mistralic-expert-15 / checkpoint-800 /trainer_state.json
pharaouk's picture
Training in progress, step 800, checkpoint
a2a9ef3
raw
history blame
99.9 kB
{
"best_metric": 0.7511235475540161,
"best_model_checkpoint": "experts/mistralic-expert-15/checkpoint-600",
"epoch": 0.13144382830149928,
"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": 0.8743,
"step": 1
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.0512,
"step": 2
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 1.0398,
"step": 3
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8428,
"step": 4
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8539,
"step": 5
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6822,
"step": 6
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7817,
"step": 7
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8251,
"step": 8
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.641,
"step": 9
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9213,
"step": 10
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8051,
"step": 11
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9494,
"step": 12
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7953,
"step": 13
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6879,
"step": 14
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7866,
"step": 15
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7557,
"step": 16
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6147,
"step": 17
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7999,
"step": 18
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9288,
"step": 19
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6545,
"step": 20
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6837,
"step": 21
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8564,
"step": 22
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6611,
"step": 23
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6956,
"step": 24
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5079,
"step": 25
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6142,
"step": 26
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7226,
"step": 27
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7037,
"step": 28
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8852,
"step": 29
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7209,
"step": 30
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8991,
"step": 31
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5839,
"step": 32
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6147,
"step": 33
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9075,
"step": 34
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6618,
"step": 35
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8367,
"step": 36
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7012,
"step": 37
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9249,
"step": 38
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5889,
"step": 39
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8997,
"step": 40
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7283,
"step": 41
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8314,
"step": 42
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7005,
"step": 43
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5693,
"step": 44
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8334,
"step": 45
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9775,
"step": 46
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7298,
"step": 47
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9334,
"step": 48
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7225,
"step": 49
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8088,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7466,
"step": 51
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4508,
"step": 52
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4574,
"step": 53
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6886,
"step": 54
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7615,
"step": 55
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5989,
"step": 56
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 1.0519,
"step": 57
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9321,
"step": 58
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6666,
"step": 59
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5781,
"step": 60
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9561,
"step": 61
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5952,
"step": 62
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7159,
"step": 63
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6358,
"step": 64
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.4376,
"step": 65
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6991,
"step": 66
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6598,
"step": 67
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7651,
"step": 68
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7757,
"step": 69
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7273,
"step": 70
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8523,
"step": 71
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6573,
"step": 72
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5618,
"step": 73
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6906,
"step": 74
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7202,
"step": 75
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8629,
"step": 76
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6166,
"step": 77
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8685,
"step": 78
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8057,
"step": 79
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8424,
"step": 80
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9356,
"step": 81
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8011,
"step": 82
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5423,
"step": 83
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6254,
"step": 84
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7161,
"step": 85
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6061,
"step": 86
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5743,
"step": 87
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.609,
"step": 88
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7292,
"step": 89
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.9335,
"step": 90
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.678,
"step": 91
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7919,
"step": 92
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6743,
"step": 93
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9325,
"step": 94
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6499,
"step": 95
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5631,
"step": 96
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8618,
"step": 97
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.683,
"step": 98
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.635,
"step": 99
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7615,
"step": 100
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8628,
"step": 101
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7456,
"step": 102
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8235,
"step": 103
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.916,
"step": 104
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0372,
"step": 105
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9372,
"step": 106
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6709,
"step": 107
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.1999,
"step": 108
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7315,
"step": 109
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7147,
"step": 110
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6933,
"step": 111
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.844,
"step": 112
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7172,
"step": 113
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7477,
"step": 114
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.835,
"step": 115
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.0017,
"step": 116
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5665,
