Fathom: upload expert-e6-detection/training_log.json
Browse files
adapters/expert-e6-detection/training_log.json
ADDED
|
@@ -0,0 +1,634 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"loss": 1.9427,
|
| 4 |
+
"grad_norm": 0.9304783344268799,
|
| 5 |
+
"learning_rate": 1.8e-05,
|
| 6 |
+
"entropy": 1.4844072341918946,
|
| 7 |
+
"num_tokens": 236335.0,
|
| 8 |
+
"mean_token_accuracy": 0.5745298147201539,
|
| 9 |
+
"epoch": 0.016,
|
| 10 |
+
"step": 10
|
| 11 |
+
},
|
| 12 |
+
{
|
| 13 |
+
"loss": 1.8013,
|
| 14 |
+
"grad_norm": 0.723517656326294,
|
| 15 |
+
"learning_rate": 3.8e-05,
|
| 16 |
+
"entropy": 1.7161851286888123,
|
| 17 |
+
"num_tokens": 471120.0,
|
| 18 |
+
"mean_token_accuracy": 0.5827198266983032,
|
| 19 |
+
"epoch": 0.032,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"loss": 1.4628,
|
| 24 |
+
"grad_norm": 1.4500237703323364,
|
| 25 |
+
"learning_rate": 5.8e-05,
|
| 26 |
+
"entropy": 1.5137762308120728,
|
| 27 |
+
"num_tokens": 713049.0,
|
| 28 |
+
"mean_token_accuracy": 0.6408967107534409,
|
| 29 |
+
"epoch": 0.048,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"loss": 1.0328,
|
| 34 |
+
"grad_norm": 0.41915225982666016,
|
| 35 |
+
"learning_rate": 7.800000000000001e-05,
|
| 36 |
+
"entropy": 1.0311345964670182,
|
| 37 |
+
"num_tokens": 950524.0,
|
| 38 |
+
"mean_token_accuracy": 0.7369566798210144,
|
| 39 |
+
"epoch": 0.064,
|
| 40 |
+
"step": 40
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"loss": 0.9175,
|
| 44 |
+
"grad_norm": 0.3350801169872284,
|
| 45 |
+
"learning_rate": 9.8e-05,
|
| 46 |
+
"entropy": 0.9200641274452209,
|
| 47 |
+
"num_tokens": 1181734.0,
|
| 48 |
+
"mean_token_accuracy": 0.7648449391126633,
|
| 49 |
+
"epoch": 0.08,
|
| 50 |
+
"step": 50
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"loss": 0.8685,
|
| 54 |
+
"grad_norm": 0.33790573477745056,
|
| 55 |
+
"learning_rate": 9.993956318446873e-05,
|
| 56 |
+
"entropy": 0.8751217514276505,
|
| 57 |
+
"num_tokens": 1417485.0,
|
| 58 |
+
"mean_token_accuracy": 0.7721945941448212,
|
| 59 |
+
"epoch": 0.096,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"loss": 0.8228,
|
| 64 |
+
"grad_norm": 0.3235560357570648,
|
| 65 |
+
"learning_rate": 9.973083336646172e-05,
|
| 66 |
+
"entropy": 0.8255001336336136,
|
| 67 |
+
"num_tokens": 1654923.0,
|
| 68 |
+
"mean_token_accuracy": 0.7804904520511627,
|
| 69 |
+
"epoch": 0.112,
|
| 70 |
+
"step": 70
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"loss": 0.7851,
|
| 74 |
+
"grad_norm": 0.34785518050193787,
|
| 75 |
+
"learning_rate": 9.937368719576374e-05,
|
| 76 |
+
"entropy": 0.7923080742359161,
|
| 77 |
+
"num_tokens": 1893268.0,
|
| 78 |
+
"mean_token_accuracy": 0.7877459049224853,
|
| 79 |
+
"epoch": 0.128,
|
| 80 |
+
"step": 80
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"loss": 0.784,
|
| 84 |
+
"grad_norm": 0.33706751465797424,
|
| 85 |
+
"learning_rate": 9.886919053691883e-05,
|
| 86 |
+
"entropy": 0.7805845379829407,
|
| 87 |
+
"num_tokens": 2131912.0,
|
| 88 |
+
"mean_token_accuracy": 0.7893630683422088,
|
| 89 |
+
"epoch": 0.144,
|
| 90 |
+
"step": 90
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"loss": 0.7473,
|
| 94 |
+
