Commit
·
a7f840a
1
Parent(s):
2f8b7f9
End of training
Browse files- all_results.json +9 -9
- eval_results.json +5 -5
- train_results.json +5 -5
- trainer_state.json +1190 -761
all_results.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second": 0.
|
| 6 |
"eval_steps_per_second": 0.291,
|
| 7 |
-
"eval_wer":
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples_per_second":
|
| 11 |
-
"train_steps_per_second": 0.
|
| 12 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"eval_loss": 0.5983160138130188,
|
| 4 |
+
"eval_runtime": 2218.8828,
|
| 5 |
+
"eval_samples_per_second": 0.581,
|
| 6 |
"eval_steps_per_second": 0.291,
|
| 7 |
+
"eval_wer": 27.49828990734407,
|
| 8 |
+
"train_loss": 0.0937204750527475,
|
| 9 |
+
"train_runtime": 49250.3081,
|
| 10 |
+
"train_samples_per_second": 2.524,
|
| 11 |
+
"train_steps_per_second": 0.158
|
| 12 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second": 0.
|
| 6 |
"eval_steps_per_second": 0.291,
|
| 7 |
-
"eval_wer":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"eval_loss": 0.5983160138130188,
|
| 4 |
+
"eval_runtime": 2218.8828,
|
| 5 |
+
"eval_samples_per_second": 0.581,
|
| 6 |
"eval_steps_per_second": 0.291,
|
| 7 |
+
"eval_wer": 27.49828990734407
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"train_loss": 0.0937204750527475,
|
| 4 |
+
"train_runtime": 49250.3081,
|
| 5 |
+
"train_samples_per_second": 2.524,
|
| 6 |
+
"train_steps_per_second": 0.158
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1519 +1,1948 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "/home/rcj2772/data_sets/Seneca/output_dir/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 0.0,
|
| 12 |
-
"learning_rate":
|
| 13 |
-
"loss": 4.
|
| 14 |
"step": 25
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.01,
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss":
|
| 20 |
"step": 50
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.01,
|
| 24 |
-
"learning_rate": 1.
|
| 25 |
-
"loss":
|
| 26 |
"step": 75
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch": 0.
|
| 30 |
-
"learning_rate": 1.
|
| 31 |
-
"loss":
|
| 32 |
"step": 100
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.02,
|
| 36 |
-
"learning_rate": 2.
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 125
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch": 0.
|
| 42 |
-
"learning_rate": 2.
|
| 43 |
-
"loss":
|
| 44 |
"step": 150
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"learning_rate": 3.
|
| 49 |
-
"loss":
|
| 50 |
"step": 175
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 0.03,
|
| 54 |
-
"learning_rate": 3.
|
| 55 |
-
"loss":
|
| 56 |
"step": 200
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch": 0.
|
| 60 |
-
"learning_rate": 4.
|
| 61 |
-
"loss":
|
| 62 |
"step": 225
|
| 63 |
},
|
| 64 |
{
|
| 65 |
-
"epoch": 0.
|
| 66 |
-
"learning_rate": 4.
|
| 67 |
-
"loss": 0.
|
| 68 |
"step": 250
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch": 0.
|
| 72 |
-
"learning_rate": 5.
|
| 73 |
-
"loss": 0.
|
| 74 |
"step": 275
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch": 0.
|
| 78 |
-
"learning_rate": 5.
|
| 79 |
-
"loss": 0.
|
| 80 |
"step": 300
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch": 0
|
| 84 |
-
"learning_rate": 6.
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 325
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch":
|
| 90 |
-
"learning_rate": 6.
|
| 91 |
-
"loss": 0.
|
| 92 |
"step": 350
|
| 93 |
},
|
| 94 |
{
|
| 95 |
-
"epoch":
|
| 96 |
-
"learning_rate": 7.
|
| 97 |
-
"loss": 0.
|
| 98 |
"step": 375
|
| 99 |
},
|
| 100 |
{
|
| 101 |
-
"epoch":
|
| 102 |
-
"learning_rate": 7.
|
| 103 |
-
"loss": 0.
|
| 104 |
"step": 400
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch":
|
| 108 |
-
"learning_rate": 8.
|
| 109 |
-
"loss": 0.
|
| 110 |
"step": 425
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"epoch":
|
| 114 |
-
"learning_rate": 8.
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 450
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch":
|
| 120 |
-
"learning_rate": 9.
|
| 121 |
-
"loss": 0.
|
| 122 |
"step": 475
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch":
|
| 126 |
-
"learning_rate": 9.
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 500
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch":
|
| 132 |
-
"learning_rate": 9.
|
| 133 |
-
"loss": 0.
|
| 134 |
"step": 525
|
| 135 |
},
|
| 136 |
{
|
| 137 |
-
"epoch":
|
| 138 |
-
"learning_rate": 9.
|
| 139 |
-
"loss": 0.
|
| 140 |
"step": 550
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"epoch":
|
| 144 |
-
"learning_rate": 9.
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 575
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch":
|
| 150 |
-
"learning_rate": 9.
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 600
|
| 153 |
},
|
| 154 |
{
|
| 155 |
-
"epoch": 0
|
| 156 |
-
"learning_rate": 9.
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 625
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch": 0
|
| 162 |
-
"learning_rate": 9.
|
| 163 |
-
"loss": 0.
|
| 164 |
"step": 650
|
| 165 |
},
|
| 166 |
{
|
| 167 |
-
"epoch":
|
| 168 |
-
"learning_rate": 9.
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 675
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch":
|
| 174 |
-
"learning_rate": 9.
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 700
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch":
|
| 180 |
-
"learning_rate": 9.
|
| 181 |
-
"loss": 0.
|
| 182 |
"step": 725
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch":
|
| 186 |
-
"learning_rate": 9.
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 750
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch":
|
| 192 |
-
"learning_rate": 9.
|
| 193 |
-
"loss": 0.
|
| 194 |
"step": 775
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch":
|
| 198 |
-
"learning_rate": 9.
|
| 199 |
-
"loss": 0.
|
| 200 |
"step": 800
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch":
|
| 204 |
-
"learning_rate": 9.
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 825
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch":
|
| 210 |
-
"learning_rate": 9.
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 850
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch":
|
| 216 |
-
"learning_rate": 9.
|
| 217 |
-
"loss": 0.
|
| 218 |
"step": 875
|
| 219 |
},
|
| 220 |
{
|
| 221 |
-
"epoch":
|
| 222 |
-
"learning_rate": 9.
|
| 223 |
-
"loss": 0.
|
| 224 |
"step": 900
|
| 225 |
},
|
| 226 |
{
|
| 227 |
-
"epoch":
|
| 228 |
-
"learning_rate": 9.
|
| 229 |
-
"loss": 0.
|
| 230 |
"step": 925
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"epoch": 0
|
| 234 |
-
"learning_rate": 9.
|
| 235 |
-
"loss": 0.
|
| 236 |
"step": 950
|
| 237 |
},
|
| 238 |
{
|
| 239 |
-
"epoch":
|
| 240 |
-
"learning_rate": 9.
|
| 241 |
-
"loss": 0.
|
| 242 |
"step": 975
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch":
|
| 246 |
-
"learning_rate": 9.