"step": 117
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6992,
"step": 118
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9693,
"step": 119
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.574,
"step": 120
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8181,
"step": 121
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6321,
"step": 122
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.85,
"step": 123
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6986,
"step": 124
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7242,
"step": 125
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7909,
"step": 126
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7143,
"step": 127
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5493,
"step": 128
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7625,
"step": 129
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7199,
"step": 130
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.3807,
"step": 131
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.322,
"step": 132
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.582,
"step": 133
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.575,
"step": 134
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8918,
"step": 135
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 1.1352,
"step": 136
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9177,
"step": 137
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9293,
"step": 138
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5509,
"step": 139
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9554,
"step": 140
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5094,
"step": 141
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5929,
"step": 142
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9733,
"step": 143
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5117,
"step": 144
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6024,
"step": 145
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6401,
"step": 146
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.924,
"step": 147
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9399,
"step": 148
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.972,
"step": 149
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6407,
"step": 150
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8665,
"step": 151
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7225,
"step": 152
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6445,
"step": 153
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7775,
"step": 154
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7517,
"step": 155
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9464,
"step": 156
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5912,
"step": 157
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6161,
"step": 158
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8201,
"step": 159
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6348,
"step": 160
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0841,
"step": 161
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8405,
"step": 162
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9317,
"step": 163
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7158,
"step": 164
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9286,
"step": 165
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9918,
"step": 166
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9909,
"step": 167
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7513,
"step": 168
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8818,
"step": 169
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8644,
"step": 170
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7351,
"step": 171
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7574,
"step": 172
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6465,
"step": 173
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7679,
"step": 174
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8849,
"step": 175
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0791,
"step": 176
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8741,
"step": 177
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7822,
"step": 178
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8417,
"step": 179
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7317,
"step": 180
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0245,
"step": 181
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6746,
"step": 182
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6059,
"step": 183
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9795,
"step": 184
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9011,
"step": 185
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6384,
"step": 186
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7801,
"step": 187
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.81,
"step": 188
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8018,
"step": 189
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6842,
"step": 190
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7468,
"step": 191
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8598,
"step": 192
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9457,
"step": 193
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5614,
"step": 194
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6668,
"step": 195
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8057,
"step": 196
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6429,
"step": 197
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6506,
"step": 198
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8127,
"step": 199
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9115,
"step": 200
},
{
"epoch": 0.03,
"eval_loss": 0.7577090859413147,
"eval_runtime": 131.6982,
"eval_samples_per_second": 7.593,
"eval_steps_per_second": 3.797,
"step": 200
},
{
"epoch": 0.03,
"mmlu_eval_accuracy": 0.5965101027344765,
"mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
"mmlu_eval_accuracy_anatomy": 0.5,
"mmlu_eval_accuracy_astronomy": 0.75,
"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": 0.375,
"mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
"mmlu_eval_accuracy_college_mathematics": 0.36363636363636365,
"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": 0.5625,
"mmlu_eval_accuracy_elementary_mathematics": 0.4878048780487805,
"mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
"mmlu_eval_accuracy_global_facts": 0.3,
"mmlu_eval_accuracy_high_school_biology": 0.5625,
"mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
"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": 0.6511627906976745,
"mmlu_eval_accuracy_high_school_mathematics": 0.20689655172413793,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6153846153846154,
"mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
"mmlu_eval_accuracy_high_school_psychology": 0.8333333333333334,
"mmlu_eval_accuracy_high_school_statistics": 0.43478260869565216,
"mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
"mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
"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": 0.7222222222222222,
"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": 0.7558139534883721,
"mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
"mmlu_eval_accuracy_moral_scenarios": 0.34,
"mmlu_eval_accuracy_nutrition": 0.7272727272727273,
"mmlu_eval_accuracy_philosophy": 0.7647058823529411,
"mmlu_eval_accuracy_prehistory": 0.6,
"mmlu_eval_accuracy_professional_accounting": 0.5806451612903226,
"mmlu_eval_accuracy_professional_law": 0.4,
"mmlu_eval_accuracy_professional_medicine": 0.5806451612903226,
"mmlu_eval_accuracy_professional_psychology": 0.6086956521739131,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.6666666666666666,
"mmlu_eval_accuracy_sociology": 0.9090909090909091,
"mmlu_eval_accuracy_us_foreign_policy": 0.9090909090909091,
"mmlu_eval_accuracy_virology": 0.5,
"mmlu_eval_accuracy_world_religions": 0.8421052631578947,
"mmlu_loss": 1.1431999449708132,
"step": 200
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6858,
"step": 201
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4792,
"step": 202
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9803,
"step": 203
