mistralic-expert-23 / checkpoint-600 /trainer_state.json
pharaouk's picture
Training in progress, step 600, checkpoint
2d92987
raw
history blame
75 kB
{
"best_metric": 0.5821242928504944,
"best_model_checkpoint": "experts/mistralic-expert-23/checkpoint-600",
"epoch": 0.14020329477742727,
"eval_steps": 200,
"global_step": 600,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6926,
"step": 1
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6169,
"step": 2
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.539,
"step": 3
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5802,
"step": 4
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5176,
"step": 5
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6371,
"step": 6
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5969,
"step": 7
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6545,
"step": 8
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7967,
"step": 9
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5244,
"step": 10
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9098,
"step": 11
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.7216,
"step": 12
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5279,
"step": 13
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.784,
"step": 14
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.9971,
"step": 15
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5255,
"step": 16
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.6404,
"step": 17
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.8234,
"step": 18
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.529,
"step": 19
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.5078,
"step": 20
},
{
"epoch": 0.0,
"learning_rate": 0.0002,
"loss": 0.4981,
"step": 21
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8841,
"step": 22
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5563,
"step": 23
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5759,
"step": 24
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5739,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5891,
"step": 26
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.454,
"step": 27
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5685,
"step": 28
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5723,
"step": 29
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.516,
"step": 30
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7419,
"step": 31
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7529,
"step": 32
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6331,
"step": 33
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7149,
"step": 34
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7413,
"step": 35
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5488,
"step": 36
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8207,
"step": 37
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5589,
"step": 38
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5907,
"step": 39
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6014,
"step": 40
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5317,
"step": 41
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5762,
"step": 42
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5986,
"step": 43
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5758,
"step": 44
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7376,
"step": 45
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6161,
"step": 46
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5554,
"step": 47
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5997,
"step": 48
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5616,
"step": 49
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5376,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6303,
"step": 51
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5722,
"step": 52
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5588,
"step": 53
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5734,
"step": 54
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5353,
"step": 55
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5713,
"step": 56
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6006,
"step": 57
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.7782,
"step": 58
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5286,
"step": 59
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6655,
"step": 60
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.6916,
"step": 61
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.8167,
"step": 62
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5933,
"step": 63
},
{
"epoch": 0.01,
"learning_rate": 0.0002,
"loss": 0.5054,
"step": 64
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5764,
"step": 65
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.4551,
"step": 66
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6122,
"step": 67
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6257,
"step": 68
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6027,
"step": 69
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5796,
"step": 70
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6817,
"step": 71
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5495,
"step": 72
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5997,
"step": 73
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.837,
"step": 74
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5642,
"step": 75
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5644,
"step": 76
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5589,
"step": 77
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6021,
"step": 78
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6688,
"step": 79
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5716,
"step": 80
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.75,
"step": 81
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6068,
"step": 82
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5987,
"step": 83
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7231,
"step": 84
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5786,
"step": 85
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6502,
"step": 86
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6702,
"step": 87
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6432,
"step": 88
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5064,
"step": 89
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5188,
"step": 90
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.9853,
"step": 91
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5188,
"step": 92
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8395,
"step": 93
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5267,
"step": 94
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5357,
"step": 95
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5574,
"step": 96
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5883,
"step": 97
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6675,
"step": 98
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5527,
"step": 99
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6302,
"step": 100
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.8618,
"step": 101
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6703,