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 1000
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch":
|
| 252 |
-
"eval_loss": 0.
|
| 253 |
-
"eval_runtime":
|
| 254 |
-
"eval_samples_per_second": 0.
|
| 255 |
-
"eval_steps_per_second": 0.
|
| 256 |
-
"eval_wer":
|
| 257 |
"step": 1000
|
| 258 |
},
|
| 259 |
{
|
| 260 |
-
"epoch":
|
| 261 |
-
"learning_rate": 9.
|
| 262 |
-
"loss": 0.
|
| 263 |
"step": 1025
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch":
|
| 267 |
-
"learning_rate": 9.
|
| 268 |
-
"loss": 0.
|
| 269 |
"step": 1050
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch":
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 1075
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch":
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 1100
|
| 282 |
},
|
| 283 |
{
|
| 284 |
-
"epoch":
|
| 285 |
-
"learning_rate":
|
| 286 |
-
"loss": 0.
|
| 287 |
"step": 1125
|
| 288 |
},
|
| 289 |
{
|
| 290 |
-
"epoch":
|
| 291 |
-
"learning_rate":
|
| 292 |
-
"loss": 0.
|
| 293 |
"step": 1150
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"epoch":
|
| 297 |
-
"learning_rate":
|
| 298 |
-
"loss": 0.
|
| 299 |
"step": 1175
|
| 300 |
},
|
| 301 |
{
|
| 302 |
-
"epoch":
|
| 303 |
-
"learning_rate":
|
| 304 |
-
"loss": 0.
|
| 305 |
"step": 1200
|
| 306 |
},
|
| 307 |
{
|
| 308 |
-
"epoch":
|
| 309 |
-
"learning_rate":
|
| 310 |
-
"loss": 0.
|
| 311 |
"step": 1225
|
| 312 |
},
|
| 313 |
{
|
| 314 |
-
"epoch": 0
|
| 315 |
-
"learning_rate": 8.
|
| 316 |
-
"loss": 0.
|
| 317 |
"step": 1250
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch": 0
|
| 321 |
-
"learning_rate": 8.
|
| 322 |
-
"loss": 0.
|
| 323 |
"step": 1275
|
| 324 |
},
|
| 325 |
{
|
| 326 |
-
"epoch":
|
| 327 |
-
"learning_rate": 8.
|
| 328 |
-
"loss": 0.
|
| 329 |
"step": 1300
|
| 330 |
},
|
| 331 |
{
|
| 332 |
-
"epoch":
|
| 333 |
-
"learning_rate": 8.
|
| 334 |
-
"loss": 0.
|
| 335 |
"step": 1325
|
| 336 |
},
|
| 337 |
{
|
| 338 |
-
"epoch":
|
| 339 |
-
"learning_rate": 8.
|
| 340 |
-
"loss": 0.
|
| 341 |
"step": 1350
|
| 342 |
},
|
| 343 |
{
|
| 344 |
-
"epoch":
|
| 345 |
-
"learning_rate": 8.
|
| 346 |
-
"loss": 0.
|
| 347 |
"step": 1375
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch":
|
| 351 |
-
"learning_rate": 8.
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 1400
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch":
|
| 357 |
-
"learning_rate": 8.
|
| 358 |
-
"loss": 0.
|
| 359 |
"step": 1425
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch":
|
| 363 |
-
"learning_rate": 8.
|
| 364 |
-
"loss": 0.
|
| 365 |
"step": 1450
|
| 366 |
},
|
| 367 |
{
|
| 368 |
-
"epoch":
|
| 369 |
-
"learning_rate": 8.
|
| 370 |
-
"loss": 0.
|
| 371 |
"step": 1475
|
| 372 |
},
|
| 373 |
{
|
| 374 |
-
"epoch":
|
| 375 |
-
"learning_rate": 8.
|
| 376 |
-
"loss": 0.
|
| 377 |
"step": 1500
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch":
|
| 381 |
-
"learning_rate": 8.
|
| 382 |
-
"loss": 0.
|
| 383 |
"step": 1525
|
| 384 |
},
|
| 385 |
{
|
| 386 |
-
"epoch":
|
| 387 |
-
"learning_rate": 8.
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 1550
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch": 0
|
| 393 |
-
"learning_rate": 8.
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 1575
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch":
|
| 399 |
-
"learning_rate": 8.
|
| 400 |
-
"loss": 0.
|
| 401 |
"step": 1600
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch":
|
| 405 |
-
"learning_rate":
|
| 406 |
-
"loss": 0.
|
| 407 |
"step": 1625
|
| 408 |
},
|
| 409 |
{
|
| 410 |
-
"epoch":
|
| 411 |
-
"learning_rate":
|
| 412 |
-
"loss": 0.
|
| 413 |
"step": 1650
|
| 414 |
},
|
| 415 |
{
|
| 416 |
-
"epoch":
|
| 417 |
-
"learning_rate":
|
| 418 |
-
"loss": 0.
|
| 419 |
"step": 1675
|
| 420 |
},
|
| 421 |
{
|
| 422 |
-
"epoch":
|
| 423 |
-
"learning_rate":
|
| 424 |
-
"loss": 0.
|
| 425 |
"step": 1700
|
| 426 |
},
|
| 427 |
{
|
| 428 |
-
"epoch":
|
| 429 |
-
"learning_rate":
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 1725
|
| 432 |
},
|
| 433 |
{
|
| 434 |
-
"epoch":
|
| 435 |
-
"learning_rate":
|
| 436 |
-
"loss": 0.
|
| 437 |
"step": 1750
|
| 438 |
},
|
| 439 |
{
|
| 440 |
-
"epoch":
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss": 0.
|
| 443 |
"step": 1775
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch":
|
| 447 |
-
"learning_rate":
|
| 448 |
-
"loss": 0.
|
| 449 |
"step": 1800
|
| 450 |
},
|
| 451 |
{
|
| 452 |
-
"epoch":
|
| 453 |
-
"learning_rate":
|
| 454 |
-
"loss": 0.
|
| 455 |
"step": 1825
|
| 456 |
},
|
| 457 |
{
|
| 458 |
-
"epoch":
|
| 459 |
-
"learning_rate":
|
| 460 |
-
"loss": 0.
|
| 461 |
"step": 1850
|
| 462 |
},
|
| 463 |
{
|
| 464 |
-
"epoch": 0
|
| 465 |
-
"learning_rate":
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 1875
|
| 468 |
},
|
| 469 |
{
|
| 470 |
-
"epoch":
|
| 471 |
-
"learning_rate":
|
| 472 |
-
"loss": 0.
|
| 473 |
"step": 1900
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"epoch":
|
| 477 |
-
"learning_rate":
|
| 478 |
-
"loss": 0.
|
| 479 |
"step": 1925
|
| 480 |
},
|
| 481 |
{
|
| 482 |
-
"epoch":
|
| 483 |
-
"learning_rate":
|
| 484 |
-
"loss": 0.
|
| 485 |
"step": 1950
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch":
|
| 489 |
-
"learning_rate": 7.
|
| 490 |
-
"loss": 0.
|
| 491 |
"step": 1975
|
| 492 |
},
|
| 493 |
{
|
| 494 |
-
"epoch":
|
| 495 |
-
"learning_rate": 7.
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 2000
|
| 498 |
},
|
| 499 |
{
|
| 500 |
-
"epoch":
|
| 501 |
-
"eval_loss": 0.
|
| 502 |
-
"eval_runtime":
|
| 503 |
-
"eval_samples_per_second": 0.
|
| 504 |
-
"eval_steps_per_second": 0.
|
| 505 |
-
"eval_wer":
|
| 506 |
"step": 2000
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch":
|
| 510 |
-
"learning_rate": 7.
|
| 511 |
-
"loss": 0.
|
| 512 |
"step": 2025
|
| 513 |
},
|
| 514 |
{
|
| 515 |
-
"epoch":
|
| 516 |
-
"learning_rate": 7.