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9486,
"step": 204
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7161,
"step": 205
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.635,
"step": 206
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4869,
"step": 207
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6713,
"step": 208
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6963,
"step": 209
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7726,
"step": 210
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 1.0783,
"step": 211
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.9185,
"step": 212
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.812,
"step": 213
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9379,
"step": 214
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9187,
"step": 215
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6081,
"step": 216
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7071,
"step": 217
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7934,
"step": 218
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7151,
"step": 219
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8808,
"step": 220
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6993,
"step": 221
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8453,
"step": 222
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7213,
"step": 223
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6957,
"step": 224
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9296,
"step": 225
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7856,
"step": 226
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5761,
"step": 227
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8645,
"step": 228
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6516,
"step": 229
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8441,
"step": 230
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.3882,
"step": 231
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0478,
"step": 232
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8387,
"step": 233
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7188,
"step": 234
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9145,
"step": 235
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8007,
"step": 236
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8292,
"step": 237
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5326,
"step": 238
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.4012,
"step": 239
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9156,
"step": 240
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6651,
"step": 241
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.9193,
"step": 242
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7638,
"step": 243
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6298,
"step": 244
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7544,
"step": 245
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7319,
"step": 246
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.807,
"step": 247
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7104,
"step": 248
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8149,
"step": 249
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.534,
"step": 250
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7246,
"step": 251
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7361,
"step": 252
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7693,
"step": 253
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8553,
"step": 254
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6042,
"step": 255
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6617,
"step": 256
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7081,
"step": 257
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7198,
"step": 258
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8238,
"step": 259
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7848,
"step": 260
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8787,
"step": 261
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8322,
"step": 262
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0992,
"step": 263
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7046,
"step": 264
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4757,
"step": 265
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4978,
"step": 266
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8656,
"step": 267
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8534,
"step": 268
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0157,
"step": 269
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6066,
"step": 270
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8877,
"step": 271
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8856,
"step": 272
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8631,
"step": 273
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0294,
"step": 274
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.596,
"step": 275
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8029,
"step": 276
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6784,
"step": 277
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8277,
"step": 278
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9855,
"step": 279
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7293,
"step": 280
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8036,
"step": 281
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8581,
"step": 282
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6941,
"step": 283
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9493,
"step": 284
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9682,
"step": 285
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7554,
"step": 286
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8585,
"step": 287
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5798,
"step": 288
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.814,
"step": 289
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7303,
"step": 290
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7244,
"step": 291
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7552,
"step": 292
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0223,
"step": 293
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7623,
"step": 294
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7383,
"step": 295
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5089,
"step": 296
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4504,
"step": 297
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.2205,
"step": 298
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9076,
"step": 299
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9371,
"step": 300
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.1473,
"step": 301
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8726,
"step": 302
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7765,
"step": 303
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7015,
"step": 304
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7357,
"step": 305
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.1318,
"step": 306
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7922,
"step": 307
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7434,
"step": 308
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9389,
"step": 309
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0817,
"step": 310
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6492,
"step": 311
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8057,
"step": 312
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8286,
"step": 313
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7106,
"step": 314
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9558,
"step": 315
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7318,
"step": 316
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 1.0106,
"step": 317
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.923,
"step": 318
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8891,
"step": 319
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7447,
"step": 320
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7532,
"step": 321
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7922,
"step": 322
