Training in progress, epoch 1
Browse files- all_results.json +22 -22
- eval_results.json +9 -9
- generated_predictions.txt +0 -0
- predict_results.json +9 -9
- pytorch_model.bin +1 -1
- runs/Mar22_12-27-47_kogecha/events.out.tfevents.1711082385.kogecha +3 -0
- runs/Mar26_19-29-34_kogecha/events.out.tfevents.1711448988.kogecha +3 -0
- train_results.json +4 -4
- trainer_state.json +137 -137
- training_args.bin +1 -1
all_results.json
CHANGED
|
@@ -1,28 +1,28 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"eval_gen_len":
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_rouge1":
|
| 6 |
-
"eval_rouge2":
|
| 7 |
-
"eval_rougeL":
|
| 8 |
-
"eval_rougeLsum":
|
| 9 |
-
"eval_runtime":
|
| 10 |
"eval_samples": 632,
|
| 11 |
-
"eval_samples_per_second": 1.
|
| 12 |
-
"eval_steps_per_second": 1.
|
| 13 |
-
"predict_gen_len":
|
| 14 |
-
"predict_loss":
|
| 15 |
-
"predict_rouge1":
|
| 16 |
-
"predict_rouge2": 2.
|
| 17 |
-
"predict_rougeL": 10.
|
| 18 |
-
"predict_rougeLsum":
|
| 19 |
-
"predict_runtime":
|
| 20 |
"predict_samples": 535,
|
| 21 |
-
"predict_samples_per_second": 1.
|
| 22 |
-
"predict_steps_per_second": 1.
|
| 23 |
-
"train_loss":
|
| 24 |
-
"train_runtime":
|
| 25 |
"train_samples": 4332,
|
| 26 |
-
"train_samples_per_second": 5.
|
| 27 |
-
"train_steps_per_second": 5.
|
| 28 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"eval_gen_len": 57.35284810126582,
|
| 4 |
+
"eval_loss": 3.128607988357544,
|
| 5 |
+
"eval_rouge1": 13.7182,
|
| 6 |
+
"eval_rouge2": 2.311,
|
| 7 |
+
"eval_rougeL": 9.1726,
|
| 8 |
+
"eval_rougeLsum": 11.5058,
|
| 9 |
+
"eval_runtime": 359.9186,
|
| 10 |
"eval_samples": 632,
|
| 11 |
+
"eval_samples_per_second": 1.756,
|
| 12 |
+
"eval_steps_per_second": 1.756,
|
| 13 |
+
"predict_gen_len": 52.28971962616822,
|
| 14 |
+
"predict_loss": 3.2279164791107178,
|
| 15 |
+
"predict_rouge1": 16.0977,
|
| 16 |
+
"predict_rouge2": 2.9966,
|
| 17 |
+
"predict_rougeL": 10.7567,
|
| 18 |
+
"predict_rougeLsum": 12.5377,
|
| 19 |
+
"predict_runtime": 273.2778,
|
| 20 |
"predict_samples": 535,
|
| 21 |
+
"predict_samples_per_second": 1.958,
|
| 22 |
+
"predict_steps_per_second": 1.958,
|
| 23 |
+
"train_loss": 0.23670565184904047,
|
| 24 |
+
"train_runtime": 3830.6771,
|
| 25 |
"train_samples": 4332,
|
| 26 |
+
"train_samples_per_second": 5.654,
|
| 27 |
+
"train_steps_per_second": 5.654
|
| 28 |
}
|
eval_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"eval_gen_len":
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_rouge1":
|
| 6 |
-
"eval_rouge2":
|
| 7 |
-
"eval_rougeL":
|
| 8 |
-
"eval_rougeLsum":
|
| 9 |
-
"eval_runtime":
|
| 10 |
"eval_samples": 632,
|
| 11 |
-
"eval_samples_per_second": 1.