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 2050
|
| 519 |
},
|
| 520 |
{
|
| 521 |
-
"epoch":
|
| 522 |
-
"learning_rate": 7.
|
| 523 |
-
"loss": 0.
|
| 524 |
"step": 2075
|
| 525 |
},
|
| 526 |
{
|
| 527 |
-
"epoch":
|
| 528 |
-
"learning_rate": 7.
|
| 529 |
-
"loss": 0.
|
| 530 |
"step": 2100
|
| 531 |
},
|
| 532 |
{
|
| 533 |
-
"epoch":
|
| 534 |
-
"learning_rate": 7.
|
| 535 |
-
"loss": 0.
|
| 536 |
"step": 2125
|
| 537 |
},
|
| 538 |
{
|
| 539 |
-
"epoch":
|
| 540 |
-
"learning_rate": 7.
|
| 541 |
-
"loss": 0.
|
| 542 |
"step": 2150
|
| 543 |
},
|
| 544 |
{
|
| 545 |
-
"epoch": 0
|
| 546 |
-
"learning_rate":
|
| 547 |
-
"loss": 0.
|
| 548 |
"step": 2175
|
| 549 |
},
|
| 550 |
{
|
| 551 |
-
"epoch": 0
|
| 552 |
-
"learning_rate":
|
| 553 |
-
"loss": 0.
|
| 554 |
"step": 2200
|
| 555 |
},
|
| 556 |
{
|
| 557 |
-
"epoch":
|
| 558 |
-
"learning_rate":
|
| 559 |
-
"loss": 0.
|
| 560 |
"step": 2225
|
| 561 |
},
|
| 562 |
{
|
| 563 |
-
"epoch":
|
| 564 |
-
"learning_rate":
|
| 565 |
-
"loss": 0.
|
| 566 |
"step": 2250
|
| 567 |
},
|
| 568 |
{
|
| 569 |
-
"epoch":
|
| 570 |
-
"learning_rate":
|
| 571 |
-
"loss": 0.
|
| 572 |
"step": 2275
|
| 573 |
},
|
| 574 |
{
|
| 575 |
-
"epoch":
|
| 576 |
-
"learning_rate":
|
| 577 |
-
"loss": 0.
|
| 578 |
"step": 2300
|
| 579 |
},
|
| 580 |
{
|
| 581 |
-
"epoch":
|
| 582 |
-
"learning_rate":
|
| 583 |
-
"loss": 0.
|
| 584 |
"step": 2325
|
| 585 |
},
|
| 586 |
{
|
| 587 |
-
"epoch":
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 2350
|
| 591 |
},
|
| 592 |
{
|
| 593 |
-
"epoch":
|
| 594 |
-
"learning_rate":
|
| 595 |
-
"loss": 0.
|
| 596 |
"step": 2375
|
| 597 |
},
|
| 598 |
{
|
| 599 |
-
"epoch":
|
| 600 |
-
"learning_rate":
|
| 601 |
-
"loss": 0.
|
| 602 |
"step": 2400
|
| 603 |
},
|
| 604 |
{
|
| 605 |
-
"epoch":
|
| 606 |
-
"learning_rate":
|
| 607 |
-
"loss": 0.
|
| 608 |
"step": 2425
|
| 609 |
},
|
| 610 |
{
|
| 611 |
-
"epoch":
|
| 612 |
-
"learning_rate":
|
| 613 |
-
"loss": 0.
|
| 614 |
"step": 2450
|
| 615 |
},
|
| 616 |
{
|
| 617 |
-
"epoch":
|
| 618 |
-
"learning_rate":
|
| 619 |
-
"loss": 0.
|
| 620 |
"step": 2475
|
| 621 |
},
|
| 622 |
{
|
| 623 |
-
"epoch":
|
| 624 |
-
"learning_rate":
|
| 625 |
-
"loss": 0.
|
| 626 |
"step": 2500
|
| 627 |
},
|
| 628 |
{
|
| 629 |
-
"epoch":
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 2525
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch":
|
| 636 |
-
"learning_rate":
|
| 637 |
-
"loss": 0.
|
| 638 |
"step": 2550
|
| 639 |
},
|
| 640 |
{
|
| 641 |
-
"epoch":
|
| 642 |
-
"learning_rate":
|
| 643 |
-
"loss": 0.
|
| 644 |
"step": 2575
|
| 645 |
},
|
| 646 |
{
|
| 647 |
-
"epoch":
|
| 648 |
-
"learning_rate":
|
| 649 |
-
"loss": 0.
|
| 650 |
"step": 2600
|
| 651 |
},
|
| 652 |
{
|
| 653 |
-
"epoch":
|
| 654 |
-
"learning_rate":
|
| 655 |
-
"loss": 0.
|
| 656 |
"step": 2625
|
| 657 |
},
|
| 658 |
{
|
| 659 |
-
"epoch":
|
| 660 |
-
"learning_rate":
|
| 661 |
-
"loss": 0.
|
| 662 |
"step": 2650
|
| 663 |
},
|
| 664 |
{
|
| 665 |
-
"epoch":
|
| 666 |
-
"learning_rate":
|
| 667 |
-
"loss": 0.
|
| 668 |
"step": 2675
|
| 669 |
},
|
| 670 |
{
|
| 671 |
-
"epoch":
|
| 672 |
-
"learning_rate": 6.
|
| 673 |
-
"loss": 0.
|
| 674 |
"step": 2700
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch":
|
| 678 |
-
"learning_rate":
|
| 679 |
-
"loss": 0.
|
| 680 |
"step": 2725
|
| 681 |
},
|
| 682 |
{
|
| 683 |
-
"epoch":
|
| 684 |
-
"learning_rate":
|
| 685 |
-
"loss": 0.
|
| 686 |
"step": 2750
|
| 687 |
},
|
| 688 |
{
|
| 689 |
-
"epoch":
|
| 690 |
-
"learning_rate":
|
| 691 |
-
"loss": 0.
|
| 692 |
"step": 2775
|
| 693 |
},
|
| 694 |
{
|
| 695 |
-
"epoch":
|
| 696 |
-
"learning_rate":
|
| 697 |
-
"loss": 0.
|
| 698 |
"step": 2800
|
| 699 |
},
|
| 700 |
{
|
| 701 |
-
"epoch":
|
| 702 |
-
"learning_rate":
|
| 703 |
-
"loss": 0.
|
| 704 |
"step": 2825
|
| 705 |
},
|
| 706 |
{
|
| 707 |
-
"epoch":
|
| 708 |
-
"learning_rate":
|
| 709 |
-
"loss": 0.
|
| 710 |
"step": 2850
|
| 711 |
},
|
| 712 |
{
|
| 713 |
-
"epoch":
|
| 714 |
-
"learning_rate":
|
| 715 |
-
"loss": 0.
|
| 716 |
"step": 2875
|
| 717 |
},
|
| 718 |
{
|
| 719 |
-
"epoch":
|
| 720 |
-
"learning_rate":
|
| 721 |
-
"loss": 0.
|
| 722 |
"step": 2900
|
| 723 |
},
|
| 724 |
{
|
| 725 |
-
"epoch":
|
| 726 |
-
"learning_rate":
|
| 727 |
-
"loss": 0.