"step": 102
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.5242,
"step": 103
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.653,
"step": 104
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.6386,
"step": 105
},
{
"epoch": 0.02,
"learning_rate": 0.0002,
"loss": 0.7444,
"step": 106
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5107,
"step": 107
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6885,
"step": 108
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5642,
"step": 109
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6477,
"step": 110
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6824,
"step": 111
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5575,
"step": 112
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5927,
"step": 113
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5312,
"step": 114
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.63,
"step": 115
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4778,
"step": 116
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5675,
"step": 117
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4322,
"step": 118
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5693,
"step": 119
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6603,
"step": 120
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4729,
"step": 121
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8347,
"step": 122
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6119,
"step": 123
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7242,
"step": 124
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5018,
"step": 125
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.7959,
"step": 126
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.606,
"step": 127
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5727,
"step": 128
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6316,
"step": 129
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6095,
"step": 130
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5735,
"step": 131
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4528,
"step": 132
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5352,
"step": 133
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6042,
"step": 134
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.76,
"step": 135
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.4522,
"step": 136
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5826,
"step": 137
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5408,
"step": 138
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5659,
"step": 139
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.551,
"step": 140
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5766,
"step": 141
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5377,
"step": 142
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.8203,
"step": 143
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5066,
"step": 144
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5001,
"step": 145
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.5727,
"step": 146
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6109,
"step": 147
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6467,
"step": 148
},
{
"epoch": 0.03,
"learning_rate": 0.0002,
"loss": 0.6917,
"step": 149
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6373,
"step": 150
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6928,
"step": 151
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5866,
"step": 152
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.659,
"step": 153
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5483,
"step": 154
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5335,
"step": 155
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7856,
"step": 156
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5694,
"step": 157
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8404,
"step": 158
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.575,
"step": 159
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6436,
"step": 160
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4149,
"step": 161
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6085,
"step": 162
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5949,
"step": 163
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6294,
"step": 164
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5303,
"step": 165
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7256,
"step": 166
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6095,
"step": 167
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7129,
"step": 168
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6088,
"step": 169
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4735,
"step": 170
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4963,
"step": 171
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.8293,
"step": 172
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6143,
"step": 173
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5835,
"step": 174
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.7986,
"step": 175
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6569,
"step": 176
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6295,
"step": 177
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4927,
"step": 178
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5401,
"step": 179
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6203,
"step": 180
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5045,
"step": 181
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5779,
"step": 182
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5386,
"step": 183
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.4483,
"step": 184
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5543,
"step": 185
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 1.0266,
"step": 186
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5621,
"step": 187
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5884,
"step": 188
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5395,
"step": 189
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6764,
"step": 190
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.5938,
"step": 191
},
{
"epoch": 0.04,
"learning_rate": 0.0002,
"loss": 0.6306,
"step": 192
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8946,
"step": 193
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6613,
"step": 194
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4926,
"step": 195
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4874,
"step": 196
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6966,
"step": 197
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7493,
"step": 198
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.62,
"step": 199
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8005,
"step": 200
},
{
"epoch": 0.05,
"eval_loss": 0.600193977355957,
"eval_runtime": 156.5363,
"eval_samples_per_second": 6.388,
"eval_steps_per_second": 3.194,
"step": 200
},
{
"epoch": 0.05,
"mmlu_eval_accuracy": 0.6110525160674652,
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
"mmlu_eval_accuracy_anatomy": 0.5714285714285714,
"mmlu_eval_accuracy_astronomy": 0.75,
"mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
