Commit
•
5f101ea
1
Parent(s):
06bbcd3
End of training
Browse files- all_results.json +7 -0
- train_results.json +7 -0
- trainer_state.json +322 -0
all_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.96,
|
| 3 |
+
"train_loss": 1.711962864949153,
|
| 4 |
+
"train_runtime": 11907.5247,
|
| 5 |
+
"train_samples_per_second": 26.874,
|
| 6 |
+
"train_steps_per_second": 0.279
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.96,
|
| 3 |
+
"train_loss": 1.711962864949153,
|
| 4 |
+
"train_runtime": 11907.5247,
|
| 5 |
+
"train_samples_per_second": 26.874,
|
| 6 |
+
"train_steps_per_second": 0.279
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,322 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 15.961630695443645,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 3328,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"eval_f1": 0.8988,
|
| 14 |
+
"eval_gen_len": 29.90909090909091,
|
| 15 |
+
"eval_loss": 1.8075228929519653,
|
| 16 |
+
"eval_precision": 0.901,
|
| 17 |
+
"eval_recall": 0.897,
|
| 18 |
+
"eval_rouge1": 0.411,
|
| 19 |
+
"eval_rouge2": 0.1689,
|
| 20 |
+
"eval_rougeL": 0.3152,
|
| 21 |
+
"eval_rougeLsum": 0.3155,
|
| 22 |
+
"eval_runtime": 182.3495,
|
| 23 |
+
"eval_samples_per_second": 6.032,
|
| 24 |
+
"eval_steps_per_second": 0.378,
|
| 25 |
+
"step": 208
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 2.0,
|
| 29 |
+
"eval_f1": 0.904,
|
| 30 |
+
"eval_gen_len": 29.907272727272726,
|
| 31 |
+
"eval_loss": 1.7311877012252808,
|
| 32 |
+
"eval_precision": 0.9059,
|
| 33 |
+
"eval_recall": 0.9024,
|
| 34 |
+
"eval_rouge1": 0.4379,
|
| 35 |
+
"eval_rouge2": 0.1893,
|
| 36 |
+
"eval_rougeL": 0.3442,
|
| 37 |
+
"eval_rougeLsum": 0.3446,
|
| 38 |
+
"eval_runtime": 160.7204,
|
| 39 |
+
"eval_samples_per_second": 6.844,
|
| 40 |
+
"eval_steps_per_second": 0.429,
|
| 41 |
+
"step": 417
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 2.4,
|
| 45 |
+
"learning_rate": 1.699519230769231e-05,
|
| 46 |
+
"loss": 2.0112,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 3.0,
|
| 51 |
+
"eval_f1": 0.9055,
|
| 52 |
+
"eval_gen_len": 30.017272727272726,
|
| 53 |
+
"eval_loss": 1.6986640691757202,
|
| 54 |
+
"eval_precision": 0.9075,
|
| 55 |
+
"eval_recall": 0.9039,
|
| 56 |
+
"eval_rouge1": 0.4475,
|
| 57 |
+
"eval_rouge2": 0.1978,
|
| 58 |
+
"eval_rougeL": 0.352,
|
| 59 |
+
"eval_rougeLsum": 0.3525,
|
| 60 |
+
"eval_runtime": 161.6717,
|
| 61 |
+
"eval_samples_per_second": 6.804,
|
| 62 |
+
"eval_steps_per_second": 0.427,
|
| 63 |
+
"step": 625
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 4.0,
|
| 67 |
+
"eval_f1": 0.9063,
|
| 68 |
+
"eval_gen_len": 30.061818181818182,
|
| 69 |
+
"eval_loss": 1.676792025566101,
|
| 70 |
+
"eval_precision": 0.9082,
|
| 71 |
+
"eval_recall": 0.9047,
|
| 72 |
+