|
| 728 |
"step": 2925
|
| 729 |
},
|
| 730 |
{
|
| 731 |
-
"epoch":
|
| 732 |
-
"learning_rate":
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 2950
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"epoch":
|
| 738 |
-
"learning_rate":
|
| 739 |
-
"loss": 0.
|
| 740 |
"step": 2975
|
| 741 |
},
|
| 742 |
{
|
| 743 |
-
"epoch":
|
| 744 |
-
"learning_rate":
|
| 745 |
-
"loss": 0.
|
| 746 |
"step": 3000
|
| 747 |
},
|
| 748 |
{
|
| 749 |
-
"epoch":
|
| 750 |
-
"eval_loss": 0.
|
| 751 |
-
"eval_runtime":
|
| 752 |
-
"eval_samples_per_second": 0.
|
| 753 |
-
"eval_steps_per_second": 0.
|
| 754 |
-
"eval_wer":
|
| 755 |
"step": 3000
|
| 756 |
},
|
| 757 |
{
|
| 758 |
-
"epoch":
|
| 759 |
-
"learning_rate":
|
| 760 |
-
"loss": 0.
|
| 761 |
"step": 3025
|
| 762 |
},
|
| 763 |
{
|
| 764 |
-
"epoch":
|
| 765 |
-
"learning_rate":
|
| 766 |
-
"loss": 0.
|
| 767 |
"step": 3050
|
| 768 |
},
|
| 769 |
{
|
| 770 |
-
"epoch":
|
| 771 |
-
"learning_rate":
|
| 772 |
-
"loss": 0.
|
| 773 |
"step": 3075
|
| 774 |
},
|
| 775 |
{
|
| 776 |
-
"epoch":
|
| 777 |
-
"learning_rate":
|
| 778 |
-
"loss": 0.
|
| 779 |
"step": 3100
|
| 780 |
},
|
| 781 |
{
|
| 782 |
-
"epoch":
|
| 783 |
-
"learning_rate":
|
| 784 |
-
"loss": 0.
|
| 785 |
"step": 3125
|
| 786 |
},
|
| 787 |
{
|
| 788 |
-
"epoch":
|
| 789 |
-
"learning_rate":
|
| 790 |
-
"loss": 0.
|
| 791 |
"step": 3150
|
| 792 |
},
|
| 793 |
{
|
| 794 |
-
"epoch":
|
| 795 |
-
"learning_rate":
|
| 796 |
-
"loss": 0.
|
| 797 |
"step": 3175
|
| 798 |
},
|
| 799 |
{
|
| 800 |
-
"epoch":
|
| 801 |
-
"learning_rate":
|
| 802 |
-
"loss": 0.
|
| 803 |
"step": 3200
|
| 804 |
},
|
| 805 |
{
|
| 806 |
-
"epoch":
|
| 807 |
-
"learning_rate":
|
| 808 |
-
"loss": 0.
|
| 809 |
"step": 3225
|
| 810 |
},
|
| 811 |
{
|
| 812 |
-
"epoch":
|
| 813 |
-
"learning_rate":
|
| 814 |
-
"loss": 0.
|
| 815 |
"step": 3250
|
| 816 |
},
|
| 817 |
{
|
| 818 |
-
"epoch":
|
| 819 |
-
"learning_rate":
|
| 820 |
-
"loss": 0.
|
| 821 |
"step": 3275
|
| 822 |
},
|
| 823 |
{
|
| 824 |
-
"epoch":
|
| 825 |
-
"learning_rate":
|
| 826 |
-
"loss": 0.
|
| 827 |
"step": 3300
|
| 828 |
},
|
| 829 |
{
|
| 830 |
-
"epoch":
|
| 831 |
-
"learning_rate":
|
| 832 |
-
"loss": 0.
|
| 833 |
"step": 3325
|
| 834 |
},
|
| 835 |
{
|
| 836 |
-
"epoch":
|
| 837 |
-
"learning_rate":
|
| 838 |
-
"loss": 0.
|
| 839 |
"step": 3350
|
| 840 |
},
|
| 841 |
{
|
| 842 |
-
"epoch":
|
| 843 |
-
"learning_rate":
|
| 844 |
-
"loss": 0.
|
| 845 |
"step": 3375
|
| 846 |
},
|
| 847 |
{
|
| 848 |
-
"epoch":
|
| 849 |
-
"learning_rate":
|
| 850 |
-
"loss": 0.
|
| 851 |
"step": 3400
|
| 852 |
},
|
| 853 |
{
|
| 854 |
-
"epoch":
|
| 855 |
-
"learning_rate":
|
| 856 |
-
"loss": 0.
|
| 857 |
"step": 3425
|
| 858 |
},
|
| 859 |
{
|
| 860 |
-
"epoch":
|
| 861 |
-
"learning_rate":
|
| 862 |
-
"loss": 0.
|
| 863 |
"step": 3450
|
| 864 |
},
|
| 865 |
{
|
| 866 |
-
"epoch":
|
| 867 |
-
"learning_rate":
|
| 868 |
-
"loss": 0.
|
| 869 |
"step": 3475
|
| 870 |
},
|
| 871 |
{
|
| 872 |
-
"epoch":
|
| 873 |
-
"learning_rate":
|
| 874 |
-
"loss": 0.
|
| 875 |
"step": 3500
|
| 876 |
},
|
| 877 |
{
|
| 878 |
-
"epoch":
|
| 879 |
-
"learning_rate":
|
| 880 |
-
"loss": 0.
|
| 881 |
"step": 3525
|
| 882 |
},
|
| 883 |
{
|
| 884 |
-
"epoch":
|
| 885 |
-
"learning_rate":
|
| 886 |
-
"loss": 0.
|
| 887 |
"step": 3550
|
| 888 |
},
|
| 889 |
{
|
| 890 |
-
"epoch":
|
| 891 |
-
"learning_rate":
|
| 892 |
-
"loss": 0.
|
| 893 |
"step": 3575
|
| 894 |
},
|
| 895 |
{
|
| 896 |
-
"epoch":
|
| 897 |
-
"learning_rate":
|
| 898 |
-
"loss": 0.
|
| 899 |
"step": 3600
|
| 900 |
},
|
| 901 |
{
|
| 902 |
-
"epoch":
|
| 903 |
-
"learning_rate":
|
| 904 |
-
"loss": 0.
|
| 905 |
"step": 3625
|
| 906 |
},
|
| 907 |
{
|
| 908 |
-
"epoch":
|
| 909 |
-
"learning_rate":
|
| 910 |
-
"loss": 0.
|
| 911 |
"step": 3650
|
| 912 |
},
|
| 913 |
{
|
| 914 |
-
"epoch":
|
| 915 |
-
"learning_rate":
|
| 916 |
-
"loss": 0.
|
| 917 |
"step": 3675
|
| 918 |
},
|
| 919 |
{
|
| 920 |
-
"epoch":
|
| 921 |
-
"learning_rate":
|
| 922 |
-
"loss": 0.
|
| 923 |
"step": 3700
|
| 924 |
},
|
| 925 |
{
|
| 926 |
-
"epoch":
|
| 927 |
-
"learning_rate":
|
| 928 |
-
"loss": 0.
|
| 929 |
"step": 3725
|
| 930 |
},
|
| 931 |
{
|
| 932 |
-
"epoch":
|
| 933 |
-
"learning_rate":
|
| 934 |
-
"loss": 0.