"grad_norm": 0.3280414938926697,
|
| 95 |
+
"learning_rate": 9.82188490062415e-05,
|
| 96 |
+
"entropy": 0.7436369717121124,
|
| 97 |
+
"num_tokens": 2371393.0,
|
| 98 |
+
"mean_token_accuracy": 0.7966683268547058,
|
| 99 |
+
"epoch": 0.16,
|
| 100 |
+
"step": 100
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"loss": 0.7439,
|
| 104 |
+
"grad_norm": 0.3483212888240814,
|
| 105 |
+
"learning_rate": 9.742460347846587e-05,
|
| 106 |
+
"entropy": 0.7480131626129151,
|
| 107 |
+
"num_tokens": 2608341.0,
|
| 108 |
+
"mean_token_accuracy": 0.7970409005880356,
|
| 109 |
+
"epoch": 0.176,
|
| 110 |
+
"step": 110
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"loss": 0.741,
|
| 114 |
+
"grad_norm": 0.34795472025871277,
|
| 115 |
+
"learning_rate": 9.648882429441257e-05,
|
| 116 |
+
"entropy": 0.7411326467990875,
|
| 117 |
+
"num_tokens": 2845727.0,
|
| 118 |
+
"mean_token_accuracy": 0.7980288416147232,
|
| 119 |
+
"epoch": 0.192,
|
| 120 |
+
"step": 120
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"loss": 0.7874,
|
| 124 |
+
"grad_norm": 0.4145350754261017,
|
| 125 |
+
"learning_rate": 9.541430418696018e-05,
|
| 126 |
+
"entropy": 0.7868094980716706,
|
| 127 |
+
"num_tokens": 3082504.0,
|
| 128 |
+
"mean_token_accuracy": 0.7893972307443619,
|
| 129 |
+
"epoch": 0.208,
|
| 130 |
+
"step": 130
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"loss": 0.7625,
|
| 134 |
+
"grad_norm": 1.9233369827270508,
|
| 135 |
+
"learning_rate": 9.420424994643266e-05,
|
| 136 |
+
"entropy": 0.7673351168632507,
|
| 137 |
+
"num_tokens": 3318727.0,
|
| 138 |
+
"mean_token_accuracy": 0.7942031443119049,
|
| 139 |
+
"epoch": 0.224,
|
| 140 |
+
"step": 140
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"loss": 0.7391,
|
| 144 |
+
"grad_norm": 0.43922004103660583,
|
| 145 |
+
"learning_rate": 9.28622728502766e-05,
|
| 146 |
+
"entropy": 0.7414912819862366,
|
| 147 |
+
"num_tokens": 3551849.0,
|
| 148 |
+
"mean_token_accuracy": 0.7978828340768814,
|
| 149 |
+
"epoch": 0.24,
|
| 150 |
+
"step": 150
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"loss": 0.702,
|
| 154 |
+
"grad_norm": 0.3687349259853363,
|
| 155 |
+
"learning_rate": 9.139237788559e-05,
|
| 156 |
+
"entropy": 0.7002590656280517,
|
| 157 |
+
"num_tokens": 3784063.0,
|
| 158 |
+
"mean_token_accuracy": 0.8059363931417465,
|
| 159 |
+
"epoch": 0.256,
|
| 160 |
+
"step": 160
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"loss": 0.7214,
|
| 164 |
+
"grad_norm": 0.3384023904800415,
|
| 165 |
+
"learning_rate": 8.979895179666673e-05,
|
| 166 |
+
"entropy": 0.7226290732622147,
|
| 167 |
+
"num_tokens": 4030658.0,
|
| 168 |
+
"mean_token_accuracy": 0.80358906686306,
|
| 169 |
+
"epoch": 0.272,
|
| 170 |
+
"step": 170
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"loss": 0.7132,
|
| 174 |
+
"grad_norm": 0.39796605706214905,
|
| 175 |
+
"learning_rate": 8.808674999322728e-05,
|
| 176 |
+
"entropy": 0.7169670343399048,
|
| 177 |
+
"num_tokens": 4264808.0,
|
| 178 |
+
"mean_token_accuracy": 0.8048111826181412,
|
| 179 |
+
"epoch": 0.288,
|
| 180 |
+
"step": 180
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"loss": 0.721,
|
| 184 |
+
"grad_norm": 0.4140866696834564,
|
