mmlu_eval_accuracy_virology": 0.4444444444444444,
"mmlu_eval_accuracy_world_religions": 0.7894736842105263,
"mmlu_loss": 1.4297099713012382,
"step": 400
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6895,
"step": 401
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5892,
"step": 402
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5107,
"step": 403
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6283,
"step": 404
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.562,
"step": 405
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7264,
"step": 406
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6245,
"step": 407
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.696,
"step": 408
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.749,
"step": 409
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7172,
"step": 410
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6991,
"step": 411
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8584,
"step": 412
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6207,
"step": 413
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5251,
"step": 414
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5696,
"step": 415
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.664,
"step": 416
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6096,
"step": 417
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6649,
"step": 418
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7113,
"step": 419
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5898,
"step": 420
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7025,
"step": 421
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4703,
"step": 422
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5672,
"step": 423
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.834,
"step": 424
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6655,
"step": 425
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6936,
"step": 426
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7467,
"step": 427
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5863,
"step": 428
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.919,
"step": 429
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7066,
"step": 430
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6046,
"step": 431
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6821,
"step": 432
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6478,
"step": 433
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.604,
"step": 434
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6641,
"step": 435
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6329,
"step": 436
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6918,
"step": 437
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6216,
"step": 438
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7208,
"step": 439
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6226,
"step": 440
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6756,
"step": 441
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5551,
"step": 442
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.661,
"step": 443
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6459,
"step": 444
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5447,
"step": 445
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8029,
"step": 446
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5886,
"step": 447
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6214,
"step": 448
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7573,
"step": 449
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4937,
"step": 450
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6315,
"step": 451
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6872,
"step": 452
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6719,
"step": 453
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6434,
"step": 454
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.882,
"step": 455
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7921,
"step": 456
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.657,
"step": 457
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.613,
"step": 458
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.9572,
"step": 459
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.753,
"step": 460
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.562,
"step": 461
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7242,
"step": 462
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5071,
"step": 463
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6417,
"step": 464
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7622,
"step": 465
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6064,
"step": 466
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6344,
"step": 467
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8129,
"step": 468
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5671,
"step": 469
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6653,
"step": 470
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8753,
"step": 471
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7669,
"step": 472
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7124,
"step": 473
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5637,
"step": 474
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6063,
"step": 475
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8913,
"step": 476
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7081,
"step": 477
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6921,
"step": 478
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6802,
"step": 479
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7924,
"step": 480
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.663,
"step": 481
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8256,
"step": 482
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8907,
"step": 483
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6274,
"step": 484
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6014,
"step": 485
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7409,
"step": 486
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6689,
"step": 487
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8057,
"step": 488
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.714,
"step": 489
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7829,
"step": 490
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7454,
"step": 491
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6404,
"step": 492
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6949,
"step": 493
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7472,
"step": 494
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6409,
"step": 495
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7146,
"step": 496
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7208,
"step": 497
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6796,
"step": 498
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6793,
"step": 499
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8412,
"step": 500
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.714,
"step": 501
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6695,
"step": 502
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7764,
"step": 503
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5901,
"step": 504
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7264,
"step": 505
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4611,
"step": 506
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6388,
"step": 507
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7984,
"step": 508
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5643,
"step": 509
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8223,
"step": 510
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.653,
"step": 511
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8065,
"step": 512
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8131,
"step": 513
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6669,
"step": 514
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6538,
"step": 515
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6628,
"step": 516
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7123,
"step": 517
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7038,
"step": 518
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7661,