"mmlu_eval_accuracy_clinical_knowledge": 0.5862068965517241,
"mmlu_eval_accuracy_college_biology": 0.625,
"mmlu_eval_accuracy_college_chemistry": 0.375,
"mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
"mmlu_eval_accuracy_college_mathematics": 0.45454545454545453,
"mmlu_eval_accuracy_college_medicine": 0.6818181818181818,
"mmlu_eval_accuracy_college_physics": 0.5454545454545454,
"mmlu_eval_accuracy_computer_security": 0.6363636363636364,
"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": 0.21428571428571427,
"mmlu_eval_accuracy_global_facts": 0.3,
"mmlu_eval_accuracy_high_school_biology": 0.625,
"mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
"mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_european_history": 0.8333333333333334,
"mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.6046511627906976,
"mmlu_eval_accuracy_high_school_mathematics": 0.3103448275862069,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6538461538461539,
"mmlu_eval_accuracy_high_school_physics": 0.23529411764705882,
"mmlu_eval_accuracy_high_school_psychology": 0.8666666666666667,
"mmlu_eval_accuracy_high_school_statistics": 0.43478260869565216,
"mmlu_eval_accuracy_high_school_us_history": 0.7727272727272727,
"mmlu_eval_accuracy_high_school_world_history": 0.6153846153846154,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.5833333333333334,
"mmlu_eval_accuracy_international_law": 0.9230769230769231,
"mmlu_eval_accuracy_jurisprudence": 0.6363636363636364,
"mmlu_eval_accuracy_logical_fallacies": 0.6111111111111112,
"mmlu_eval_accuracy_machine_learning": 0.5454545454545454,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.92,
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
"mmlu_eval_accuracy_miscellaneous": 0.7325581395348837,
"mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
"mmlu_eval_accuracy_moral_scenarios": 0.37,
"mmlu_eval_accuracy_nutrition": 0.7272727272727273,
"mmlu_eval_accuracy_philosophy": 0.7352941176470589,
"mmlu_eval_accuracy_prehistory": 0.6,
"mmlu_eval_accuracy_professional_accounting": 0.5161290322580645,
"mmlu_eval_accuracy_professional_law": 0.4,
"mmlu_eval_accuracy_professional_medicine": 0.6774193548387096,
"mmlu_eval_accuracy_professional_psychology": 0.6231884057971014,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.7407407407407407,
"mmlu_eval_accuracy_sociology": 0.9090909090909091,
"mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
"mmlu_eval_accuracy_virology": 0.5555555555555556,
"mmlu_eval_accuracy_world_religions": 0.8947368421052632,
"mmlu_loss": 1.4672958862532524,
"step": 200
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4891,
"step": 201
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5071,
"step": 202
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7988,
"step": 203
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.542,
"step": 204
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.501,
"step": 205
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6016,
"step": 206
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7817,
"step": 207
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5444,
"step": 208
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7871,
"step": 209
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6683,
"step": 210
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4517,
"step": 211
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8223,
"step": 212
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.536,
"step": 213
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5169,
"step": 214
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6521,
"step": 215
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5043,
"step": 216
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5211,
"step": 217
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5449,
"step": 218
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5034,
"step": 219
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8192,
"step": 220
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5212,
"step": 221
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6259,
"step": 222
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5675,
"step": 223
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5851,
"step": 224
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.7381,
"step": 225
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.6499,
"step": 226
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8992,
"step": 227
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5907,
"step": 228
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.602,
"step": 229
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5298,
"step": 230
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.4669,
"step": 231
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8426,
"step": 232
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.8515,
"step": 233
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5737,
"step": 234
},
{
"epoch": 0.05,
"learning_rate": 0.0002,
"loss": 0.5501,
"step": 235
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4967,
"step": 236
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6454,
"step": 237
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4659,
"step": 238
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.579,
"step": 239
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7209,
"step": 240
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5715,
"step": 241
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4989,
"step": 242
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6546,
"step": 243
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7433,
"step": 244
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5624,
"step": 245
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5985,
"step": 246
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5385,
"step": 247
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6101,
"step": 248
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5485,
"step": 249
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6936,
"step": 250
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.551,
"step": 251
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5149,
"step": 252
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5189,
"step": 253
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7124,
"step": 254
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6658,
"step": 255
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5099,
"step": 256
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4779,
"step": 257
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5552,
"step": 258
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5756,
"step": 259
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5334,
"step": 260
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6678,
"step": 261
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7578,
"step": 262
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7903,
"step": 263
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5694,
"step": 264
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7169,
"step": 265
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4679,
"step": 266
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6381,
"step": 267
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5931,