| 185 |
+
"learning_rate": 8.626088235840726e-05,
|
| 186 |
+
"entropy": 0.7286807566881179,
|
| 187 |
+
"num_tokens": 4500615.0,
|
| 188 |
+
"mean_token_accuracy": 0.8025989532470703,
|
| 189 |
+
"epoch": 0.304,
|
| 190 |
+
"step": 190
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"loss": 0.693,
|
| 194 |
+
"grad_norm": 0.35368794202804565,
|
| 195 |
+
"learning_rate": 8.43267979988576e-05,
|
| 196 |
+
"entropy": 0.6942046850919723,
|
| 197 |
+
"num_tokens": 4727422.0,
|
| 198 |
+
"mean_token_accuracy": 0.8079935729503631,
|
| 199 |
+
"epoch": 0.32,
|
| 200 |
+
"step": 200
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"loss": 0.6868,
|
| 204 |
+
"grad_norm": 0.36108162999153137,
|
| 205 |
+
"learning_rate": 8.229026898246885e-05,
|
| 206 |
+
"entropy": 0.6922369688749314,
|
| 207 |
+
"num_tokens": 4970679.0,
|
| 208 |
+
"mean_token_accuracy": 0.8092731922864914,
|
| 209 |
+
"epoch": 0.336,
|
| 210 |
+
"step": 210
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"loss": 0.6775,
|
| 214 |
+
"grad_norm": 0.41774165630340576,
|
| 215 |
+
"learning_rate": 8.015737311225172e-05,
|
| 216 |
+
"entropy": 0.6763984620571136,
|
| 217 |
+
"num_tokens": 5205857.0,
|
| 218 |
+
"mean_token_accuracy": 0.8115724682807922,
|
| 219 |
+
"epoch": 0.352,
|
| 220 |
+
"step": 220
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"loss": 0.6551,
|
| 224 |
+
"grad_norm": 0.3897970914840698,
|
| 225 |
+
"learning_rate": 7.793447578778427e-05,
|
| 226 |
+
"entropy": 0.6601776152849197,
|
| 227 |
+
"num_tokens": 5440774.0,
|
| 228 |
+
"mean_token_accuracy": 0.8163723379373551,
|
| 229 |
+
"epoch": 0.368,
|
| 230 |
+
"step": 230
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"loss": 0.694,
|
| 234 |
+
"grad_norm": 0.4008243978023529,
|
| 235 |
+
"learning_rate": 7.56282110083577e-05,
|
| 236 |
+
"entropy": 0.7006691724061966,
|
| 237 |
+
"num_tokens": 5681945.0,
|
| 238 |
+
"mean_token_accuracy": 0.8072743356227875,
|
| 239 |
+
"epoch": 0.384,
|
| 240 |
+
"step": 240
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"loss": 0.6635,
|
| 244 |
+
"grad_norm": 0.38448140025138855,
|
| 245 |
+
"learning_rate": 7.324546157451508e-05,
|
| 246 |
+
"entropy": 0.6688638418912888,
|
| 247 |
+
"num_tokens": 5911653.0,
|
| 248 |
+
"mean_token_accuracy": 0.8145491361618042,
|
| 249 |
+
"epoch": 0.4,
|
| 250 |
+
"step": 250
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"loss": 0.6778,
|
| 254 |
+
"grad_norm": 0.3691696226596832,
|
| 255 |
+
"learning_rate": 7.079333854706938e-05,
|
| 256 |
+
"entropy": 0.6748764783143997,
|
| 257 |
+
"num_tokens": 6153386.0,
|
| 258 |
+
"mean_token_accuracy": 0.8123239934444427,
|
| 259 |
+
"epoch": 0.416,
|
| 260 |
+
"step": 260
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"loss": 0.6931,
|
| 264 |
+
"grad_norm": 0.34841006994247437,
|
| 265 |
+
"learning_rate": 6.827916002490304e-05,
|
| 266 |
+
"entropy": 0.698054638504982,
|
| 267 |
+
"num_tokens": 6392270.0,
|
| 268 |
+
"mean_token_accuracy": 0.8090349048376083,
|
| 269 |
+
"epoch": 0.432,
|
| 270 |
+
"step": 270
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"loss": 0.6833,
|
| 274 |
+
"grad_norm": 0.36822986602783203,
|
| 275 |
+