"step": 519
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7713,
"step": 520
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6057,
"step": 521
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7759,
"step": 522
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.752,
"step": 523
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6083,
"step": 524
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.564,
"step": 525
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6842,
"step": 526
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8257,
"step": 527
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.668,
"step": 528
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8132,
"step": 529
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.691,
"step": 530
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7447,
"step": 531
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8735,
"step": 532
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8874,
"step": 533
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6187,
"step": 534
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6091,
"step": 535
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7805,
"step": 536
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7125,
"step": 537
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7744,
"step": 538
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7623,
"step": 539
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.616,
"step": 540
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5048,
"step": 541
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6342,
"step": 542
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6586,
"step": 543
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7099,
"step": 544
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7082,
"step": 545
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6549,
"step": 546
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6562,
"step": 547
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6766,
"step": 548
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.8085,
"step": 549
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5737,
"step": 550
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.8658,
"step": 551
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6835,
"step": 552
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6107,
"step": 553
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6473,
"step": 554
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5264,
"step": 555
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5308,
"step": 556
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7845,
"step": 557
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6702,
"step": 558
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6707,
"step": 559
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.715,
"step": 560
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.8565,
"step": 561
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5826,
"step": 562
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6412,
"step": 563
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7312,
"step": 564
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6294,
"step": 565
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5465,
"step": 566
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.8667,
"step": 567
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.8091,
"step": 568
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7047,
"step": 569
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5156,
"step": 570
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6275,
"step": 571
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5292,
"step": 572
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.727,
"step": 573
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6511,
"step": 574
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5755,
"step": 575
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.7141,
"step": 576
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5442,
"step": 577
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6237,
"step": 578
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6656,
"step": 579
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.7612,
"step": 580
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5665,
"step": 581
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.706,
"step": 582
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6703,
"step": 583
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.64,
"step": 584
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5549,
"step": 585
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5237,
"step": 586
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6167,
"step": 587
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.595,
"step": 588
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.8437,
"step": 589
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.7083,
"step": 590
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.849,
"step": 591
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.7253,
"step": 592
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6806,
"step": 593
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6226,
"step": 594
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6413,
"step": 595
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.8792,
"step": 596
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5808,
"step": 597
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.8768,
"step": 598
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.8693,
"step": 599
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.8851,
"step": 600
},
{
"epoch": 0.15,
"eval_loss": 0.681043803691864,
"eval_runtime": 95.159,
"eval_samples_per_second": 10.509,
"eval_steps_per_second": 5.254,
"step": 600
},
{
"epoch": 0.15,
"mmlu_eval_accuracy": 0.5878984403883732,
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
"mmlu_eval_accuracy_anatomy": 0.5714285714285714,
"mmlu_eval_accuracy_astronomy": 0.8125,
"mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
"mmlu_eval_accuracy_clinical_knowledge": 0.6551724137931034,
"mmlu_eval_accuracy_college_biology": 0.625,
"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": 0.6363636363636364,
"mmlu_eval_accuracy_college_physics": 0.36363636363636365,
"mmlu_eval_accuracy_computer_security": 0.7272727272727273,
"mmlu_eval_accuracy_conceptual_physics": 0.4230769230769231,
"mmlu_eval_accuracy_econometrics": 0.5,
"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": 0.5,
"mmlu_eval_accuracy_high_school_biology": 0.53125,
"mmlu_eval_accuracy_high_school_chemistry": 0.3181818181818182,
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
"mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
"mmlu_eval_accuracy_high_school_geography": 0.7727272727272727,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.627906976744186,
"mmlu_eval_accuracy_high_school_mathematics": 0.1724137931034483,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6153846153846154,
"mmlu_eval_accuracy_high_school_physics": 0.23529411764705882,
"mmlu_eval_accuracy_high_school_psychology": 0.8833333333333333,
"mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
"mmlu_eval_accuracy_high_school_us_history": 0.6818181818181818,
"mmlu_eval_accuracy_high_school_world_history": 0.7692307692307693,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.5,
"mmlu_eval_accuracy_international_law": 0.9230769230769231,
"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": 0.84,
"mmlu_eval_accuracy_medical_genetics": 1.0,
"mmlu_eval_accuracy_miscellaneous": 0.7441860465116279,
"mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
"mmlu_eval_accuracy_moral_scenarios": 0.34,
"mmlu_eval_accuracy_nutrition": 0.6666666666666666,
"mmlu_eval_accuracy_philosophy": 0.7352941176470589,
"mmlu_eval_accuracy_prehistory": 0.5142857142857142,
"mmlu_eval_accuracy_professional_accounting": 0.5483870967741935,
"mmlu_eval_accuracy_professional_law": 0.4176470588235294,
"mmlu_eval_accuracy_professional_medicine": 0.6129032258064516,
"mmlu_eval_accuracy_professional_psychology": 0.6231884057971014,
"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": 0.5,
"mmlu_eval_accuracy_world_religions": 0.7894736842105263,
"mmlu_loss": 1.5582773469135593,
"step": 600
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6205,
"step": 601
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6323,