"step": 268
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5818,
"step": 269
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7225,
"step": 270
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6526,
"step": 271
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5631,
"step": 272
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.7555,
"step": 273
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.6211,
"step": 274
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5831,
"step": 275
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.5269,
"step": 276
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.4996,
"step": 277
},
{
"epoch": 0.06,
"learning_rate": 0.0002,
"loss": 0.8797,
"step": 278
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4943,
"step": 279
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8619,
"step": 280
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6191,
"step": 281
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4933,
"step": 282
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.744,
"step": 283
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4481,
"step": 284
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5396,
"step": 285
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5108,
"step": 286
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4857,
"step": 287
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5603,
"step": 288
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5778,
"step": 289
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6088,
"step": 290
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.52,
"step": 291
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6123,
"step": 292
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5051,
"step": 293
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5057,
"step": 294
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.601,
"step": 295
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.9173,
"step": 296
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5194,
"step": 297
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6205,
"step": 298
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5811,
"step": 299
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.8153,
"step": 300
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5872,
"step": 301
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6062,
"step": 302
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4371,
"step": 303
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5227,
"step": 304
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5014,
"step": 305
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4599,
"step": 306
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5638,
"step": 307
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4859,
"step": 308
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5348,
"step": 309
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4462,
"step": 310
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.6522,
"step": 311
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7238,
"step": 312
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4864,
"step": 313
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7379,
"step": 314
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4141,
"step": 315
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7453,
"step": 316
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.4621,
"step": 317
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.5194,
"step": 318
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7686,
"step": 319
},
{
"epoch": 0.07,
"learning_rate": 0.0002,
"loss": 0.7024,
"step": 320
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6979,
"step": 321
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.717,
"step": 322
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5238,
"step": 323
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5048,
"step": 324
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5091,
"step": 325
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7207,
"step": 326
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5248,
"step": 327
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8079,
"step": 328
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6206,
"step": 329
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6146,
"step": 330
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5315,
"step": 331
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.53,
"step": 332
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.794,
"step": 333
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4968,
"step": 334
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5732,
"step": 335
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5856,
"step": 336
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.469,
"step": 337
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.638,
"step": 338
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7457,
"step": 339
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5851,
"step": 340
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6549,
"step": 341
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5412,
"step": 342
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7014,
"step": 343
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.727,
"step": 344
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5415,
"step": 345
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5822,
"step": 346
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.585,
"step": 347
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5198,
"step": 348
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5274,
"step": 349
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5288,
"step": 350
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5649,
"step": 351
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5385,
"step": 352
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6064,
"step": 353
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5652,
"step": 354
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4739,
"step": 355
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.4917,
"step": 356
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6023,
"step": 357
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.7409,
"step": 358
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.6717,
"step": 359
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.8324,
"step": 360
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5896,
"step": 361
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5608,
"step": 362
},
{
"epoch": 0.08,
"learning_rate": 0.0002,
"loss": 0.5038,
"step": 363
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5431,
"step": 364
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7616,
"step": 365
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5106,
"step": 366
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4998,
"step": 367
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5058,
"step": 368
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.536,
"step": 369
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5248,
"step": 370
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.728,
"step": 371
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4739,