"learning_rate": 6.571042930488474e-05,
|
| 276 |
+
"entropy": 0.6903767049312591,
|
| 277 |
+
"num_tokens": 6628878.0,
|
| 278 |
+
"mean_token_accuracy": 0.8101357609033585,
|
| 279 |
+
"epoch": 0.448,
|
| 280 |
+
"step": 280
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"loss": 0.6587,
|
| 284 |
+
"grad_norm": 0.3246494233608246,
|
| 285 |
+
"learning_rate": 6.309481248908235e-05,
|
| 286 |
+
"entropy": 0.6623686283826828,
|
| 287 |
+
"num_tokens": 6866329.0,
|
| 288 |
+
"mean_token_accuracy": 0.8156650364398956,
|
| 289 |
+
"epoch": 0.464,
|
| 290 |
+
"step": 290
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"loss": 0.6773,
|
| 294 |
+
"grad_norm": 0.3834746778011322,
|
| 295 |
+
"learning_rate": 6.044011560610153e-05,
|
| 296 |
+
"entropy": 0.6803743422031403,
|
| 297 |
+
"num_tokens": 7096086.0,
|
| 298 |
+
"mean_token_accuracy": 0.8116874188184738,
|
| 299 |
+
"epoch": 0.48,
|
| 300 |
+
"step": 300
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"loss": 0.6648,
|
| 304 |
+
"grad_norm": 0.33701953291893005,
|
| 305 |
+
"learning_rate": 5.775426131482811e-05,
|
| 306 |
+
"entropy": 0.6692022681236267,
|
| 307 |
+
"num_tokens": 7332753.0,
|
| 308 |
+
"mean_token_accuracy": 0.8145207434892654,
|
| 309 |
+
"epoch": 0.496,
|
| 310 |
+
"step": 310
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"loss": 0.6334,
|
| 314 |
+
"grad_norm": 0.3483853340148926,
|
| 315 |
+
"learning_rate": 5.5045265260100234e-05,
|
| 316 |
+
"entropy": 0.6391948401927948,
|
| 317 |
+
"num_tokens": 7562764.0,
|
| 318 |
+
"mean_token_accuracy": 0.8210806369781494,
|
| 319 |
+
"epoch": 0.512,
|
| 320 |
+
"step": 320
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"loss": 0.6424,
|
| 324 |
+
"grad_norm": 0.2986826002597809,
|
| 325 |
+
"learning_rate": 5.232121215087369e-05,
|
| 326 |
+
"entropy": 0.6503159761428833,
|
| 327 |
+
"num_tokens": 7810030.0,
|
| 328 |
+
"mean_token_accuracy": 0.8197412937879562,
|
| 329 |
+
"epoch": 0.528,
|
| 330 |
+
"step": 330
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"loss": 0.6784,
|
| 334 |
+
"grad_norm": 0.3261159360408783,
|
| 335 |
+
"learning_rate": 4.959023163227284e-05,
|
| 336 |
+
"entropy": 0.6812007486820221,
|
| 337 |
+
"num_tokens": 8046816.0,
|
| 338 |
+
"mean_token_accuracy": 0.8112091690301895,
|
| 339 |
+
"epoch": 0.544,
|
| 340 |
+
"step": 340
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"loss": 0.6711,
|
| 344 |
+
"grad_norm": 0.37481242418289185,
|
| 345 |
+
"learning_rate": 4.6860474023534335e-05,
|
| 346 |
+
"entropy": 0.6745136559009552,
|
| 347 |
+
"num_tokens": 8279743.0,
|
| 348 |
+
"mean_token_accuracy": 0.8127087384462357,
|
| 349 |
+
"epoch": 0.56,
|
| 350 |
+
"step": 350
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"loss": 0.6573,
|
| 354 |
+
"grad_norm": 0.38329869508743286,
|
| 355 |
+
"learning_rate": 4.4140085994251136e-05,
|
| 356 |
+
"entropy": 0.657204994559288,
|
| 357 |
+
"num_tokens": 8513323.0,
|
| 358 |
+
"mean_token_accuracy": 0.8160663604736328,
|
| 359 |
+
"epoch": 0.576,
|
| 360 |
+
"step": 360
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"loss": 0.6356,
|
| 364 |
+
"grad_norm": 0.4051035940647125,
|
| 365 |
+