|
| 12 |
-
"eval_steps_per_second": 1.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"eval_gen_len": 57.35284810126582,
|
| 4 |
+
"eval_loss": 3.128607988357544,
|
| 5 |
+
"eval_rouge1": 13.7182,
|
| 6 |
+
"eval_rouge2": 2.311,
|
| 7 |
+
"eval_rougeL": 9.1726,
|
| 8 |
+
"eval_rougeLsum": 11.5058,
|
| 9 |
+
"eval_runtime": 359.9186,
|
| 10 |
"eval_samples": 632,
|
| 11 |
+
"eval_samples_per_second": 1.756,
|
| 12 |
+
"eval_steps_per_second": 1.756
|
| 13 |
}
|
generated_predictions.txt
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
predict_results.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
-
"predict_gen_len":
|
| 3 |
-
"predict_loss":
|
| 4 |
-
"predict_rouge1":
|
| 5 |
-
"predict_rouge2": 2.
|
| 6 |
-
"predict_rougeL": 10.
|
| 7 |
-
"predict_rougeLsum":
|
| 8 |
-
"predict_runtime":
|
| 9 |
"predict_samples": 535,
|
| 10 |
-
"predict_samples_per_second": 1.
|
| 11 |
-
"predict_steps_per_second": 1.
|
| 12 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"predict_gen_len": 52.28971962616822,
|
| 3 |
+
"predict_loss": 3.2279164791107178,
|
| 4 |
+
"predict_rouge1": 16.0977,
|
| 5 |
+
"predict_rouge2": 2.9966,
|
| 6 |
+
"predict_rougeL": 10.7567,
|
| 7 |
+
"predict_rougeLsum": 12.5377,
|
| 8 |
+
"predict_runtime": 273.2778,
|
| 9 |
"predict_samples": 535,
|
| 10 |
+
"predict_samples_per_second": 1.958,
|
| 11 |
+
"predict_steps_per_second": 1.958
|
| 12 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 501807853
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:172991dc78c9918db2529f7adc381600a30e38b8aec3163b28636fb580629e59
|
| 3 |
size 501807853
|
runs/Mar22_12-27-47_kogecha/events.out.tfevents.1711082385.kogecha
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:195f4236267b24b54e5763b09fad05e35237ee01c5f4f42d56115a70237c9de8
|
| 3 |
+
size 575
|
runs/Mar26_19-29-34_kogecha/events.out.tfevents.1711448988.kogecha
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e8a3e81acffc3b7e3a1d8bede7cc9f6b98afafc216ade878e4afd75083aa7664
|
| 3 |
+
size 7434
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 4332,
|
| 6 |
-
"train_samples_per_second": 5.
|
| 7 |
-
"train_steps_per_second": 5.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"train_loss": 0.23670565184904047,
|
| 4 |
+
"train_runtime": 3830.6771,
|
| 5 |
"train_samples": 4332,
|
| 6 |
+
"train_samples_per_second": 5.654,
|
| 7 |
+
"train_steps_per_second": 5.654
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "saved/tobyoki-pairwise/bart-base-japanese/BaseModel/checkpoint-
|
| 4 |
"epoch": 5.0,
|
| 5 |
"global_step": 21660,
|
| 6 |
"is_hyper_param_search": false,
|
|
@@ -9,335 +9,335 @@
|
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 0.12,
|
| 12 |
-
"learning_rate":
|
| 13 |
-
"loss":
|
| 14 |
"step": 500
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.23,
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss":
|
| 20 |
"step": 1000
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.35,
|
| 24 |
-
"learning_rate":
|
| 25 |
-
"loss":
|
| 26 |
"step": 1500
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 0.46,
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss":
|
| 32 |
"step": 2000
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.58,
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss":
|
| 38 |
"step": 2500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.69,
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss":
|
| 44 |
"step": 3000
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.81,
|
| 48 |
-
"learning_rate":
|
| 49 |
-
"loss":
|
| 50 |
"step": 3500
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 0.92,
|
| 54 |
-
"learning_rate":
|
| 55 |
-
"loss":
|
| 56 |
"step": 4000
|
| 57 |
},
|
| 58 |
{
|
| 59 |
"epoch": 1.0,
|
| 60 |
-
"eval_gen_len":
|
| 61 |
-
"eval_loss": 2.