"eval_rouge1": 0.4514,
|
| 73 |
+
"eval_rouge2": 0.1981,
|
| 74 |
+
"eval_rougeL": 0.357,
|
| 75 |
+
"eval_rougeLsum": 0.3573,
|
| 76 |
+
"eval_runtime": 160.8885,
|
| 77 |
+
"eval_samples_per_second": 6.837,
|
| 78 |
+
"eval_steps_per_second": 0.429,
|
| 79 |
+
"step": 834
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 4.8,
|
| 83 |
+
"learning_rate": 1.3990384615384615e-05,
|
| 84 |
+
"loss": 1.7647,
|
| 85 |
+
"step": 1000
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 5.0,
|
| 89 |
+
"eval_f1": 0.9068,
|
| 90 |
+
"eval_gen_len": 30.326363636363638,
|
| 91 |
+
"eval_loss": 1.661742091178894,
|
| 92 |
+
"eval_precision": 0.9084,
|
| 93 |
+
"eval_recall": 0.9055,
|
| 94 |
+
"eval_rouge1": 0.4537,
|
| 95 |
+
"eval_rouge2": 0.2003,
|
| 96 |
+
"eval_rougeL": 0.3592,
|
| 97 |
+
"eval_rougeLsum": 0.3595,
|
| 98 |
+
"eval_runtime": 162.5609,
|
| 99 |
+
"eval_samples_per_second": 6.767,
|
| 100 |
+
"eval_steps_per_second": 0.424,
|
| 101 |
+
"step": 1042
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 6.0,
|
| 105 |
+
"eval_f1": 0.9072,
|
| 106 |
+
"eval_gen_len": 30.082727272727272,
|
| 107 |
+
"eval_loss": 1.6502095460891724,
|
| 108 |
+
"eval_precision": 0.9089,
|
| 109 |
+
"eval_recall": 0.9057,
|
| 110 |
+
"eval_rouge1": 0.4554,
|
| 111 |
+
"eval_rouge2": 0.2021,
|
| 112 |
+
"eval_rougeL": 0.3607,
|
| 113 |
+
"eval_rougeLsum": 0.361,
|
| 114 |
+
"eval_runtime": 159.1789,
|
| 115 |
+
"eval_samples_per_second": 6.91,
|
| 116 |
+
"eval_steps_per_second": 0.433,
|
| 117 |
+
"step": 1251
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 7.0,
|
| 121 |
+
"eval_f1": 0.908,
|
| 122 |
+
"eval_gen_len": 30.021818181818183,
|
| 123 |
+
"eval_loss": 1.6416178941726685,
|
| 124 |
+
"eval_precision": 0.9099,
|
| 125 |
+
"eval_recall": 0.9064,
|
| 126 |
+
"eval_rouge1": 0.4592,
|
| 127 |
+
"eval_rouge2": 0.2052,
|
| 128 |
+
"eval_rougeL": 0.3639,
|
| 129 |
+
"eval_rougeLsum": 0.3641,
|
| 130 |
+
"eval_runtime": 156.6401,
|
| 131 |
+
"eval_samples_per_second": 7.022,
|
| 132 |
+
"eval_steps_per_second": 0.441,
|
| 133 |
+
"step": 1459
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 7.19,
|
| 137 |
+
"learning_rate": 1.0985576923076924e-05,
|
| 138 |
+
"loss": 1.6948,
|
| 139 |
+
"step": 1500
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 8.0,
|
| 143 |
+
"eval_f1": 0.9081,
|
| 144 |
+
"eval_gen_len": 30.78272727272727,
|
| 145 |
+
"eval_loss": 1.6360372304916382,
|
| 146 |
+
"eval_precision": 0.909,
|
| 147 |
+
"eval_recall": 0.9074,
|
| 148 |
+
"eval_rouge1": 0.4612,
|
| 149 |
+
"eval_rouge2": 0.2054,
|
| 150 |
+
"eval_rougeL": 0.3649,
|
| 151 |
+
"eval_rougeLsum": 0.365,
|
| 152 |
+
"eval_runtime": 160.7238,
|
| 153 |
+
"eval_samples_per_second": 6.844,
|
| 154 |
+
"eval_steps_per_second": 0.429,
|
| 155 |
+
"step": 1668