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6687,
"step": 323
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6815,
"step": 324
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9777,
"step": 325
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7946,
"step": 326
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6075,
"step": 327
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8379,
"step": 328
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6908,
"step": 329
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6723,
"step": 330
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6847,
"step": 331
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.9873,
"step": 332
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6398,
"step": 333
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6382,
"step": 334
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5199,
"step": 335
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6147,
"step": 336
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6788,
"step": 337
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7803,
"step": 338
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8861,
"step": 339
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6744,
"step": 340
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6291,
"step": 341
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8696,
"step": 342
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8301,
"step": 343
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7515,
"step": 344
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6998,
"step": 345
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6413,
"step": 346
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6048,
"step": 347
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9201,
"step": 348
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7271,
"step": 349
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8123,
"step": 350
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7396,
"step": 351
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6787,
"step": 352
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7128,
"step": 353
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7682,
"step": 354
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4681,
"step": 355
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7411,
"step": 356
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6903,
"step": 357
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.054,
"step": 358
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0388,
"step": 359
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5839,
"step": 360
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9394,
"step": 361
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8511,
"step": 362
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7703,
"step": 363
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8453,
"step": 364
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8057,
"step": 365
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7747,
"step": 366
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7674,
"step": 367
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7861,
"step": 368
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7925,
"step": 369
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.951,
"step": 370
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9278,
"step": 371
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.781,
"step": 372
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6763,
"step": 373
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5281,
"step": 374
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5816,
"step": 375
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9666,
"step": 376
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5588,
"step": 377
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.0995,
"step": 378
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 1.2138,
"step": 379
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8365,
"step": 380
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6347,
"step": 381
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7131,
"step": 382
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7783,
"step": 383
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5475,
"step": 384
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6053,
"step": 385
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9546,
"step": 386
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9732,
"step": 387
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.621,
"step": 388
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6031,
"step": 389
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6867,
"step": 390
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8615,
"step": 391
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.9171,
"step": 392
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.678,
"step": 393
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.881,
"step": 394
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.435,
"step": 395
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7515,
"step": 396
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7088,
"step": 397
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8921,
"step": 398
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7842,
"step": 399
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7468,
"step": 400
},
{
"epoch": 0.07,
"eval_loss": 0.7549907565116882,
"eval_runtime": 131.6241,
"eval_samples_per_second": 7.597,
"eval_steps_per_second": 3.799,
"step": 400
},
{
"epoch": 0.07,
"mmlu_eval_accuracy": 0.5905395038007227,
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
"mmlu_eval_accuracy_anatomy": 0.5714285714285714,
"mmlu_eval_accuracy_astronomy": 0.8125,
"mmlu_eval_accuracy_business_ethics": 0.8181818181818182,
"mmlu_eval_accuracy_clinical_knowledge": 0.4827586206896552,
"mmlu_eval_accuracy_college_biology": 0.625,
"mmlu_eval_accuracy_college_chemistry": 0.25,
"mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
"mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
"mmlu_eval_accuracy_college_medicine": 0.6363636363636364,
"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": 0.5833333333333334,
"mmlu_eval_accuracy_electrical_engineering": 0.6875,
"mmlu_eval_accuracy_elementary_mathematics": 0.43902439024390244,
"mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
"mmlu_eval_accuracy_global_facts": 0.5,
"mmlu_eval_accuracy_high_school_biology": 0.59375,
"mmlu_eval_accuracy_high_school_chemistry": 0.3181818181818182,
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
"mmlu_eval_accuracy_high_school_european_history": 0.7222222222222222,
"mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.6046511627906976,
"mmlu_eval_accuracy_high_school_mathematics": 0.3793103448275862,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6153846153846154,
"mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
"mmlu_eval_accuracy_high_school_psychology": 0.7833333333333333,
"mmlu_eval_accuracy_high_school_statistics": 0.5217391304347826,
"mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
"mmlu_eval_accuracy_high_school_world_history": 0.6153846153846154,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.5833333333333334,
"mmlu_eval_accuracy_international_law": 1.0,
"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": 0.88,
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
"mmlu_eval_accuracy_miscellaneous": 0.7674418604651163,
"mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
"mmlu_eval_accuracy_moral_scenarios": 0.34,
"mmlu_eval_accuracy_nutrition": 0.6060606060606061,
"mmlu_eval_accuracy_philosophy": 0.7352941176470589,
"mmlu_eval_accuracy_prehistory": 0.4857142857142857,
"mmlu_eval_accuracy_professional_accounting": 0.5806451612903226,
"mmlu_eval_accuracy_professional_law": 0.38235294117647056,
"mmlu_eval_accuracy_professional_medicine": 0.6774193548387096,
"mmlu_eval_accuracy_professional_psychology": 0.5652173913043478,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.5925925925925926,
"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": 0.8421052631578947,
"mmlu_loss": 1.1786961650412637,
"step": 400
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8926,
"step": 401
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8226,
"step": 402