|
| 62 |
-
"eval_rouge1":
|
| 63 |
-
"eval_rouge2":
|
| 64 |
-
"eval_rougeL":
|
| 65 |
-
"eval_rougeLsum":
|
| 66 |
-
"eval_runtime":
|
| 67 |
-
"eval_samples_per_second":
|
| 68 |
-
"eval_steps_per_second":
|
| 69 |
"step": 4332
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 1.04,
|
| 73 |
-
"learning_rate":
|
| 74 |
-
"loss":
|
| 75 |
"step": 4500
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 1.15,
|
| 79 |
-
"learning_rate":
|
| 80 |
-
"loss":
|
| 81 |
"step": 5000
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 1.27,
|
| 85 |
-
"learning_rate":
|
| 86 |
-
"loss":
|
| 87 |
"step": 5500
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 1.39,
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss":
|
| 93 |
"step": 6000
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 1.5,
|
| 97 |
-
"learning_rate":
|
| 98 |
-
"loss":
|
| 99 |
"step": 6500
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 1.62,
|
| 103 |
-
"learning_rate":
|
| 104 |
-
"loss":
|
| 105 |
"step": 7000
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 1.73,
|
| 109 |
-
"learning_rate":
|
| 110 |
-
"loss":
|
| 111 |
"step": 7500
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 1.85,
|
| 115 |
-
"learning_rate":
|
| 116 |
-
"loss":
|
| 117 |
"step": 8000
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 1.96,
|
| 121 |
-
"learning_rate":
|
| 122 |
-
"loss":
|
| 123 |
"step": 8500
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 2.0,
|
| 127 |
-
"eval_gen_len":
|
| 128 |
-
"eval_loss":
|
| 129 |
-
"eval_rouge1":
|
| 130 |
-
"eval_rouge2":
|
| 131 |
-
"eval_rougeL":
|
| 132 |
-
"eval_rougeLsum":
|
| 133 |
-
"eval_runtime":
|
| 134 |
-
"eval_samples_per_second":
|
| 135 |
-
"eval_steps_per_second":
|
| 136 |
"step": 8664
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 2.08,
|
| 140 |
-
"learning_rate":
|
| 141 |
-
"loss":
|
| 142 |
"step": 9000
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 2.19,
|
| 146 |
-
"learning_rate":
|
| 147 |
-
"loss":
|
| 148 |
"step": 9500
|
| 149 |
},
|
| 150 |
{
|
| 151 |
"epoch": 2.31,
|
| 152 |
-
"learning_rate":
|
| 153 |
-
"loss":
|
| 154 |
"step": 10000
|
| 155 |
},
|
| 156 |
{
|
| 157 |
"epoch": 2.42,
|
| 158 |
-
"learning_rate":
|
| 159 |
-
"loss":
|
| 160 |
"step": 10500
|
| 161 |
},
|
| 162 |
{
|
| 163 |
"epoch": 2.54,
|
| 164 |
-
"learning_rate":
|
| 165 |
-
"loss":
|
| 166 |
"step": 11000
|
| 167 |
},
|
| 168 |
{
|
| 169 |
"epoch": 2.65,
|
| 170 |
-
"learning_rate":
|
| 171 |
-
"loss":
|
| 172 |
"step": 11500
|
| 173 |
},
|
| 174 |
{
|
| 175 |
"epoch": 2.77,
|
| 176 |
-
"learning_rate":
|
| 177 |
-
"loss":
|
| 178 |
"step": 12000
|
| 179 |
},
|
| 180 |
{
|
| 181 |
"epoch": 2.89,
|
| 182 |
-
"learning_rate":
|
| 183 |
-
"loss":
|
| 184 |
"step": 12500
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 3.0,
|
| 188 |
-
"eval_gen_len":
|
| 189 |
-
"eval_loss":
|
| 190 |
-
"eval_rouge1":
|
| 191 |
-
"eval_rouge2":
|
| 192 |
-
"eval_rougeL":
|
| 193 |
-
"eval_rougeLsum":
|
| 194 |
-
"eval_runtime":
|
| 195 |
-
"eval_samples_per_second":
|
| 196 |
-
"eval_steps_per_second":
|
| 197 |
"step": 12996
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 3.0,
|
| 201 |
-
"learning_rate": 1.
|
| 202 |
-
"loss":
|
| 203 |
"step": 13000
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 3.12,
|
| 207 |
-
"learning_rate": 1.
|
| 208 |
-
"loss":
|
| 209 |
"step": 13500
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 3.23,
|
| 213 |
-
"learning_rate": 1.