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 9.0,
|
| 159 |
+
"eval_f1": 0.9083,
|
| 160 |
+
"eval_gen_len": 30.62909090909091,
|
| 161 |
+
"eval_loss": 1.6301532983779907,
|
| 162 |
+
"eval_precision": 0.9095,
|
| 163 |
+
"eval_recall": 0.9074,
|
| 164 |
+
"eval_rouge1": 0.4621,
|
| 165 |
+
"eval_rouge2": 0.2062,
|
| 166 |
+
"eval_rougeL": 0.3645,
|
| 167 |
+
"eval_rougeLsum": 0.3647,
|
| 168 |
+
"eval_runtime": 159.8662,
|
| 169 |
+
"eval_samples_per_second": 6.881,
|
| 170 |
+
"eval_steps_per_second": 0.432,
|
| 171 |
+
"step": 1876
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 9.59,
|
| 175 |
+
"learning_rate": 7.980769230769232e-06,
|
| 176 |
+
"loss": 1.6501,
|
| 177 |
+
"step": 2000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 10.0,
|
| 181 |
+
"eval_f1": 0.9083,
|
| 182 |
+
"eval_gen_len": 30.48181818181818,
|
| 183 |
+
"eval_loss": 1.6264721155166626,
|
| 184 |
+
"eval_precision": 0.9095,
|
| 185 |
+
"eval_recall": 0.9073,
|
| 186 |
+
"eval_rouge1": 0.4606,
|
| 187 |
+
"eval_rouge2": 0.2051,
|
| 188 |
+
"eval_rougeL": 0.3651,
|
| 189 |
+
"eval_rougeLsum": 0.3655,
|
| 190 |
+
"eval_runtime": 157.2711,
|
| 191 |
+
"eval_samples_per_second": 6.994,
|
| 192 |
+
"eval_steps_per_second": 0.439,
|
| 193 |
+
"step": 2085
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 11.0,
|
| 197 |
+
"eval_f1": 0.9087,
|
| 198 |
+
"eval_gen_len": 30.806363636363635,
|
| 199 |
+
"eval_loss": 1.6229554414749146,
|
| 200 |
+
"eval_precision": 0.9097,
|
| 201 |
+
"eval_recall": 0.908,
|
| 202 |
+
"eval_rouge1": 0.4625,
|
| 203 |
+
"eval_rouge2": 0.2073,
|
| 204 |
+
"eval_rougeL": 0.3658,
|
| 205 |
+
"eval_rougeLsum": 0.366,
|
| 206 |
+
"eval_runtime": 157.1263,
|
| 207 |
+
"eval_samples_per_second": 7.001,
|
| 208 |
+
"eval_steps_per_second": 0.439,
|
| 209 |
+
"step": 2293
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 11.99,
|
| 213 |
+
"learning_rate": 4.975961538461539e-06,
|
| 214 |
+
"loss": 1.6222,
|
| 215 |
+
"step": 2500
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 12.0,
|
| 219 |
+
"eval_f1": 0.909,
|
| 220 |
+
"eval_gen_len": 30.55272727272727,
|
| 221 |
+
"eval_loss": 1.6204967498779297,
|
| 222 |
+
"eval_precision": 0.9103,
|
| 223 |
+
"eval_recall": 0.9081,
|
| 224 |
+
"eval_rouge1": 0.4644,
|
| 225 |
+
"eval_rouge2": 0.2082,
|
| 226 |
+
"eval_rougeL": 0.3674,
|
| 227 |
+
"eval_rougeLsum": 0.3679,
|
| 228 |
+
"eval_runtime": 155.7494,
|
| 229 |
+
"eval_samples_per_second": 7.063,
|
| 230 |
+
"eval_steps_per_second": 0.443,
|
| 231 |
+
"step": 2502
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 13.0,
|
| 235 |
+
"eval_f1": 0.909,
|
| 236 |
+
"eval_gen_len": 30.805454545454545,
|
| 237 |
+
"eval_loss": 1.618751883506775,
|
| 238 |
+
"eval_precision": 0.9101,
|
| 239 |
+
"eval_recall": 0.9083,
|
| 240 |