"step": 372
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7373,
"step": 373
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5395,
"step": 374
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5658,
"step": 375
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4556,
"step": 376
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5087,
"step": 377
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5215,
"step": 378
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4547,
"step": 379
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6626,
"step": 380
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.666,
"step": 381
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7555,
"step": 382
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6575,
"step": 383
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6711,
"step": 384
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5149,
"step": 385
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6033,
"step": 386
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5753,
"step": 387
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5267,
"step": 388
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.664,
"step": 389
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6835,
"step": 390
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5874,
"step": 391
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5413,
"step": 392
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.579,
"step": 393
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5294,
"step": 394
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5736,
"step": 395
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5794,
"step": 396
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6177,
"step": 397
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5843,
"step": 398
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6333,
"step": 399
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.7496,
"step": 400
},
{
"epoch": 0.09,
"eval_loss": 0.5899355411529541,
"eval_runtime": 156.5886,
"eval_samples_per_second": 6.386,
"eval_steps_per_second": 3.193,
"step": 400
},
{
"epoch": 0.09,
"mmlu_eval_accuracy": 0.6068635133301271,
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
"mmlu_eval_accuracy_anatomy": 0.5,
"mmlu_eval_accuracy_astronomy": 0.6875,
"mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
"mmlu_eval_accuracy_clinical_knowledge": 0.6206896551724138,
"mmlu_eval_accuracy_college_biology": 0.6875,
"mmlu_eval_accuracy_college_chemistry": 0.25,
"mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
"mmlu_eval_accuracy_college_mathematics": 0.5454545454545454,
"mmlu_eval_accuracy_college_medicine": 0.6363636363636364,
"mmlu_eval_accuracy_college_physics": 0.5454545454545454,
"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": 0.8125,
"mmlu_eval_accuracy_elementary_mathematics": 0.43902439024390244,
"mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
"mmlu_eval_accuracy_global_facts": 0.4,
"mmlu_eval_accuracy_high_school_biology": 0.65625,
"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": 0.8636363636363636,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7619047619047619,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.5813953488372093,
"mmlu_eval_accuracy_high_school_mathematics": 0.27586206896551724,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6923076923076923,
"mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
"mmlu_eval_accuracy_high_school_psychology": 0.8666666666666667,
"mmlu_eval_accuracy_high_school_statistics": 0.4782608695652174,
"mmlu_eval_accuracy_high_school_us_history": 0.8181818181818182,
"mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
"mmlu_eval_accuracy_human_aging": 0.6956521739130435,
"mmlu_eval_accuracy_human_sexuality": 0.5833333333333334,
"mmlu_eval_accuracy_international_law": 0.8461538461538461,
"mmlu_eval_accuracy_jurisprudence": 0.6363636363636364,
"mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
"mmlu_eval_accuracy_machine_learning": 0.5454545454545454,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.88,
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
"mmlu_eval_accuracy_miscellaneous": 0.7441860465116279,
"mmlu_eval_accuracy_moral_disputes": 0.6052631578947368,
"mmlu_eval_accuracy_moral_scenarios": 0.37,
"mmlu_eval_accuracy_nutrition": 0.696969696969697,
"mmlu_eval_accuracy_philosophy": 0.7058823529411765,
"mmlu_eval_accuracy_prehistory": 0.6,
"mmlu_eval_accuracy_professional_accounting": 0.4838709677419355,
"mmlu_eval_accuracy_professional_law": 0.4117647058823529,
"mmlu_eval_accuracy_professional_medicine": 0.6774193548387096,
"mmlu_eval_accuracy_professional_psychology": 0.6231884057971014,
"mmlu_eval_accuracy_public_relations": 0.5,
"mmlu_eval_accuracy_security_studies": 0.6666666666666666,
"mmlu_eval_accuracy_sociology": 0.9545454545454546,
"mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
"mmlu_eval_accuracy_virology": 0.5,
"mmlu_eval_accuracy_world_religions": 0.7894736842105263,
"mmlu_loss": 1.496563289402048,
"step": 400
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5013,
"step": 401
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.509,
"step": 402
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.6059,
"step": 403
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.4962,
"step": 404
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5252,
"step": 405
},
{
"epoch": 0.09,
"learning_rate": 0.0002,
"loss": 0.5063,
"step": 406
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4901,
"step": 407
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.636,
"step": 408
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5539,
"step": 409
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6965,
"step": 410
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.767,
"step": 411
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7795,
"step": 412
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.488,
"step": 413
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7407,
"step": 414
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6471,
"step": 415
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5734,
"step": 416
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5481,
"step": 417
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6453,
"step": 418
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4902,
"step": 419
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6816,
"step": 420
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5906,
"step": 421
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5285,
"step": 422
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5178,
"step": 423
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5605,
"step": 424
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7357,
"step": 425
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6648,
"step": 426
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5083,
"step": 427
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4889,
"step": 428
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5004,
"step": 429
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4759,
"step": 430
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6473,
"step": 431
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5438,
"step": 432
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5149,