|
| 214 |
-
"loss":
|
| 215 |
"step": 14000
|
| 216 |
},
|
| 217 |
{
|
| 218 |
"epoch": 3.35,
|
| 219 |
-
"learning_rate":
|
| 220 |
-
"loss":
|
| 221 |
"step": 14500
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 3.46,
|
| 225 |
-
"learning_rate":
|
| 226 |
-
"loss":
|
| 227 |
"step": 15000
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 3.58,
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss":
|
| 233 |
"step": 15500
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 3.69,
|
| 237 |
-
"learning_rate":
|
| 238 |
-
"loss":
|
| 239 |
"step": 16000
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 3.81,
|
| 243 |
-
"learning_rate":
|
| 244 |
-
"loss":
|
| 245 |
"step": 16500
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 3.92,
|
| 249 |
-
"learning_rate":
|
| 250 |
-
"loss":
|
| 251 |
"step": 17000
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 4.0,
|
| 255 |
-
"eval_gen_len":
|
| 256 |
-
"eval_loss":
|
| 257 |
-
"eval_rouge1":
|
| 258 |
-
"eval_rouge2":
|
| 259 |
-
"eval_rougeL":
|
| 260 |
-
"eval_rougeLsum":
|
| 261 |
-
"eval_runtime":
|
| 262 |
-
"eval_samples_per_second": 1.
|
| 263 |
-
"eval_steps_per_second": 1.
|
| 264 |
"step": 17328
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 4.04,
|
| 268 |
-
"learning_rate":
|
| 269 |
-
"loss":
|
| 270 |
"step": 17500
|
| 271 |
},
|
| 272 |
{
|
| 273 |
"epoch": 4.16,
|
| 274 |
-
"learning_rate":
|
| 275 |
-
"loss":
|
| 276 |
"step": 18000
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 4.27,
|
| 280 |
-
"learning_rate":
|
| 281 |
-
"loss":
|
| 282 |
"step": 18500
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 4.39,
|
| 286 |
-
"learning_rate":
|
| 287 |
-
"loss":
|
| 288 |
"step": 19000
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 4.5,
|
| 292 |
-
"learning_rate":
|
| 293 |
-
"loss":
|
| 294 |
"step": 19500
|
| 295 |
},
|
| 296 |
{
|
| 297 |
"epoch": 4.62,
|
| 298 |
-
"learning_rate":
|
| 299 |
-
"loss":
|
| 300 |
"step": 20000
|
| 301 |
},
|
| 302 |
{
|
| 303 |
"epoch": 4.73,
|
| 304 |
-
"learning_rate":
|
| 305 |
-
"loss":
|
| 306 |
"step": 20500
|
| 307 |
},
|
| 308 |
{
|
| 309 |
"epoch": 4.85,
|
| 310 |
-
"learning_rate":
|
| 311 |
-
"loss":
|
| 312 |
"step": 21000
|
| 313 |
},
|
| 314 |
{
|
| 315 |
"epoch": 4.96,
|
| 316 |
-
"learning_rate":
|
| 317 |
-
"loss":
|
| 318 |
"step": 21500
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 5.0,
|
| 322 |
-
"eval_gen_len":
|
| 323 |
-
"eval_loss":
|
| 324 |
-
"eval_rouge1":
|
| 325 |
-
"eval_rouge2":
|
| 326 |
-
"eval_rougeL":
|
| 327 |
-
"eval_rougeLsum":
|
| 328 |
-
"eval_runtime":
|
| 329 |
-
"eval_samples_per_second": 1.
|
| 330 |
-
"eval_steps_per_second": 1.
|
| 331 |
"step": 21660
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 5.0,
|
| 335 |
"step": 21660,
|
| 336 |
"total_flos": 3568971203850240.0,
|
| 337 |
-
"train_loss":
|
| 338 |
-
"train_runtime":
|
| 339 |
-
"train_samples_per_second": 5.
|
| 340 |
-
"train_steps_per_second": 5.