+
"eval_rouge1": 0.4648,
|
| 241 |
+
"eval_rouge2": 0.2087,
|
| 242 |
+
"eval_rougeL": 0.3681,
|
| 243 |
+
"eval_rougeLsum": 0.3683,
|
| 244 |
+
"eval_runtime": 158.59,
|
| 245 |
+
"eval_samples_per_second": 6.936,
|
| 246 |
+
"eval_steps_per_second": 0.435,
|
| 247 |
+
"step": 2710
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 14.0,
|
| 251 |
+
"eval_f1": 0.9093,
|
| 252 |
+
"eval_gen_len": 30.670909090909092,
|
| 253 |
+
"eval_loss": 1.6172130107879639,
|
| 254 |
+
"eval_precision": 0.9104,
|
| 255 |
+
"eval_recall": 0.9084,
|
| 256 |
+
"eval_rouge1": 0.4654,
|
| 257 |
+
"eval_rouge2": 0.2097,
|
| 258 |
+
"eval_rougeL": 0.3685,
|
| 259 |
+
"eval_rougeLsum": 0.3689,
|
| 260 |
+
"eval_runtime": 159.3516,
|
| 261 |
+
"eval_samples_per_second": 6.903,
|
| 262 |
+
"eval_steps_per_second": 0.433,
|
| 263 |
+
"step": 2919
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 14.39,
|
| 267 |
+
"learning_rate": 1.971153846153846e-06,
|
| 268 |
+
"loss": 1.6048,
|
| 269 |
+
"step": 3000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 15.0,
|
| 273 |
+
"eval_f1": 0.9093,
|
| 274 |
+
"eval_gen_len": 30.630909090909093,
|
| 275 |
+
"eval_loss": 1.6168792247772217,
|
| 276 |
+
"eval_precision": 0.9104,
|
| 277 |
+
"eval_recall": 0.9084,
|
| 278 |
+
"eval_rouge1": 0.465,
|
| 279 |
+
"eval_rouge2": 0.21,
|
| 280 |
+
"eval_rougeL": 0.3693,
|
| 281 |
+
"eval_rougeLsum": 0.3697,
|
| 282 |
+
"eval_runtime": 158.4169,
|
| 283 |
+
"eval_samples_per_second": 6.944,
|
| 284 |
+
"eval_steps_per_second": 0.436,
|
| 285 |
+
"step": 3127
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 15.96,
|
| 289 |
+
"eval_f1": 0.9091,
|
| 290 |
+
"eval_gen_len": 30.619090909090907,
|
| 291 |
+
"eval_loss": 1.6166560649871826,
|
| 292 |
+
"eval_precision": 0.9102,
|
| 293 |
+
"eval_recall": 0.9083,
|
| 294 |
+
"eval_rouge1": 0.4649,
|
| 295 |
+
"eval_rouge2": 0.2096,
|
| 296 |
+
"eval_rougeL": 0.3686,
|
| 297 |
+
"eval_rougeLsum": 0.3688,
|
| 298 |
+
"eval_runtime": 157.7317,
|
| 299 |
+
"eval_samples_per_second": 6.974,
|
| 300 |
+
"eval_steps_per_second": 0.437,
|
| 301 |
+
"step": 3328
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 15.96,
|
| 305 |
+
"step": 3328,
|
| 306 |
+
"total_flos": 4.6122414295331635e+17,
|
| 307 |
+
"train_loss": 1.711962864949153,
|
| 308 |
+
"train_runtime": 11907.5247,
|
| 309 |
+
"train_samples_per_second": 26.874,
|
| 310 |
+
"train_steps_per_second": 0.279
|
| 311 |
+
}
|
| 312 |
+
],
|
| 313 |
+
"logging_steps": 500,
|
| 314 |
+
"max_steps": 3328,
|
| 315 |
+
"num_input_tokens_seen": 0,
|
| 316 |
+
"num_train_epochs": 16,
|
| 317 |
+
"save_steps": 500,
|
| 318 |
+
"total_flos": 4.6122414295331635e+17,
|
| 319 |
+
"train_batch_size": 24,
|
| 320 |
+
"trial_name": null,
|
| 321 |
+
"trial_params": null
|
| 322 |
+
}
|