"step": 602
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6582,
"step": 603
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.7214,
"step": 604
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.7337,
"step": 605
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6967,
"step": 606
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5925,
"step": 607
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.7246,
"step": 608
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5934,
"step": 609
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.5778,
"step": 610
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.7097,
"step": 611
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6428,
"step": 612
},
{
"epoch": 0.15,
"learning_rate": 0.0002,
"loss": 0.6802,
"step": 613
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.8198,
"step": 614
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.7251,
"step": 615
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.712,
"step": 616
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.7109,
"step": 617
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6471,
"step": 618
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5935,
"step": 619
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5067,
"step": 620
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.7164,
"step": 621
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6579,
"step": 622
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5533,
"step": 623
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6435,
"step": 624
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5523,
"step": 625
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.64,
"step": 626
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5963,
"step": 627
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5808,
"step": 628
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.7357,
"step": 629
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.7685,
"step": 630
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6112,
"step": 631
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5771,
"step": 632
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6807,
"step": 633
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.7864,
"step": 634
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.7146,
"step": 635
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.652,
"step": 636
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.8465,
"step": 637
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5468,
"step": 638
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.7368,
"step": 639
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6861,
"step": 640
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.6096,
"step": 641
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.876,
"step": 642
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.7318,
"step": 643
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.7272,
"step": 644
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.7233,
"step": 645
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5766,
"step": 646
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.8112,
"step": 647
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.509,
"step": 648
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.662,
"step": 649
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.5192,
"step": 650
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.8249,
"step": 651
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.8286,
"step": 652
},
{
"epoch": 0.16,
"learning_rate": 0.0002,
"loss": 0.726,
"step": 653
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.556,
"step": 654
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6986,
"step": 655
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.8957,
"step": 656
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6839,
"step": 657
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5587,
"step": 658
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6421,
"step": 659
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.7218,
"step": 660
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6266,
"step": 661
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6571,
"step": 662
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.7651,
"step": 663
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.628,
"step": 664
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.7103,
"step": 665
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6629,
"step": 666
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6109,
"step": 667
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5732,
"step": 668
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.627,
"step": 669
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.57,
"step": 670
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.589,
"step": 671
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.68,
"step": 672
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6583,
"step": 673
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6384,
"step": 674
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6129,
"step": 675
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5543,
"step": 676
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.7953,
"step": 677
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6257,
"step": 678
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5456,
"step": 679
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5136,
"step": 680
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4626,
"step": 681
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.7061,
"step": 682
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6561,
"step": 683
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.4717,
"step": 684
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.8333,
"step": 685
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5653,
"step": 686
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6158,
"step": 687
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6273,
"step": 688
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.5875,
"step": 689
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.7503,
"step": 690
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.6726,
"step": 691
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.8295,
"step": 692
},
{
"epoch": 0.17,
"learning_rate": 0.0002,
"loss": 0.7537,
"step": 693
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.7053,
"step": 694
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.7355,
"step": 695
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.7471,
"step": 696
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6294,
"step": 697
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.9196,
"step": 698
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.676,
"step": 699
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.666,
"step": 700
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6315,
"step": 701
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6036,
"step": 702
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.7621,
"step": 703
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.7786,
"step": 704
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.7091,
"step": 705
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5635,
"step": 706
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5863,
"step": 707
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.7182,
"step": 708
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6654,
"step": 709
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6554,
"step": 710
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6784,
"step": 711
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6425,
"step": 712
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6923,
"step": 713
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6026,
"step": 714
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6085,
"step": 715
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.8186,
"step": 716
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.9611,
"step": 717
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6353,
"step": 718
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6169,
"step": 719
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.8306,
"step": 720
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.7393,
"step": 721
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6107,