|
| 341 |
}
|
| 342 |
],
|
| 343 |
"max_steps": 21660,
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 13.7182,
|
| 3 |
+
"best_model_checkpoint": "saved/tobyoki-pairwise/bart-base-japanese/BaseModel/checkpoint-8664",
|
| 4 |
"epoch": 5.0,
|
| 5 |
"global_step": 21660,
|
| 6 |
"is_hyper_param_search": false,
|
|
|
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 0.12,
|
| 12 |
+
"learning_rate": 4.8845798707294554e-05,
|
| 13 |
+
"loss": 2.1628,
|
| 14 |
"step": 500
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.23,
|
| 18 |
+
"learning_rate": 4.7691597414589107e-05,
|
| 19 |
+
"loss": 1.3515,
|
| 20 |
"step": 1000
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.35,
|
| 24 |
+
"learning_rate": 4.653739612188366e-05,
|
| 25 |
+
"loss": 0.9577,
|
| 26 |
"step": 1500
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 0.46,
|
| 30 |
+
"learning_rate": 4.538319482917821e-05,
|
| 31 |
+
"loss": 0.7042,
|
| 32 |
"step": 2000
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.58,
|
| 36 |
+
"learning_rate": 4.422899353647276e-05,
|
| 37 |
+
"loss": 0.5935,
|
| 38 |
"step": 2500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.69,
|
| 42 |
+
"learning_rate": 4.3074792243767315e-05,
|
| 43 |
+
"loss": 0.4658,
|
| 44 |
"step": 3000
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.81,
|
| 48 |
+
"learning_rate": 4.192059095106187e-05,
|
| 49 |
+
"loss": 0.3854,
|
| 50 |
"step": 3500
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 0.92,
|
| 54 |
+
"learning_rate": 4.076638965835642e-05,
|
| 55 |
+
"loss": 0.2994,
|
| 56 |
"step": 4000
|
| 57 |
},
|
| 58 |
{
|
| 59 |
"epoch": 1.0,
|
| 60 |
+
"eval_gen_len": 55.063291139240505,
|
| 61 |
+
"eval_loss": 2.7883288860321045,
|
| 62 |
+
"eval_rouge1": 11.1611,
|
| 63 |
+
"eval_rouge2": 1.7768,
|
| 64 |
+
"eval_rougeL": 7.5158,
|
| 65 |
+
"eval_rougeLsum": 9.6222,
|
| 66 |
+
"eval_runtime": 352.9978,
|
| 67 |
+
"eval_samples_per_second": 1.79,
|
| 68 |
+
"eval_steps_per_second": 1.79,
|
| 69 |
"step": 4332
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 1.04,
|
| 73 |
+
"learning_rate": 3.961218836565097e-05,
|
| 74 |
+
"loss": 0.2806,
|
| 75 |
"step": 4500
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 1.15,
|
| 79 |
+
"learning_rate": 3.845798707294552e-05,
|
| 80 |
+
"loss": 0.263,
|
| 81 |
"step": 5000
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 1.27,
|
| 85 |
+
"learning_rate": 3.7303785780240075e-05,
|
| 86 |
+
"loss": 0.2187,
|
| 87 |
"step": 5500
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 1.39,
|
| 91 |
+
"learning_rate": 3.614958448753463e-05,
|
| 92 |
+
"loss": 0.1761,
|
| 93 |
"step": 6000
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 1.5,
|
| 97 |
+
"learning_rate": 3.499538319482918e-05,
|
| 98 |
+
"loss": 0.1664,
|
| 99 |
"step": 6500
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 1.62,
|
| 103 |
+
"learning_rate": 3.384118190212373e-05,
|
| 104 |
+
"loss": 0.1896,
|
| 105 |
"step": 7000
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 1.73,
|
| 109 |
+
"learning_rate": 3.2686980609418284e-05,
|
| 110 |
+
"loss": 0.1627,
|
| 111 |
"step": 7500
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 1.85,
|
| 115 |
+
"learning_rate": 3.1532779316712836e-05,
|
| 116 |
+
"loss": 0.1146,
|
| 117 |
"step": 8000