|
| 935 |
"step": 3750
|
| 936 |
},
|
| 937 |
{
|
| 938 |
-
"epoch":
|
| 939 |
-
"learning_rate":
|
| 940 |
-
"loss": 0.
|
| 941 |
"step": 3775
|
| 942 |
},
|
| 943 |
{
|
| 944 |
-
"epoch":
|
| 945 |
-
"learning_rate":
|
| 946 |
-
"loss": 0.
|
| 947 |
"step": 3800
|
| 948 |
},
|
| 949 |
{
|
| 950 |
-
"epoch":
|
| 951 |
-
"learning_rate":
|
| 952 |
-
"loss": 0.
|
| 953 |
"step": 3825
|
| 954 |
},
|
| 955 |
{
|
| 956 |
-
"epoch":
|
| 957 |
-
"learning_rate":
|
| 958 |
-
"loss": 0.
|
| 959 |
"step": 3850
|
| 960 |
},
|
| 961 |
{
|
| 962 |
-
"epoch":
|
| 963 |
-
"learning_rate":
|
| 964 |
-
"loss": 0.
|
| 965 |
"step": 3875
|
| 966 |
},
|
| 967 |
{
|
| 968 |
-
"epoch":
|
| 969 |
-
"learning_rate":
|
| 970 |
-
"loss": 0.
|
| 971 |
"step": 3900
|
| 972 |
},
|
| 973 |
{
|
| 974 |
-
"epoch":
|
| 975 |
-
"learning_rate":
|
| 976 |
-
"loss": 0.
|
| 977 |
"step": 3925
|
| 978 |
},
|
| 979 |
{
|
| 980 |
-
"epoch":
|
| 981 |
-
"learning_rate":
|
| 982 |
-
"loss": 0.
|
| 983 |
"step": 3950
|
| 984 |
},
|
| 985 |
{
|
| 986 |
-
"epoch":
|
| 987 |
-
"learning_rate":
|
| 988 |
-
"loss": 0.
|
| 989 |
"step": 3975
|
| 990 |
},
|
| 991 |
{
|
| 992 |
-
"epoch":
|
| 993 |
-
"learning_rate":
|
| 994 |
-
"loss": 0.
|
| 995 |
"step": 4000
|
| 996 |
},
|
| 997 |
{
|
| 998 |
-
"epoch":
|
| 999 |
-
"eval_loss": 0.
|
| 1000 |
-
"eval_runtime":
|
| 1001 |
-
"eval_samples_per_second": 0.
|
| 1002 |
-
"eval_steps_per_second": 0.
|
| 1003 |
-
"eval_wer":
|
| 1004 |
"step": 4000
|
| 1005 |
},
|
| 1006 |
{
|
| 1007 |
-
"epoch":
|
| 1008 |
-
"learning_rate":
|
| 1009 |
-
"loss": 0.
|
| 1010 |
"step": 4025
|
| 1011 |
},
|
| 1012 |
{
|
| 1013 |
-
"epoch":
|
| 1014 |
-
"learning_rate":
|
| 1015 |
-
"loss": 0.
|
| 1016 |
"step": 4050
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
-
"epoch":
|
| 1020 |
-
"learning_rate":
|
| 1021 |
-
"loss": 0.
|
| 1022 |
"step": 4075
|
| 1023 |
},
|
| 1024 |
{
|
| 1025 |
-
"epoch":
|
| 1026 |
-
"learning_rate":
|
| 1027 |
-
"loss": 0.
|
| 1028 |
"step": 4100
|
| 1029 |
},
|
| 1030 |
{
|
| 1031 |
-
"epoch":
|
| 1032 |
-
"learning_rate":
|
| 1033 |
-
"loss": 0.
|
| 1034 |
"step": 4125
|
| 1035 |
},
|
| 1036 |
{
|
| 1037 |
-
"epoch":
|
| 1038 |
-
"learning_rate":
|
| 1039 |
-
"loss": 0.
|
| 1040 |
"step": 4150
|
| 1041 |
},
|
| 1042 |
{
|
| 1043 |
-
"epoch":
|
| 1044 |
-
"learning_rate":
|
| 1045 |
-
"loss": 0.
|
| 1046 |
"step": 4175
|
| 1047 |
},
|
| 1048 |
{
|
| 1049 |
-
"epoch":
|
| 1050 |
-
"learning_rate":
|
| 1051 |
-
"loss": 0.
|
| 1052 |
"step": 4200
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
-
"epoch":
|
| 1056 |
-
"learning_rate":
|
| 1057 |
-
"loss": 0.
|
| 1058 |
"step": 4225
|
| 1059 |
},
|
| 1060 |
{
|
| 1061 |
-
"epoch":
|
| 1062 |
-
"learning_rate":
|
| 1063 |
-
"loss": 0.
|
| 1064 |
"step": 4250
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
-
"epoch":
|
| 1068 |
-
"learning_rate":
|
| 1069 |
-
"loss": 0.
|
| 1070 |
"step": 4275
|
| 1071 |
},
|
| 1072 |
{
|
| 1073 |
-
"epoch":
|
| 1074 |
-
"learning_rate":
|
| 1075 |
-
"loss": 0.
|
| 1076 |
"step": 4300
|
| 1077 |
},
|
| 1078 |
{
|
| 1079 |
-
"epoch":
|
| 1080 |
-
"learning_rate":
|
| 1081 |
-
"loss": 0.
|
| 1082 |
"step": 4325
|
| 1083 |
},
|
| 1084 |
{
|
| 1085 |
-
"epoch":
|
| 1086 |
-
"learning_rate":
|
| 1087 |
-
"loss": 0.
|
| 1088 |
"step": 4350
|
| 1089 |
},
|
| 1090 |
{
|
| 1091 |
-
"epoch":
|
| 1092 |
-
"learning_rate":
|
| 1093 |
-
"loss": 0.
|
| 1094 |
"step": 4375
|
| 1095 |
},
|
| 1096 |
{
|
| 1097 |
-
"epoch":
|
| 1098 |
-
"learning_rate":
|
| 1099 |
-
"loss": 0.
|
| 1100 |
"step": 4400
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
-
"epoch":
|
| 1104 |
-
"learning_rate":
|
| 1105 |
-
"loss": 0.
|
| 1106 |
"step": 4425
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
-
"epoch":
|
| 1110 |
-
"learning_rate":
|
| 1111 |
-
"loss": 0.
|
| 1112 |
"step": 4450
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
-
"epoch":
|
| 1116 |
-
"learning_rate":
|
| 1117 |
-
"loss": 0.
|
| 1118 |
"step": 4475
|
| 1119 |
},
|
| 1120 |
{
|
| 1121 |
-
"epoch":
|
| 1122 |
-
"learning_rate":
|
| 1123 |
-
"loss": 0.
|
| 1124 |
"step": 4500
|
| 1125 |
},
|
| 1126 |
{
|
| 1127 |
-
"epoch":
|
| 1128 |
-
"learning_rate":
|
| 1129 |
-
"loss": 0.
|
| 1130 |
"step": 4525
|
| 1131 |
},
|
| 1132 |
{
|
| 1133 |
-
"epoch":
|
| 1134 |
-
"learning_rate":
|
| 1135 |
-
"loss": 0.
|
| 1136 |
"step": 4550
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
-
"epoch":
|
| 1140 |
-
"learning_rate":
|
| 1141 |
-
"loss": 0.
|
| 1142 |
"step": 4575
|
| 1143 |
},
|
| 1144 |
{
|
| 1145 |
-
"epoch":
|
| 1146 |
-
"learning_rate":
|
| 1147 |
-
"loss": 0.