"step": 433
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.839,
"step": 434
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5581,
"step": 435
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5813,
"step": 436
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4878,
"step": 437
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5984,
"step": 438
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7623,
"step": 439
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.6718,
"step": 440
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5112,
"step": 441
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.4414,
"step": 442
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5551,
"step": 443
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5107,
"step": 444
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5318,
"step": 445
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.504,
"step": 446
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.5137,
"step": 447
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.588,
"step": 448
},
{
"epoch": 0.1,
"learning_rate": 0.0002,
"loss": 0.7416,
"step": 449
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5064,
"step": 450
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4845,
"step": 451
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.8342,
"step": 452
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5693,
"step": 453
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4511,
"step": 454
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4945,
"step": 455
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5943,
"step": 456
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6719,
"step": 457
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5331,
"step": 458
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6232,
"step": 459
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7195,
"step": 460
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7247,
"step": 461
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7871,
"step": 462
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5414,
"step": 463
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5728,
"step": 464
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5263,
"step": 465
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5105,
"step": 466
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5922,
"step": 467
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5316,
"step": 468
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4836,
"step": 469
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4781,
"step": 470
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.645,
"step": 471
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6311,
"step": 472
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5401,
"step": 473
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5423,
"step": 474
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.7188,
"step": 475
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.587,
"step": 476
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.46,
"step": 477
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5724,
"step": 478
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5907,
"step": 479
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.484,
"step": 480
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5095,
"step": 481
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5197,
"step": 482
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6144,
"step": 483
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4964,
"step": 484
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4705,
"step": 485
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.4844,
"step": 486
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6951,
"step": 487
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5948,
"step": 488
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5728,
"step": 489
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6655,
"step": 490
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.6506,
"step": 491
},
{
"epoch": 0.11,
"learning_rate": 0.0002,
"loss": 0.5119,
"step": 492
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6653,
"step": 493
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5105,
"step": 494
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5486,
"step": 495
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7635,
"step": 496
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5753,
"step": 497
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5336,
"step": 498
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5103,
"step": 499
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4841,
"step": 500
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5479,
"step": 501
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6147,
"step": 502
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4874,
"step": 503
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7884,
"step": 504
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6535,
"step": 505
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6987,
"step": 506
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6371,
"step": 507
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5822,
"step": 508
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7897,
"step": 509
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4869,
"step": 510
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6293,
"step": 511
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.7633,
"step": 512
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6217,
"step": 513
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5522,
"step": 514
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5368,
"step": 515
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5667,
"step": 516
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5541,
"step": 517
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5003,
"step": 518
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5217,
"step": 519
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.467,
"step": 520
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4416,
"step": 521
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4778,
"step": 522
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5085,
"step": 523
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.6101,
"step": 524
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4561,
"step": 525
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.56,
"step": 526
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5205,
"step": 527
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4748,
"step": 528
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5813,
"step": 529
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5873,
"step": 530
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.5845,
"step": 531
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.4624,
"step": 532
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.546,
"step": 533
},
{
"epoch": 0.12,
"learning_rate": 0.0002,
"loss": 0.9573,
"step": 534
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5634,
"step": 535
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.3786,
"step": 536
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7356,