"learning_rate": 4.143718625150854e-05,
|
| 366 |
+
"entropy": 0.6373627185821533,
|
| 367 |
+
"num_tokens": 8744995.0,
|
| 368 |
+
"mean_token_accuracy": 0.8212027668952941,
|
| 369 |
+
"epoch": 0.592,
|
| 370 |
+
"step": 370
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"loss": 0.6721,
|
| 374 |
+
"grad_norm": 0.34723058342933655,
|
| 375 |
+
"learning_rate": 3.875984131047135e-05,
|
| 376 |
+
"entropy": 0.6792070597410202,
|
| 377 |
+
"num_tokens": 8976168.0,
|
| 378 |
+
"mean_token_accuracy": 0.8130046099424362,
|
| 379 |
+
"epoch": 0.608,
|
| 380 |
+
"step": 380
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"loss": 0.6579,
|
| 384 |
+
"grad_norm": 0.3466307520866394,
|
| 385 |
+
"learning_rate": 3.6116041420732103e-05,
|
| 386 |
+
"entropy": 0.6591042011976243,
|
| 387 |
+
"num_tokens": 9217079.0,
|
| 388 |
+
"mean_token_accuracy": 0.8165153503417969,
|
| 389 |
+
"epoch": 0.624,
|
| 390 |
+
"step": 390
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"loss": 0.655,
|
| 394 |
+
"grad_norm": 0.3664017915725708,
|
| 395 |
+
"learning_rate": 3.3513676720265934e-05,
|
| 396 |
+
"entropy": 0.6642171382904053,
|
| 397 |
+
"num_tokens": 9448775.0,
|
| 398 |
+
"mean_token_accuracy": 0.8166080981492996,
|
| 399 |
+
"epoch": 0.64,
|
| 400 |
+
"step": 400
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"loss": 0.6423,
|
| 404 |
+
"grad_norm": 0.35895591974258423,
|
| 405 |
+
"learning_rate": 3.0960513688157754e-05,
|
| 406 |
+
"entropy": 0.6416015654802323,
|
| 407 |
+
"num_tokens": 9685315.0,
|
| 408 |
+
"mean_token_accuracy": 0.8195697069168091,
|
| 409 |
+
"epoch": 0.656,
|
| 410 |
+
"step": 410
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"loss": 0.6279,
|
| 414 |
+
"grad_norm": 0.3837035894393921,
|
| 415 |
+
"learning_rate": 2.846417196637613e-05,
|
| 416 |
+
"entropy": 0.6338138192892074,
|
| 417 |
+
"num_tokens": 9922050.0,
|
| 418 |
+
"mean_token_accuracy": 0.8214808911085129,
|
| 419 |
+
"epoch": 0.672,
|
| 420 |
+
"step": 420
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"loss": 0.6638,
|
| 424 |
+
"grad_norm": 0.3423479199409485,
|
| 425 |
+
"learning_rate": 2.603210161976687e-05,
|
| 426 |
+
"entropy": 0.6659493237733841,
|
| 427 |
+
"num_tokens": 10152825.0,
|
| 428 |
+
"mean_token_accuracy": 0.8158191829919815,
|
| 429 |
+
"epoch": 0.688,
|
| 430 |
+
"step": 430
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"loss": 0.6551,
|
| 434 |
+
"grad_norm": 0.3698706328868866,
|
| 435 |
+
"learning_rate": 2.3671560902131447e-05,
|
| 436 |
+
"entropy": 0.654210501909256,
|
| 437 |
+
"num_tokens": 10389209.0,
|
| 438 |
+
"mean_token_accuracy": 0.8156277418136597,
|
| 439 |
+
"epoch": 0.704,
|
| 440 |
+
"step": 440
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"loss": 0.6283,
|
| 444 |
+
"grad_norm": 0.38697871565818787,
|
| 445 |
+
"learning_rate": 2.1389594594744864e-05,
|
| 446 |
+
"entropy": 0.6335553735494613,
|
| 447 |
+
"num_tokens": 10627038.0,
|
| 448 |
+
"mean_token_accuracy": 0.8218480378389359,
|
| 449 |
+
"epoch": 0.72,
|
| 450 |
+
"step": 450
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"loss": 0.6455,
|
| 454 |
+
"grad_norm": 0.33989399671554565,
|
| 455 |
+