|
| 1148 |
"step": 4600
|
| 1149 |
},
|
| 1150 |
{
|
| 1151 |
-
"epoch":
|
| 1152 |
-
"learning_rate":
|
| 1153 |
-
"loss": 0.
|
| 1154 |
"step": 4625
|
| 1155 |
},
|
| 1156 |
{
|
| 1157 |
-
"epoch":
|
| 1158 |
-
"learning_rate":
|
| 1159 |
-
"loss": 0.
|
| 1160 |
"step": 4650
|
| 1161 |
},
|
| 1162 |
{
|
| 1163 |
-
"epoch":
|
| 1164 |
-
"learning_rate":
|
| 1165 |
-
"loss": 0.
|
| 1166 |
"step": 4675
|
| 1167 |
},
|
| 1168 |
{
|
| 1169 |
-
"epoch":
|
| 1170 |
-
"learning_rate":
|
| 1171 |
-
"loss": 0.
|
| 1172 |
"step": 4700
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
-
"epoch":
|
| 1176 |
-
"learning_rate":
|
| 1177 |
-
"loss": 0.
|
| 1178 |
"step": 4725
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
-
"epoch":
|
| 1182 |
-
"learning_rate":
|
| 1183 |
-
"loss": 0.
|
| 1184 |
"step": 4750
|
| 1185 |
},
|
| 1186 |
{
|
| 1187 |
-
"epoch":
|
| 1188 |
-
"learning_rate":
|
| 1189 |
-
"loss": 0.
|
| 1190 |
"step": 4775
|
| 1191 |
},
|
| 1192 |
{
|
| 1193 |
-
"epoch":
|
| 1194 |
-
"learning_rate":
|
| 1195 |
-
"loss": 0.
|
| 1196 |
"step": 4800
|
| 1197 |
},
|
| 1198 |
{
|
| 1199 |
-
"epoch":
|
| 1200 |
-
"learning_rate":
|
| 1201 |
-
"loss": 0.
|
| 1202 |
"step": 4825
|
| 1203 |
},
|
| 1204 |
{
|
| 1205 |
-
"epoch":
|
| 1206 |
-
"learning_rate":
|
| 1207 |
-
"loss": 0.
|
| 1208 |
"step": 4850
|
| 1209 |
},
|
| 1210 |
{
|
| 1211 |
-
"epoch":
|
| 1212 |
-
"learning_rate":
|
| 1213 |
-
"loss": 0.
|
| 1214 |
"step": 4875
|
| 1215 |
},
|
| 1216 |
{
|
| 1217 |
-
"epoch":
|
| 1218 |
-
"learning_rate":
|
| 1219 |
-
"loss": 0.
|
| 1220 |
"step": 4900
|
| 1221 |
},
|
| 1222 |
{
|
| 1223 |
-
"epoch":
|
| 1224 |
-
"learning_rate":
|
| 1225 |
-
"loss": 0.
|
| 1226 |
"step": 4925
|
| 1227 |
},
|
| 1228 |
{
|
| 1229 |
-
"epoch":
|
| 1230 |
-
"learning_rate":
|
| 1231 |
-
"loss": 0.
|
| 1232 |
"step": 4950
|
| 1233 |
},
|
| 1234 |
{
|
| 1235 |
-
"epoch":
|
| 1236 |
-
"learning_rate":
|
| 1237 |
-
"loss": 0.
|
| 1238 |
"step": 4975
|
| 1239 |
},
|
| 1240 |
{
|
| 1241 |
-
"epoch":
|
| 1242 |
-
"learning_rate":
|
| 1243 |
-
"loss": 0.
|
| 1244 |
"step": 5000
|
| 1245 |
},
|
| 1246 |
{
|
| 1247 |
-
"epoch":
|
| 1248 |
-
"eval_loss": 0.
|
| 1249 |
-
"eval_runtime":
|
| 1250 |
"eval_samples_per_second": 0.583,
|
| 1251 |
"eval_steps_per_second": 0.292,
|
| 1252 |
-
"eval_wer":
|
| 1253 |
"step": 5000
|
| 1254 |
},
|
| 1255 |
{
|
| 1256 |
-
"epoch":
|
| 1257 |
-
"learning_rate":
|
| 1258 |
-
"loss": 0.
|
| 1259 |
"step": 5025
|
| 1260 |
},
|
| 1261 |
{
|
| 1262 |
-
"epoch":
|
| 1263 |
-
"learning_rate":
|
| 1264 |
-
"loss": 0.
|
| 1265 |
"step": 5050
|
| 1266 |
},
|
| 1267 |
{
|
| 1268 |
-
"epoch":
|
| 1269 |
-
"learning_rate":
|
| 1270 |
-
"loss": 0.
|
| 1271 |
"step": 5075
|
| 1272 |
},
|
| 1273 |
{
|
| 1274 |
-
"epoch":
|
| 1275 |
-
"learning_rate":
|
| 1276 |
-
"loss": 0.
|
| 1277 |
"step": 5100
|
| 1278 |
},
|
| 1279 |
{
|
| 1280 |
-
"epoch":
|
| 1281 |
-
"learning_rate":
|
| 1282 |
-
"loss": 0.
|
| 1283 |
"step": 5125
|
| 1284 |
},
|
| 1285 |
{
|
| 1286 |
-
"epoch":
|
| 1287 |
-
"learning_rate":
|
| 1288 |
-
"loss": 0.
|
| 1289 |
"step": 5150
|
| 1290 |
},
|
| 1291 |
{
|
| 1292 |
-
"epoch":
|
| 1293 |
-
"learning_rate":
|
| 1294 |
-
"loss": 0.
|
| 1295 |
"step": 5175
|
| 1296 |
},
|
| 1297 |
{
|
| 1298 |
-
"epoch":
|
| 1299 |
-
"learning_rate":
|
| 1300 |
-
"loss": 0.
|
| 1301 |
"step": 5200
|
| 1302 |
},
|
| 1303 |
{
|
| 1304 |
-
"epoch":
|
| 1305 |
-
"learning_rate":
|
| 1306 |
-
"loss": 0.
|
| 1307 |
"step": 5225
|
| 1308 |
},
|
| 1309 |
{
|
| 1310 |
-
"epoch":
|
| 1311 |
-
"learning_rate":
|
| 1312 |
-
"loss": 0.
|
| 1313 |
"step": 5250
|
| 1314 |
},
|
| 1315 |
{
|
| 1316 |
-
"epoch":
|
| 1317 |
-
"learning_rate":
|
| 1318 |
-
"loss": 0.
|
| 1319 |
"step": 5275
|
| 1320 |
},
|
| 1321 |
{
|
| 1322 |
-
"epoch":
|
| 1323 |
-
"learning_rate":
|
| 1324 |
-
"loss": 0.
|
| 1325 |
"step": 5300
|
| 1326 |
},
|
| 1327 |
{
|
| 1328 |
-
"epoch":
|
| 1329 |
-
"learning_rate":
|
| 1330 |
-
"loss": 0.
|
| 1331 |
"step": 5325
|
| 1332 |
},
|
| 1333 |
{
|
| 1334 |
-
"epoch":
|
| 1335 |
-
"learning_rate":
|
| 1336 |
-
"loss": 0.
|
| 1337 |
"step": 5350
|
| 1338 |
},
|
| 1339 |
{
|
| 1340 |
-
"epoch":
|
| 1341 |
-
"learning_rate":
|
| 1342 |
-
"loss": 0.