"step": 537
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5748,
"step": 538
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6774,
"step": 539
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5361,
"step": 540
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.465,
"step": 541
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5905,
"step": 542
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6072,
"step": 543
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5397,
"step": 544
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5201,
"step": 545
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7455,
"step": 546
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.672,
"step": 547
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6171,
"step": 548
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5088,
"step": 549
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5438,
"step": 550
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4865,
"step": 551
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5796,
"step": 552
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6324,
"step": 553
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4357,
"step": 554
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4484,
"step": 555
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6764,
"step": 556
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5309,
"step": 557
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5603,
"step": 558
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7126,
"step": 559
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4893,
"step": 560
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5384,
"step": 561
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4642,
"step": 562
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4378,
"step": 563
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5831,
"step": 564
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5491,
"step": 565
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6361,
"step": 566
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 1.0347,
"step": 567
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.4843,
"step": 568
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.762,
"step": 569
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7908,
"step": 570
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5221,
"step": 571
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.8129,
"step": 572
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.542,
"step": 573
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5179,
"step": 574
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.5375,
"step": 575
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.6367,
"step": 576
},
{
"epoch": 0.13,
"learning_rate": 0.0002,
"loss": 0.7865,
"step": 577
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5287,
"step": 578
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5414,
"step": 579
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6124,
"step": 580
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.8515,
"step": 581
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5426,
"step": 582
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5048,
"step": 583
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.559,
"step": 584
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5914,
"step": 585
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5469,
"step": 586
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.719,
"step": 587
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5888,
"step": 588
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4804,
"step": 589
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4549,
"step": 590
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.4929,
"step": 591
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6596,
"step": 592
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5239,
"step": 593
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6946,
"step": 594
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7026,
"step": 595
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7988,
"step": 596
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.6868,
"step": 597
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7739,
"step": 598
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.5263,
"step": 599
},
{
"epoch": 0.14,
"learning_rate": 0.0002,
"loss": 0.7381,
"step": 600
},
{
"epoch": 0.14,
"eval_loss": 0.5821242928504944,
"eval_runtime": 156.5384,
"eval_samples_per_second": 6.388,
"eval_steps_per_second": 3.194,
"step": 600
},
{
"epoch": 0.14,
"mmlu_eval_accuracy": 0.5980259299067578,
"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": 0.625,
"mmlu_eval_accuracy_college_chemistry": 0.125,
"mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
"mmlu_eval_accuracy_college_mathematics": 0.6363636363636364,
"mmlu_eval_accuracy_college_medicine": 0.6818181818181818,
"mmlu_eval_accuracy_college_physics": 0.5454545454545454,
"mmlu_eval_accuracy_computer_security": 0.6363636363636364,
"mmlu_eval_accuracy_conceptual_physics": 0.5,
"mmlu_eval_accuracy_econometrics": 0.5,
"mmlu_eval_accuracy_electrical_engineering": 0.6875,
"mmlu_eval_accuracy_elementary_mathematics": 0.3902439024390244,
"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": 0.6666666666666666,
"mmlu_eval_accuracy_high_school_european_history": 0.8333333333333334,
"mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
"mmlu_eval_accuracy_high_school_government_and_politics": 0.7142857142857143,
"mmlu_eval_accuracy_high_school_macroeconomics": 0.5813953488372093,
"mmlu_eval_accuracy_high_school_mathematics": 0.3103448275862069,
"mmlu_eval_accuracy_high_school_microeconomics": 0.6923076923076923,
"mmlu_eval_accuracy_high_school_physics": 0.17647058823529413,
"mmlu_eval_accuracy_high_school_psychology": 0.85,
"mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
"mmlu_eval_accuracy_high_school_us_history": 0.6818181818181818,
"mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
"mmlu_eval_accuracy_human_sexuality": 0.5833333333333334,
"mmlu_eval_accuracy_international_law": 0.9230769230769231,
"mmlu_eval_accuracy_jurisprudence": 0.6363636363636364,
"mmlu_eval_accuracy_logical_fallacies": 0.7222222222222222,
"mmlu_eval_accuracy_machine_learning": 0.6363636363636364,
"mmlu_eval_accuracy_management": 0.9090909090909091,
"mmlu_eval_accuracy_marketing": 0.84,
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
"mmlu_eval_accuracy_miscellaneous": 0.7209302325581395,
"mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
"mmlu_eval_accuracy_moral_scenarios": 0.31,
"mmlu_eval_accuracy_nutrition": 0.6666666666666666,
"mmlu_eval_accuracy_philosophy": 0.7058823529411765,
"mmlu_eval_accuracy_prehistory": 0.6,
"mmlu_eval_accuracy_professional_accounting": 0.5161290322580645,
"mmlu_eval_accuracy_professional_law": 0.38235294117647056,
"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": 0.8636363636363636,
"mmlu_eval_accuracy_us_foreign_policy": 0.9090909090909091,
"mmlu_eval_accuracy_virology": 0.5,
"mmlu_eval_accuracy_world_religions": 0.7894736842105263,
"mmlu_loss": 1.5698812617178686,
"step": 600
}
],
"logging_steps": 1,
"max_steps": 12837,
"num_train_epochs": 3,
"save_steps": 200,
"total_flos": 1.4813142740749517e+17,
"trial_name": null,
"trial_params": null
}