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7972,
"step": 403
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9732,
"step": 404
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8672,
"step": 405
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.3649,
"step": 406
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9201,
"step": 407
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9357,
"step": 408
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8635,
"step": 409
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6937,
"step": 410
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.3983,
"step": 411
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7475,
"step": 412
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6109,
"step": 413
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6539,
"step": 414
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6245,
"step": 415
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8199,
"step": 416
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9231,
"step": 417
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.739,
"step": 418
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9367,
"step": 419
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.71,
"step": 420
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7856,
"step": 421
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6392,
"step": 422
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9306,
"step": 423
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7466,
"step": 424
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7127,
"step": 425
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6587,
"step": 426
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6164,
"step": 427
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7736,
"step": 428
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.016,
"step": 429
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7706,
"step": 430
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9474,
"step": 431
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7483,
"step": 432
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.811,
"step": 433
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7163,
"step": 434
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9745,
"step": 435
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7279,
"step": 436
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9218,
"step": 437
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8183,
"step": 438
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8229,
"step": 439
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7518,
"step": 440
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7858,
"step": 441
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 1.0076,
"step": 442
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9241,
"step": 443
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8994,
"step": 444
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.438,
"step": 445
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7214,
"step": 446
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5063,
"step": 447
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6436,
"step": 448
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6496,
"step": 449
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7387,
"step": 450
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.824,
"step": 451
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5313,
"step": 452
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7908,
"step": 453
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8242,
"step": 454
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8876,
"step": 455
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7456,
"step": 456
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.752,
"step": 457
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4918,
"step": 458
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.855,
"step": 459
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8322,
"step": 460
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7973,
"step": 461
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9598,
"step": 462
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.791,
"step": 463
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5249,
"step": 464
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6198,
"step": 465
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6606,
"step": 466
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7283,
"step": 467
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8863,
"step": 468
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6965,
"step": 469
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7057,
"step": 470
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8994,
"step": 471
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6775,
"step": 472
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5412,
"step": 473
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9663,
"step": 474
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9682,
"step": 475
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9535,
"step": 476
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7155,
"step": 477
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8014,
"step": 478
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8144,
"step": 479
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.713,
"step": 480
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6225,
"step": 481
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7206,
"step": 482
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5586,
"step": 483
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8996,
"step": 484
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7745,
"step": 485
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9241,
"step": 486
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7356,
"step": 487
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6449,
"step": 488
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7228,
"step": 489
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6462,
"step": 490
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8052,
"step": 491
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9312,
"step": 492
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7832,
"step": 493
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8995,
"step": 494
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7184,
"step": 495
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7811,
"step": 496
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8659,
"step": 497
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7348,
"step": 498
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7742,
"step": 499
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8397,
"step": 500
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6825,
"step": 501
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8492,
"step": 502
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0731,
"step": 503
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.669,
"step": 504
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6464,
"step": 505
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9044,
"step": 506
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.174,
"step": 507
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9606,
"step": 508
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0218,
"step": 509
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5801,
"step": 510
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0433,
"step": 511
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8027,
"step": 512
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0204,
"step": 513
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.9669,
"step": 514
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5325,
"step": 515
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.0595,
"step": 516
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 1.002,
"step": 517
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8445,
"step": 518
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8647,
"step": 519
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7309,
"step": 520
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5321,
"step": 521
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7862,