|
| 1343 |
"step": 5375
|
| 1344 |
},
|
| 1345 |
{
|
| 1346 |
-
"epoch":
|
| 1347 |
-
"learning_rate":
|
| 1348 |
-
"loss": 0.
|
| 1349 |
"step": 5400
|
| 1350 |
},
|
| 1351 |
{
|
| 1352 |
-
"epoch":
|
| 1353 |
-
"learning_rate":
|
| 1354 |
-
"loss": 0.
|
| 1355 |
"step": 5425
|
| 1356 |
},
|
| 1357 |
{
|
| 1358 |
-
"epoch":
|
| 1359 |
-
"learning_rate":
|
| 1360 |
-
"loss": 0.
|
| 1361 |
"step": 5450
|
| 1362 |
},
|
| 1363 |
{
|
| 1364 |
-
"epoch":
|
| 1365 |
-
"learning_rate":
|
| 1366 |
-
"loss": 0.
|
| 1367 |
"step": 5475
|
| 1368 |
},
|
| 1369 |
{
|
| 1370 |
-
"epoch":
|
| 1371 |
-
"learning_rate":
|
| 1372 |
-
"loss": 0.
|
| 1373 |
"step": 5500
|
| 1374 |
},
|
| 1375 |
{
|
| 1376 |
-
"epoch":
|
| 1377 |
-
"learning_rate":
|
| 1378 |
-
"loss": 0.
|
| 1379 |
"step": 5525
|
| 1380 |
},
|
| 1381 |
{
|
| 1382 |
-
"epoch":
|
| 1383 |
-
"learning_rate":
|
| 1384 |
-
"loss": 0.
|
| 1385 |
"step": 5550
|
| 1386 |
},
|
| 1387 |
{
|
| 1388 |
-
"epoch":
|
| 1389 |
-
"learning_rate":
|
| 1390 |
-
"loss": 0.
|
| 1391 |
"step": 5575
|
| 1392 |
},
|
| 1393 |
{
|
| 1394 |
-
"epoch":
|
| 1395 |
-
"learning_rate":
|
| 1396 |
-
"loss": 0.
|
| 1397 |
"step": 5600
|
| 1398 |
},
|
| 1399 |
{
|
| 1400 |
-
"epoch":
|
| 1401 |
-
"learning_rate":
|
| 1402 |
-
"loss": 0.
|
| 1403 |
"step": 5625
|
| 1404 |
},
|
| 1405 |
{
|
| 1406 |
-
"epoch":
|
| 1407 |
-
"learning_rate":
|
| 1408 |
-
"loss": 0.
|
| 1409 |
"step": 5650
|
| 1410 |
},
|
| 1411 |
{
|
| 1412 |
-
"epoch":
|
| 1413 |
-
"learning_rate":
|
| 1414 |
-
"loss": 0.
|
| 1415 |
"step": 5675
|
| 1416 |
},
|
| 1417 |
{
|
| 1418 |
-
"epoch":
|
| 1419 |
-
"learning_rate":
|
| 1420 |
-
"loss": 0.
|
| 1421 |
"step": 5700
|
| 1422 |
},
|
| 1423 |
{
|
| 1424 |
-
"epoch":
|
| 1425 |
-
"learning_rate":
|
| 1426 |
-
"loss": 0.
|
| 1427 |
"step": 5725
|
| 1428 |
},
|
| 1429 |
{
|
| 1430 |
-
"epoch":
|
| 1431 |
-
"learning_rate":
|
| 1432 |
-
"loss": 0.
|
| 1433 |
"step": 5750
|
| 1434 |
},
|
| 1435 |
{
|
| 1436 |
-
"epoch":
|
| 1437 |
-
"learning_rate":
|
| 1438 |
-
"loss": 0.
|
| 1439 |
"step": 5775
|
| 1440 |
},
|
| 1441 |
{
|
| 1442 |
-
"epoch":
|
| 1443 |
-
"learning_rate":
|
| 1444 |
-
"loss": 0.
|
| 1445 |
"step": 5800
|
| 1446 |
},
|
| 1447 |
{
|
| 1448 |
-
"epoch":
|
| 1449 |
-
"learning_rate":
|
| 1450 |
-
"loss": 0.
|
| 1451 |
"step": 5825
|
| 1452 |
},
|
| 1453 |
{
|
| 1454 |
-
"epoch":
|
| 1455 |
-
"learning_rate": 2.
|
| 1456 |
-
"loss": 0.
|
| 1457 |
"step": 5850
|
| 1458 |
},
|
| 1459 |
{
|
| 1460 |
-
"epoch":
|
| 1461 |
-
"learning_rate": 2.
|
| 1462 |
-
"loss": 0.
|
| 1463 |
"step": 5875
|
| 1464 |
},
|
| 1465 |
{
|
| 1466 |
-
"epoch":
|
| 1467 |
-
"learning_rate":
|
| 1468 |
-
"loss": 0.
|
| 1469 |
"step": 5900
|
| 1470 |
},
|
| 1471 |
{
|
| 1472 |
-
"epoch":
|
| 1473 |
-
"learning_rate":
|
| 1474 |
-
"loss": 0.
|
| 1475 |
"step": 5925
|
| 1476 |
},
|
| 1477 |
{
|
| 1478 |
-
"epoch":
|
| 1479 |
-
"learning_rate":
|
| 1480 |
-
"loss": 0.
|
| 1481 |
"step": 5950
|
| 1482 |
},
|
| 1483 |
{
|
| 1484 |
-
"epoch":
|
| 1485 |
-
"learning_rate":
|
| 1486 |
-
"loss": 0.
|
| 1487 |
"step": 5975
|
| 1488 |
},
|
| 1489 |
{
|
| 1490 |
-
"epoch":
|
| 1491 |
-
"learning_rate":
|
| 1492 |
-
"loss": 0.
|
| 1493 |
"step": 6000
|
| 1494 |
},
|
| 1495 |
{
|
| 1496 |
-
"epoch":
|
| 1497 |
-
"eval_loss": 0.
|
| 1498 |
-
"eval_runtime":
|
| 1499 |
-
"eval_samples_per_second": 0.
|
| 1500 |
"eval_steps_per_second": 0.289,
|
| 1501 |
-
"eval_wer": 27.
|
| 1502 |
"step": 6000
|
| 1503 |
},
|
| 1504 |
{
|
| 1505 |
-
"epoch":
|
| 1506 |
-
"
|
| 1507 |
-
"
|
| 1508 |
-
"
|
| 1509 |
-
|
| 1510 |
-
|
| 1511 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1512 |
}
|
| 1513 |
],
|
| 1514 |
-
"max_steps":
|
| 1515 |
"num_train_epochs": 9223372036854775807,
|
| 1516 |
-
"total_flos": 1.