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 1.96,
|
| 121 |
+
"learning_rate": 3.0378578024007388e-05,
|
| 122 |
+
"loss": 0.1513,
|
| 123 |
"step": 8500
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 2.0,
|
| 127 |
+
"eval_gen_len": 57.35284810126582,
|
| 128 |
+
"eval_loss": 3.128607988357544,
|
| 129 |
+
"eval_rouge1": 13.7182,
|
| 130 |
+
"eval_rouge2": 2.311,
|
| 131 |
+
"eval_rougeL": 9.1726,
|
| 132 |
+
"eval_rougeLsum": 11.5058,
|
| 133 |
+
"eval_runtime": 358.8793,
|
| 134 |
+
"eval_samples_per_second": 1.761,
|
| 135 |
+
"eval_steps_per_second": 1.761,
|
| 136 |
"step": 8664
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 2.08,
|
| 140 |
+
"learning_rate": 2.922437673130194e-05,
|
| 141 |
+
"loss": 0.125,
|
| 142 |
"step": 9000
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 2.19,
|
| 146 |
+
"learning_rate": 2.8070175438596492e-05,
|
| 147 |
+
"loss": 0.1028,
|
| 148 |
"step": 9500
|
| 149 |
},
|
| 150 |
{
|
| 151 |
"epoch": 2.31,
|
| 152 |
+
"learning_rate": 2.6915974145891044e-05,
|
| 153 |
+
"loss": 0.0934,
|
| 154 |
"step": 10000
|
| 155 |
},
|
| 156 |
{
|
| 157 |
"epoch": 2.42,
|
| 158 |
+
"learning_rate": 2.5761772853185596e-05,
|
| 159 |
+
"loss": 0.0993,
|
| 160 |
"step": 10500
|
| 161 |
},
|
| 162 |
{
|
| 163 |
"epoch": 2.54,
|
| 164 |
+
"learning_rate": 2.460757156048015e-05,
|
| 165 |
+
"loss": 0.074,
|
| 166 |
"step": 11000
|
| 167 |
},
|
| 168 |
{
|
| 169 |
"epoch": 2.65,
|
| 170 |
+
"learning_rate": 2.3453370267774704e-05,
|
| 171 |
+
"loss": 0.0857,
|
| 172 |
"step": 11500
|
| 173 |
},
|
| 174 |
{
|
| 175 |
"epoch": 2.77,
|
| 176 |
+
"learning_rate": 2.2299168975069256e-05,
|
| 177 |
+
"loss": 0.0823,
|
| 178 |
"step": 12000
|
| 179 |
},
|
| 180 |
{
|
| 181 |
"epoch": 2.89,
|
| 182 |
+
"learning_rate": 2.1144967682363804e-05,
|
| 183 |
+
"loss": 0.0778,
|
| 184 |
"step": 12500
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 3.0,
|
| 188 |
+
"eval_gen_len": 48.70886075949367,
|
| 189 |
+
"eval_loss": 3.3238439559936523,
|
| 190 |
+
"eval_rouge1": 12.1173,
|
| 191 |
+
"eval_rouge2": 1.88,
|
| 192 |
+
"eval_rougeL": 8.1156,
|
| 193 |
+
"eval_rougeLsum": 10.1187,
|
| 194 |
+
"eval_runtime": 315.5777,
|
| 195 |
+
"eval_samples_per_second": 2.003,
|
| 196 |
+
"eval_steps_per_second": 2.003,
|
| 197 |
"step": 12996
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 3.0,
|
| 201 |
+
"learning_rate": 1.9990766389658356e-05,
|
| 202 |
+
"loss": 0.078,
|
| 203 |
"step": 13000
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 3.12,
|
| 207 |
+
"learning_rate": 1.883656509695291e-05,
|
| 208 |
+
"loss": 0.0546,
|
| 209 |
"step": 13500
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 3.23,
|
| 213 |
+
"learning_rate": 1.768236380424746e-05,
|
| 214 |
+
"loss": 0.062,
|
| 215 |
"step": 14000
|
| 216 |
},
|
| 217 |
{
|
| 218 |
"epoch": 3.35,
|
| 219 |
+
"learning_rate": 1.6528162511542013e-05,
|
| 220 |
+
"loss": 0.0656,
|
| 221 |
"step": 14500
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 3.46,
|
| 225 |
+
"learning_rate": 1.5373961218836565e-05,
|
| 226 |
+
"loss": 0.0555,
|
| 227 |
"step": 15000
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 3.58,
|
| 231 |
+
"learning_rate": 1.4219759926131118e-05,
|
| 232 |
+
"loss": 0.0437,
|
| 233 |
"step": 15500
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 3.69,