"learning_rate": 1.9193012981959475e-05,
|
| 456 |
+
"entropy": 0.6531029403209686,
|
| 457 |
+
"num_tokens": 10865908.0,
|
| 458 |
+
"mean_token_accuracy": 0.8178681671619416,
|
| 459 |
+
"epoch": 0.736,
|
| 460 |
+
"step": 460
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"loss": 0.6205,
|
| 464 |
+
"grad_norm": 0.37868356704711914,
|
| 465 |
+
"learning_rate": 1.7088371526639853e-05,
|
| 466 |
+
"entropy": 0.6222113043069839,
|
| 467 |
+
"num_tokens": 11105057.0,
|
| 468 |
+
"mean_token_accuracy": 0.8235622078180314,
|
| 469 |
+
"epoch": 0.752,
|
| 470 |
+
"step": 470
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"loss": 0.6771,
|
| 474 |
+
"grad_norm": 0.35013335943222046,
|
| 475 |
+
"learning_rate": 1.508195130608504e-05,
|
| 476 |
+
"entropy": 0.6883874475955963,
|
| 477 |
+
"num_tokens": 11347042.0,
|
| 478 |
+
"mean_token_accuracy": 0.8107833832502365,
|
| 479 |
+
"epoch": 0.768,
|
| 480 |
+
"step": 480
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"loss": 0.6578,
|
| 484 |
+
"grad_norm": 0.3570718467235565,
|
| 485 |
+
"learning_rate": 1.3179740266825253e-05,
|
| 486 |
+
"entropy": 0.6517588019371032,
|
| 487 |
+
"num_tokens": 11585916.0,
|
| 488 |
+
"mean_token_accuracy": 0.8149342775344849,
|
| 489 |
+
"epoch": 0.784,
|
| 490 |
+
"step": 490
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"loss": 0.6338,
|
| 494 |
+
"grad_norm": 0.3437284529209137,
|
| 495 |
+
"learning_rate": 1.1387415354235887e-05,
|
| 496 |
+
"entropy": 0.6454322338104248,
|
| 497 |
+
"num_tokens": 11824489.0,
|
| 498 |
+
"mean_token_accuracy": 0.8217582374811172,
|
| 499 |
+
"epoch": 0.8,
|
| 500 |
+
"step": 500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"loss": 0.6403,
|
| 504 |
+
"grad_norm": 0.3438985347747803,
|
| 505 |
+
"learning_rate": 9.710325570301348e-06,
|
| 506 |
+
"entropy": 0.6410525500774383,
|
| 507 |
+
"num_tokens": 12062163.0,
|
| 508 |
+
"mean_token_accuracy": 0.8210669010877609,
|
| 509 |
+
"epoch": 0.816,
|
| 510 |
+
"step": 510
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"loss": 0.6843,
|
| 514 |
+
"grad_norm": 0.3453381359577179,
|
| 515 |
+
"learning_rate": 8.153476010090788e-06,
|
| 516 |
+
"entropy": 0.6865164041519165,
|
| 517 |
+
"num_tokens": 12302722.0,
|
| 518 |
+
"mean_token_accuracy": 0.8096443176269531,
|
| 519 |
+
"epoch": 0.832,
|
| 520 |
+
"step": 520
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"loss": 0.6299,
|
| 524 |
+
"grad_norm": 0.3145497143268585,
|
| 525 |
+
"learning_rate": 6.721512924587242e-06,
|
| 526 |
+
"entropy": 0.6282441258430481,
|
| 527 |
+
"num_tokens": 12542828.0,
|
| 528 |
+
"mean_token_accuracy": 0.8224331319332123,
|
| 529 |
+
"epoch": 0.848,
|
| 530 |
+
"step": 530
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"loss": 0.6521,
|
| 534 |
+
"grad_norm": 0.7085707783699036,
|
| 535 |
+
"learning_rate": 5.418709854448489e-06,
|
| 536 |
+
"entropy": 0.658638134598732,
|
| 537 |
+
"num_tokens": 12778750.0,
|
| 538 |
+
"mean_token_accuracy": 0.8189221560955048,
|
| 539 |
+
"epoch": 0.864,
|
| 540 |
+
"step": 540
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"loss": 0.6516,
|
| 544 |
+
"grad_norm": 0.3605988025665283,
|
| 545 |
+