"step": 522
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8507,
"step": 523
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9764,
"step": 524
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5889,
"step": 525
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6066,
"step": 526
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7368,
"step": 527
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0567,
"step": 528
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8779,
"step": 529
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0253,
"step": 530
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8322,
"step": 531
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6848,
"step": 532
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5222,
"step": 533
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7199,
"step": 534
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8765,
"step": 535
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8593,
"step": 536
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8867,
"step": 537
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6727,
"step": 538
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6709,
"step": 539
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8352,
"step": 540
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.0596,
"step": 541
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7518,
"step": 542
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.57,
"step": 543
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7398,
"step": 544
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9338,
"step": 545
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7498,
"step": 546
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6023,
"step": 547
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8713,
"step": 548
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8121,
"step": 549
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 1.2485,
"step": 550
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8921,
"step": 551
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9471,
"step": 552
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4399,
"step": 553
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7338,
"step": 554
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8261,
"step": 555
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7494,
"step": 556
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7305,
"step": 557
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9084,
"step": 558
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7184,
"step": 559
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7275,
"step": 560
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8692,
"step": 561
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8769,
"step": 562
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7692,
"step": 563
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9968,
"step": 564
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6536,
"step": 565
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7829,
"step": 566
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7614,
"step": 567
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8699,
"step": 568
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8991,
"step": 569
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6783,
"step": 570
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7599,
"step": 571
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6944,
"step": 572
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.748,
"step": 573
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6174,
"step": 574
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6199,
"step": 575
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.8565,
"step": 576
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.9175,
"step": 577
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.578,
"step": 578
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7151,
"step": 579
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7445,
"step": 580
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6315,
"step": 581
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7391,
"step": 582
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8364,
"step": 583
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5496,
"step": 584
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7928,
"step": 585
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8136,
"step": 586
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5404,
"step": 587
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6769,
"step": 588
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6331,
"step": 589
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5495,
"step": 590
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7055,
"step": 591
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0395,
"step": 592
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8604,
"step": 593
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8628,
"step": 594
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4613,
"step": 595
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7412,
"step": 596
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9055,
"step": 597
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5592,
"step": 598
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0462,
"step": 599
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7479,
"step": 600
},
{
"epoch": 0.1,
"eval_loss": 0.7511235475540161,
"eval_runtime": 131.6743,
"eval_samples_per_second": 7.594,
"eval_steps_per_second": 3.797,
"step": 600
},
{
"epoch": 0.1,
"mmlu_eval_accuracy": 0.5854937013814044,
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
"mmlu_eval_accuracy_anatomy": 0.5714285714285714,
"mmlu_eval_accuracy_astronomy": 0.75,
"mmlu_eval_accuracy_business_ethics": 0.8181818181818182,
"mmlu_eval_accuracy_clinical_knowledge": 0.4827586206896552,
"mmlu_eval_accuracy_college_biology": 0.625,
"mmlu_eval_accuracy_college_chemistry": 0.375,
"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": 0.36363636363636365,
"mmlu_eval_accuracy_computer_security": 0.5454545454545454,
"mmlu_eval_accuracy_conceptual_physics": 0.5384615384615384,
"mmlu_eval_accuracy_econometrics": 0.5,
"mmlu_eval_accuracy_electrical_engineering": 0.6875,
"mmlu_eval_accuracy_elementary_mathematics": 0.4146341463414634,
"mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
"mmlu_eval_accuracy_global_facts": 0.3,
"mmlu_eval_accuracy_high_school_biology": 0.5,
"mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
"mmlu_eval_accuracy_high_school_european_history": 0.7222222222222222,
"mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.6511627906976745,
"mmlu_eval_accuracy_high_school_mathematics": 0.27586206896551724,
"mmlu_eval_accuracy_high_school_microeconomics": 0.5769230769230769,
"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": 0.8181818181818182,
"mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
"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": 0.45454545454545453,
"mmlu_eval_accuracy_logical_fallacies": 0.6111111111111112,
"mmlu_eval_accuracy_machine_learning": 0.36363636363636365,
"mmlu_eval_accuracy_management": 0.8181818181818182,
"mmlu_eval_accuracy_marketing": 0.92,
"mmlu_eval_accuracy_medical_genetics": 1.0,
"mmlu_eval_accuracy_miscellaneous": 0.7209302325581395,
"mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
"mmlu_eval_accuracy_moral_scenarios": 0.24,
"mmlu_eval_accuracy_nutrition": 0.7272727272727273,
"mmlu_eval_accuracy_philosophy": 0.6764705882352942,
"mmlu_eval_accuracy_prehistory": 0.4857142857142857,
"mmlu_eval_accuracy_professional_accounting": 0.5483870967741935,
"mmlu_eval_accuracy_professional_law": 0.36470588235294116,
"mmlu_eval_accuracy_professional_medicine": 0.6129032258064516,
"mmlu_eval_accuracy_professional_psychology": 0.6086956521739131,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.6666666666666666,
"mmlu_eval_accuracy_sociology": 0.8636363636363636,
"mmlu_eval_accuracy_us_foreign_policy": 0.9090909090909091,
"mmlu_eval_accuracy_virology": 0.6111111111111112,
"mmlu_eval_accuracy_world_religions": 0.8947368421052632,
"mmlu_loss": 0.8422373449697351,
"step": 600
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7688,
"step": 601
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8277,