|
| 237 |
+
"learning_rate": 1.306555863342567e-05,
|
| 238 |
+
"loss": 0.05,
|
| 239 |
"step": 16000
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 3.81,
|
| 243 |
+
"learning_rate": 1.1911357340720223e-05,
|
| 244 |
+
"loss": 0.0577,
|
| 245 |
"step": 16500
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 3.92,
|
| 249 |
+
"learning_rate": 1.0757156048014775e-05,
|
| 250 |
+
"loss": 0.056,
|
| 251 |
"step": 17000
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 4.0,
|
| 255 |
+
"eval_gen_len": 50.7373417721519,
|
| 256 |
+
"eval_loss": 3.4031858444213867,
|
| 257 |
+
"eval_rouge1": 11.9555,
|
| 258 |
+
"eval_rouge2": 2.0536,
|
| 259 |
+
"eval_rougeL": 8.2185,
|
| 260 |
+
"eval_rougeLsum": 10.0656,
|
| 261 |
+
"eval_runtime": 343.5063,
|
| 262 |
+
"eval_samples_per_second": 1.84,
|
| 263 |
+
"eval_steps_per_second": 1.84,
|
| 264 |
"step": 17328
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 4.04,
|
| 268 |
+
"learning_rate": 9.602954755309327e-06,
|
| 269 |
+
"loss": 0.0416,
|
| 270 |
"step": 17500
|
| 271 |
},
|
| 272 |
{
|
| 273 |
"epoch": 4.16,
|
| 274 |
+
"learning_rate": 8.448753462603879e-06,
|
| 275 |
+
"loss": 0.0393,
|
| 276 |
"step": 18000
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 4.27,
|
| 280 |
+
"learning_rate": 7.29455216989843e-06,
|
| 281 |
+
"loss": 0.0368,
|
| 282 |
"step": 18500
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 4.39,
|
| 286 |
+
"learning_rate": 6.140350877192982e-06,
|
| 287 |
+
"loss": 0.0433,
|
| 288 |
"step": 19000
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 4.5,
|
| 292 |
+
"learning_rate": 4.986149584487535e-06,
|
| 293 |
+
"loss": 0.0421,
|
| 294 |
"step": 19500
|
| 295 |
},
|
| 296 |
{
|
| 297 |
"epoch": 4.62,
|
| 298 |
+
"learning_rate": 3.831948291782087e-06,
|
| 299 |
+
"loss": 0.038,
|
| 300 |
"step": 20000
|
| 301 |
},
|
| 302 |
{
|
| 303 |
"epoch": 4.73,
|
| 304 |
+
"learning_rate": 2.6777469990766392e-06,
|
| 305 |
+
"loss": 0.0285,
|
| 306 |
"step": 20500
|
| 307 |
},
|
| 308 |
{
|
| 309 |
"epoch": 4.85,
|
| 310 |
+
"learning_rate": 1.5235457063711913e-06,
|
| 311 |
+
"loss": 0.0343,
|
| 312 |
"step": 21000
|
| 313 |
},
|
| 314 |
{
|
| 315 |
"epoch": 4.96,
|
| 316 |
+
"learning_rate": 3.693444136657433e-07,
|
| 317 |
+
"loss": 0.0364,
|
| 318 |
"step": 21500
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 5.0,
|
| 322 |
+
"eval_gen_len": 50.44462025316456,
|
| 323 |
+
"eval_loss": 3.525162696838379,
|
| 324 |
+
"eval_rouge1": 11.814,
|
| 325 |
+
"eval_rouge2": 1.7965,
|
| 326 |
+
"eval_rougeL": 8.0177,
|
| 327 |
+
"eval_rougeLsum": 9.7342,
|
| 328 |
+
"eval_runtime": 338.0819,
|
| 329 |
+
"eval_samples_per_second": 1.869,
|
| 330 |
+
"eval_steps_per_second": 1.869,
|
| 331 |
"step": 21660
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 5.0,
|
| 335 |
"step": 21660,
|
| 336 |
"total_flos": 3568971203850240.0,
|
| 337 |
+
"train_loss": 0.23670565184904047,
|
| 338 |
+
"train_runtime": 3830.6771,
|
| 339 |
+
"train_samples_per_second": 5.654,
|
| 340 |
+
"train_steps_per_second": 5.654
|
| 341 |
}
|
| 342 |
],
|
| 343 |
"max_steps": 21660,
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4475
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f7081b5cc1106070ff6198b02a5a8687054ba5449cf8518c35628d1536648b8d
|
| 3 |
size 4475
|