"step": 722
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6272,
"step": 723
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.8137,
"step": 724
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6601,
"step": 725
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.8145,
"step": 726
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.7013,
"step": 727
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.8615,
"step": 728
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5397,
"step": 729
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.5877,
"step": 730
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6517,
"step": 731
},
{
"epoch": 0.18,
"learning_rate": 0.0002,
"loss": 0.6256,
"step": 732
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.7099,
"step": 733
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5855,
"step": 734
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.8794,
"step": 735
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6913,
"step": 736
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5675,
"step": 737
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.771,
"step": 738
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5814,
"step": 739
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5479,
"step": 740
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6592,
"step": 741
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6,
"step": 742
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.714,
"step": 743
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.9334,
"step": 744
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5193,
"step": 745
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5156,
"step": 746
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.7621,
"step": 747
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.7085,
"step": 748
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.7029,
"step": 749
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.7876,
"step": 750
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6889,
"step": 751
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.7157,
"step": 752
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6886,
"step": 753
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.7936,
"step": 754
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.8475,
"step": 755
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5645,
"step": 756
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.7286,
"step": 757
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.7566,
"step": 758
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5507,
"step": 759
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.7878,
"step": 760
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6571,
"step": 761
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6373,
"step": 762
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.7095,
"step": 763
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.7905,
"step": 764
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.5663,
"step": 765
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.7229,
"step": 766
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.724,
"step": 767
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.581,
"step": 768
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.59,
"step": 769
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6692,
"step": 770
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.8525,
"step": 771
},
{
"epoch": 0.19,
"learning_rate": 0.0002,
"loss": 0.6007,
"step": 772
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5039,
"step": 773
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6594,
"step": 774
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6097,
"step": 775
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5979,
"step": 776
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6142,
"step": 777
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6798,
"step": 778
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.7565,
"step": 779
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6205,
"step": 780
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.561,
"step": 781
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6364,
"step": 782
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.7985,
"step": 783
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6478,
"step": 784
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.8059,
"step": 785
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6572,
"step": 786
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.7501,
"step": 787
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5915,
"step": 788
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.746,
"step": 789
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6507,
"step": 790
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5086,
"step": 791
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6255,
"step": 792
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6855,
"step": 793
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5474,
"step": 794
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6299,
"step": 795
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5785,
"step": 796
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.8113,
"step": 797
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.6551,
"step": 798
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5899,
"step": 799
},
{
"epoch": 0.2,
"learning_rate": 0.0002,
"loss": 0.5783,
"step": 800
},
{
"epoch": 0.2,
"eval_loss": 0.6814640164375305,
"eval_runtime": 95.15,
"eval_samples_per_second": 10.51,
"eval_steps_per_second": 5.255,
"step": 800
},
{
"epoch": 0.2,
"mmlu_eval_accuracy": 0.5718623534237226,
"mmlu_eval_accuracy_abstract_algebra": 0.18181818181818182,
"mmlu_eval_accuracy_anatomy": 0.5,
"mmlu_eval_accuracy_astronomy": 0.6875,
"mmlu_eval_accuracy_business_ethics": 0.45454545454545453,
"mmlu_eval_accuracy_clinical_knowledge": 0.6551724137931034,
"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": 0.2727272727272727,
"mmlu_eval_accuracy_college_medicine": 0.6818181818181818,
"mmlu_eval_accuracy_college_physics": 0.5454545454545454,
"mmlu_eval_accuracy_computer_security": 0.5454545454545454,
"mmlu_eval_accuracy_conceptual_physics": 0.3076923076923077,
"mmlu_eval_accuracy_econometrics": 0.5,
"mmlu_eval_accuracy_electrical_engineering": 0.5625,
"mmlu_eval_accuracy_elementary_mathematics": 0.34146341463414637,
"mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
"mmlu_eval_accuracy_global_facts": 0.5,
"mmlu_eval_accuracy_high_school_biology": 0.65625,
"mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
"mmlu_eval_accuracy_high_school_european_history": 0.7777777777777778,
"mmlu_eval_accuracy_high_school_geography": 0.7272727272727273,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.5581395348837209,
"mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6538461538461539,
"mmlu_eval_accuracy_high_school_physics": 0.23529411764705882,
"mmlu_eval_accuracy_high_school_psychology": 0.8833333333333333,
"mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
"mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
"mmlu_eval_accuracy_high_school_world_history": 0.7307692307692307,
"mmlu_eval_accuracy_human_aging": 0.6956521739130435,
"mmlu_eval_accuracy_human_sexuality": 0.5,
"mmlu_eval_accuracy_international_law": 1.0,
"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": 0.84,
"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": 0.42,
"mmlu_eval_accuracy_nutrition": 0.696969696969697,
"mmlu_eval_accuracy_philosophy": 0.7058823529411765,
"mmlu_eval_accuracy_prehistory": 0.4857142857142857,
"mmlu_eval_accuracy_professional_accounting": 0.45161290322580644,
"mmlu_eval_accuracy_professional_law": 0.4117647058823529,
"mmlu_eval_accuracy_professional_medicine": 0.5806451612903226,
"mmlu_eval_accuracy_professional_psychology": 0.5507246376811594,
"mmlu_eval_accuracy_public_relations": 0.4166666666666667,
"mmlu_eval_accuracy_security_studies": 0.5555555555555556,
"mmlu_eval_accuracy_sociology": 0.8181818181818182,
"mmlu_eval_accuracy_us_foreign_policy": 0.7272727272727273,
"mmlu_eval_accuracy_virology": 0.4444444444444444,
"mmlu_eval_accuracy_world_religions": 0.7368421052631579,
"mmlu_loss": 1.2909344176010427,
"step": 800
}
],
"logging_steps": 1,
"max_steps": 11880,
"num_train_epochs": 3,
"save_steps": 200,
"total_flos": 9.434427538912051e+16,
"trial_name": null,
"trial_params": null
}