"step": 602
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9892,
"step": 603
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.882,
"step": 604
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5368,
"step": 605
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6269,
"step": 606
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5438,
"step": 607
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5921,
"step": 608
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8068,
"step": 609
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8916,
"step": 610
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7217,
"step": 611
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9221,
"step": 612
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8893,
"step": 613
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6156,
"step": 614
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8421,
"step": 615
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7934,
"step": 616
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9781,
"step": 617
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8513,
"step": 618
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7201,
"step": 619
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9233,
"step": 620
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6561,
"step": 621
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7198,
"step": 622
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8383,
"step": 623
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8131,
"step": 624
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7621,
"step": 625
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.9067,
"step": 626
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0717,
"step": 627
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7355,
"step": 628
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.0598,
"step": 629
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8181,
"step": 630
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8185,
"step": 631
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.8387,
"step": 632
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6738,
"step": 633
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 1.2803,
"step": 634
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7685,
"step": 635
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7091,
"step": 636
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.779,
"step": 637
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6472,
"step": 638
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7769,
"step": 639
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6087,
"step": 640
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8182,
"step": 641
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5228,
"step": 642
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.3634,
"step": 643
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8811,
"step": 644
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.9722,
"step": 645
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7217,
"step": 646
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7172,
"step": 647
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.0123,
"step": 648
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.171,
"step": 649
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8549,
"step": 650
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.9442,
"step": 651
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8082,
"step": 652
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5462,
"step": 653
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7786,
"step": 654
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5764,
"step": 655
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5766,
"step": 656
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.2254,
"step": 657
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7426,
"step": 658
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7231,
"step": 659
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.9842,
"step": 660
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8422,
"step": 661
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.785,
"step": 662
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7621,
"step": 663
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8803,
"step": 664
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8239,
"step": 665
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8327,
"step": 666
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.9772,
"step": 667
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8029,
"step": 668
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5865,
"step": 669
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8794,
"step": 670
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8821,
"step": 671
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7357,
"step": 672
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6092,
"step": 673
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7127,
"step": 674
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7141,
"step": 675
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7687,
"step": 676
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.9527,
"step": 677
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8083,
"step": 678
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4753,
"step": 679
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7073,
"step": 680
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5191,
"step": 681
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6506,
"step": 682
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6917,
"step": 683
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6707,
"step": 684
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6223,
"step": 685
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7281,
"step": 686
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8363,
"step": 687
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8984,
"step": 688
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 1.0086,
"step": 689
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8176,
"step": 690
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6743,
"step": 691
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7653,
"step": 692
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.908,
"step": 693
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8402,
"step": 694
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8425,
"step": 695
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.767,
"step": 696
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8191,
"step": 697
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5398,
"step": 698
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6728,
"step": 699
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8391,
"step": 700
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7766,
"step": 701
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6116,
"step": 702
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.9081,
"step": 703
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6774,
"step": 704
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8251,
"step": 705
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.9524,
"step": 706
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.535,
"step": 707
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6306,
"step": 708
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.9171,
"step": 709
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.9689,
"step": 710
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.9917,
"step": 711
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8348,
"step": 712
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.587,
"step": 713
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5918,
"step": 714
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7032,
"step": 715
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8926,
"step": 716
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6812,
"step": 717
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.828,
"step": 718
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7452,
"step": 719
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6744,
"step": 720
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8305,