"learning_rate": 4.248954876082195e-06,
|
| 546 |
+
"entropy": 0.6549220532178879,
|
| 547 |
+
"num_tokens": 13017993.0,
|
| 548 |
+
"mean_token_accuracy": 0.8162762761116028,
|
| 549 |
+
"epoch": 0.88,
|
| 550 |
+
"step": 550
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"loss": 0.6227,
|
| 554 |
+
"grad_norm": 0.3576490581035614,
|
| 555 |
+
"learning_rate": 3.2157389980976783e-06,
|
| 556 |
+
"entropy": 0.6235672175884247,
|
| 557 |
+
"num_tokens": 13248686.0,
|
| 558 |
+
"mean_token_accuracy": 0.8248269021511078,
|
| 559 |
+
"epoch": 0.896,
|
| 560 |
+
"step": 560
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"loss": 0.6533,
|
| 564 |
+
"grad_norm": 0.35294073820114136,
|
| 565 |
+
"learning_rate": 2.322145742764181e-06,
|
| 566 |
+
"entropy": 0.6572938799858093,
|
| 567 |
+
"num_tokens": 13479101.0,
|
| 568 |
+
"mean_token_accuracy": 0.8174732059240342,
|
| 569 |
+
"epoch": 0.912,
|
| 570 |
+
"step": 570
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"loss": 0.6294,
|
| 574 |
+
"grad_norm": 0.3723297417163849,
|
| 575 |
+
"learning_rate": 1.5708419435684462e-06,
|
| 576 |
+
"entropy": 0.6358791559934616,
|
| 577 |
+
"num_tokens": 13711722.0,
|
| 578 |
+
"mean_token_accuracy": 0.8224119067192077,
|
| 579 |
+
"epoch": 0.928,
|
| 580 |
+
"step": 580
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"loss": 0.6359,
|
| 584 |
+
"grad_norm": 0.3650200068950653,
|
| 585 |
+
"learning_rate": 9.640697863354698e-07,
|
| 586 |
+
"entropy": 0.6374465972185135,
|
| 587 |
+
"num_tokens": 13947211.0,
|
| 588 |
+
"mean_token_accuracy": 0.821335780620575,
|
| 589 |
+
"epoch": 0.944,
|
| 590 |
+
"step": 590
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"loss": 0.6595,
|
| 594 |
+
"grad_norm": 0.3581603169441223,
|
| 595 |
+
"learning_rate": 5.036401176647332e-07,
|
| 596 |
+
"entropy": 0.6631550431251526,
|
| 597 |
+
"num_tokens": 14186581.0,
|
| 598 |
+
"mean_token_accuracy": 0.8151085019111634,
|
| 599 |
+
"epoch": 0.96,
|
| 600 |
+
"step": 600
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"loss": 0.6501,
|
| 604 |
+
"grad_norm": 0.36077970266342163,
|
| 605 |
+
"learning_rate": 1.9092704065222788e-07,
|
| 606 |
+
"entropy": 0.6615756928920746,
|
| 607 |
+
"num_tokens": 14416922.0,
|
| 608 |
+
"mean_token_accuracy": 0.8183146119117737,
|
| 609 |
+
"epoch": 0.976,
|
| 610 |
+
"step": 610
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"loss": 0.6589,
|
| 614 |
+
"grad_norm": 0.3554404377937317,
|
| 615 |
+
"learning_rate": 2.6863814026734102e-08,
|
| 616 |
+
"entropy": 0.6655615210533142,
|
| 617 |
+
"num_tokens": 14652110.0,
|
| 618 |
+
"mean_token_accuracy": 0.8152088820934296,
|
| 619 |
+
"epoch": 0.992,
|
| 620 |
+
"step": 620
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"train_runtime": 6311.5248,
|
| 624 |
+
"train_samples_per_second": 3.167,
|
| 625 |
+
"train_steps_per_second": 0.099,
|
| 626 |
+
"total_flos": 6.939581601096253e+18,
|
| 627 |
+
"train_loss": 0.7419871192932129,
|
| 628 |
+
"entropy": 0.6707779943943024,
|
| 629 |
+
"num_tokens": 14764501.0,
|
| 630 |
+
"mean_token_accuracy": 0.8151742398738862,
|
| 631 |
+
"epoch": 1.0,
|
| 632 |
+
"step": 625
|
| 633 |
+
}
|
| 634 |
+
]
|