"step": 721
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6471,
"step": 722
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5812,
"step": 723
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5891,
"step": 724
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.94,
"step": 725
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7092,
"step": 726
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7654,
"step": 727
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.828,
"step": 728
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7061,
"step": 729
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5904,
"step": 730
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8591,
"step": 731
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.9372,
"step": 732
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8514,
"step": 733
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8774,
"step": 734
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.9201,
"step": 735
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8356,
"step": 736
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8392,
"step": 737
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7399,
"step": 738
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5375,
"step": 739
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8915,
"step": 740
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7966,
"step": 741
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6087,
"step": 742
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6648,
"step": 743
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 1.0216,
"step": 744
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.9432,
"step": 745
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6548,
"step": 746
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8015,
"step": 747
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7979,
"step": 748
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6144,
"step": 749
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6494,
"step": 750
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8614,
"step": 751
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.948,
"step": 752
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7197,
"step": 753
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8588,
"step": 754
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.9157,
"step": 755
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.8928,
"step": 756
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7218,
"step": 757
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.9789,
"step": 758
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6592,
"step": 759
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6305,
"step": 760
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.866,
"step": 761
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7438,
"step": 762
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6438,
"step": 763
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 1.0191,
"step": 764
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7692,
"step": 765
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6723,
"step": 766
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8188,
"step": 767
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6793,
"step": 768
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8244,
"step": 769
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.9137,
"step": 770
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7096,
"step": 771
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8164,
"step": 772
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7262,
"step": 773
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7389,
"step": 774
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 1.2896,
"step": 775
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8491,
"step": 776
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8356,
"step": 777
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8435,
"step": 778
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7388,
"step": 779
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.941,
"step": 780
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8939,
"step": 781
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7314,
"step": 782
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7922,
"step": 783
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6946,
"step": 784
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.752,
"step": 785
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5049,
"step": 786
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7308,
"step": 787
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7585,
"step": 788
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8932,
"step": 789
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 1.0651,
"step": 790
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.736,
"step": 791
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6819,
"step": 792
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6963,
"step": 793
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.9918,
"step": 794
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.9083,
"step": 795
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.9174,
"step": 796
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8295,
"step": 797
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 1.004,
"step": 798
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8248,
"step": 799
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5399,
"step": 800
},
{
"epoch": 0.13,
"eval_loss": 0.7548895478248596,
"eval_runtime": 131.5977,
"eval_samples_per_second": 7.599,
"eval_steps_per_second": 3.799,
"step": 800
},
{
"epoch": 0.13,
"mmlu_eval_accuracy": 0.5966805174146648,
"mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
"mmlu_eval_accuracy_anatomy": 0.5714285714285714,
"mmlu_eval_accuracy_astronomy": 0.625,
"mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
"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": 0.5454545454545454,
"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": 0.5384615384615384,
"mmlu_eval_accuracy_econometrics": 0.5,
"mmlu_eval_accuracy_electrical_engineering": 0.6875,
"mmlu_eval_accuracy_elementary_mathematics": 0.3902439024390244,
"mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
"mmlu_eval_accuracy_global_facts": 0.4,
"mmlu_eval_accuracy_high_school_biology": 0.5625,
"mmlu_eval_accuracy_high_school_chemistry": 0.36363636363636365,
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
"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": 0.6511627906976745,
"mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6153846153846154,
"mmlu_eval_accuracy_high_school_physics": 0.23529411764705882,
"mmlu_eval_accuracy_high_school_psychology": 0.85,
"mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
"mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
"mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
"mmlu_eval_accuracy_human_aging": 0.6956521739130435,
"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": 0.6111111111111112,
"mmlu_eval_accuracy_machine_learning": 0.2727272727272727,
"mmlu_eval_accuracy_management": 0.8181818181818182,
"mmlu_eval_accuracy_marketing": 0.96,
"mmlu_eval_accuracy_medical_genetics": 1.0,
"mmlu_eval_accuracy_miscellaneous": 0.7441860465116279,
"mmlu_eval_accuracy_moral_disputes": 0.5789473684210527,
"mmlu_eval_accuracy_moral_scenarios": 0.32,
"mmlu_eval_accuracy_nutrition": 0.7575757575757576,
"mmlu_eval_accuracy_philosophy": 0.6764705882352942,
"mmlu_eval_accuracy_prehistory": 0.4857142857142857,
"mmlu_eval_accuracy_professional_accounting": 0.6129032258064516,
"mmlu_eval_accuracy_professional_law": 0.38823529411764707,
"mmlu_eval_accuracy_professional_medicine": 0.6774193548387096,
"mmlu_eval_accuracy_professional_psychology": 0.5797101449275363,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.7407407407407407,
"mmlu_eval_accuracy_sociology": 0.8636363636363636,
"mmlu_eval_accuracy_us_foreign_policy": 0.9090909090909091,
"mmlu_eval_accuracy_virology": 0.5555555555555556,
"mmlu_eval_accuracy_world_religions": 0.9473684210526315,
"mmlu_loss": 0.9437568160868189,
"step": 800
}
],
"logging_steps": 1,
"max_steps": 18258,
"num_train_epochs": 3,
"save_steps": 200,
"total_flos": 1.640282490222674e+17,
"trial_name": null,
"trial_params": null
}