|
| 1517 |
"trial_name": null,
|
| 1518 |
"trial_params": null
|
| 1519 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 27.49828990734407,
|
| 3 |
+
"best_model_checkpoint": "/home/rcj2772/data_sets/Seneca/output_dir/checkpoint-7000",
|
| 4 |
+
"epoch": 25.00231719876416,
|
| 5 |
+
"global_step": 7768,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 0.0,
|
| 12 |
+
"learning_rate": 4.2000000000000006e-07,
|
| 13 |
+
"loss": 4.3385,
|
| 14 |
"step": 25
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.01,
|
| 18 |
+
"learning_rate": 9.200000000000001e-07,
|
| 19 |
+
"loss": 3.4704,
|
| 20 |
"step": 50
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.01,
|
| 24 |
+
"learning_rate": 1.42e-06,
|
| 25 |
+
"loss": 2.4271,
|
| 26 |
"step": 75
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.01,
|
| 30 |
+
"learning_rate": 1.9200000000000003e-06,
|
| 31 |
+
"loss": 1.6065,
|
| 32 |
"step": 100
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 0.02,
|
| 36 |
+
"learning_rate": 2.42e-06,
|
| 37 |
+
"loss": 1.1884,
|
| 38 |
"step": 125
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.02,
|
| 42 |
+
"learning_rate": 2.92e-06,
|
| 43 |
+
"loss": 0.9446,
|
| 44 |
"step": 150
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.02,
|
| 48 |
+
"learning_rate": 3.4200000000000007e-06,
|
| 49 |
+
"loss": 0.7629,
|
| 50 |
"step": 175
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 0.03,
|
| 54 |
+
"learning_rate": 3.920000000000001e-06,
|
| 55 |
+
"loss": 0.7097,
|
| 56 |
"step": 200
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 0.03,
|
| 60 |
+
"learning_rate": 4.42e-06,
|
| 61 |
+
"loss": 0.6508,
|
| 62 |
"step": 225
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 0.03,
|
| 66 |
+
"learning_rate": 4.92e-06,
|
| 67 |
+
"loss": 0.5891,
|
| 68 |
"step": 250
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 0.04,
|
| 72 |
+
"learning_rate": 5.420000000000001e-06,
|
| 73 |
+
"loss": 0.5815,
|
| 74 |
"step": 275
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 0.04,
|
| 78 |
+
"learning_rate": 5.92e-06,
|
| 79 |
+
"loss": 0.5616,
|
| 80 |
"step": 300
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 1.0,
|
| 84 |
+
"learning_rate": 6.42e-06,
|
| 85 |
+
"loss": 0.455,
|
| 86 |
"step": 325
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 1.01,
|
| 90 |
+
"learning_rate": 6.92e-06,
|
| 91 |
+
"loss": 0.4164,
|
| 92 |
"step": 350
|
| 93 |
},
|
| 94 |
{
|
| 95 |
+
"epoch": 1.01,
|
| 96 |
+
"learning_rate": 7.420000000000001e-06,
|
| 97 |
+
"loss": 0.4369,
|
| 98 |
"step": 375
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 1.01,
|
| 102 |
+
"learning_rate": 7.92e-06,
|
| 103 |
+
"loss": 0.4078,
|
| 104 |
"step": 400
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 1.01,
|
| 108 |
+
"learning_rate": 8.42e-06,
|
| 109 |
+
"loss": 0.4149,
|
| 110 |
"step": 425
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"epoch": 1.02,
|
| 114 |
+
"learning_rate": 8.920000000000001e-06,
|
| 115 |
+
"loss": 0.4046,
|
| 116 |
"step": 450
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 1.02,
|
| 120 |
+
"learning_rate": 9.42e-06,
|
| 121 |
+
"loss": 0.3988,
|
| 122 |
"step": 475
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 1.02,
|
| 126 |
+
"learning_rate": 9.920000000000002e-06,
|
| 127 |
+
"loss": 0.3853,
|
| 128 |
"step": 500
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 1.03,
|
| 132 |
+
"learning_rate": 9.971106219042378e-06,
|
| 133 |
+
"loss": 0.3684,
|
| 134 |
"step": 525
|
| 135 |
},
|
| 136 |
{
|
| 137 |
+
"epoch": 1.03,
|
| 138 |
+
"learning_rate": 9.936708860759493e-06,
|
| 139 |
+
"loss": 0.375,
|
| 140 |
"step": 550
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"epoch": 1.03,
|
| 144 |
+
"learning_rate": 9.90231150247661e-06,
|
| 145 |
+
"loss": 0.3839,
|
| 146 |
"step": 575
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 1.04,
|
| 150 |
+
"learning_rate": 9.867914144193726e-06,
|
| 151 |
+
"loss": 0.3818,
|
| 152 |
"step": 600
|
| 153 |
},
|
| 154 |
{
|
| 155 |
+
"epoch": 2.0,
|
| 156 |
+
"learning_rate": 9.833516785910844e-06,
|
| 157 |
+
"loss": 0.3041,
|
| 158 |
"step": 625
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 2.0,
|
| 162 |
+
"learning_rate": 9.799119427627959e-06,
|
| 163 |
+
"loss": 0.2169,
|
| 164 |
"step": 650
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 2.01,
|
| 168 |
+
"learning_rate": 9.764722069345075e-06,
|
| 169 |
+
"loss": 0.2362,
|
| 170 |
"step": 675
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 2.01,
|
| 174 |
+
"learning_rate": 9.730324711062192e-06,
|
| 175 |
+
"loss": 0.2139,
|
| 176 |
"step": 700
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 2.01,
|
| 180 |
+
"learning_rate": 9.695927352779307e-06,
|
| 181 |
+
"loss": 0.2241,
|
| 182 |
"step": 725
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"epoch": 2.02,
|
| 186 |
+
"learning_rate": 9.661529994496423e-06,
|
| 187 |
+
"loss": 0.2259,
|
| 188 |
"step": 750
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 2.02,
|
| 192 |
+
"learning_rate": 9.62713263621354e-06,
|
| 193 |
+
"loss": 0.2239,
|
| 194 |
"step": 775
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 2.02,
|
| 198 |
+
"learning_rate": 9.592735277930656e-06,
|
| 199 |
+
"loss": 0.2189,
|
| 200 |
"step": 800
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 2.03,
|
| 204 |
+
"learning_rate": 9.558337919647771e-06,
|
| 205 |
+
"loss": 0.2327,
|
| 206 |
"step": 825
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 2.03,
|
| 210 |
+
"learning_rate": 9.523940561364889e-06,
|
| 211 |
+
"loss": 0.2268,
|
| 212 |
"step": 850
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 2.03,
|
| 216 |
+
"learning_rate": 9.489543203082004e-06,
|
| 217 |
+
"loss": 0.259,
|
| 218 |
"step": 875
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 2.04,
|
| 222 |
+
"learning_rate": 9.45514584479912e-06,
|
| 223 |
+
"loss": 0.2077,
|
| 224 |
"step": 900
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 2.04,
|
| 228 |
+
"learning_rate": 9.420748486516237e-06,
|
| 229 |
+
"loss": 0.2208,
|
| 230 |
"step": 925
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 3.0,
|
| 234 |
+
"learning_rate": 9.386351128233353e-06,
|
| 235 |
+
"loss": 0.1276,
|
| 236 |
"step": 950
|
| 237 |
},
|
| 238 |
{
|
| 239 |
+
"epoch": 3.01,
|
| 240 |
+
"learning_rate": 9.351953769950468e-06,
|
| 241 |
+
"loss": 0.1161,
|
| 242 |
"step": 975
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 3.01,
|
| 246 |
+
"learning_rate": 9.317556411667584e-06,
|
| 247 |
+
"loss": 0.1151,
|
| 248 |
"step": 1000
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 3.01,
|
| 252 |
+
"eval_loss": 0.4013126492500305,
|
| 253 |
+
"eval_runtime": 2274.8036,
|
| 254 |
+
"eval_samples_per_second": 0.567,
|
| 255 |
+
"eval_steps_per_second": 0.284,
|
| 256 |
+
"eval_wer": 31.83259747528139,
|
| 257 |
"step": 1000
|
| 258 |
},
|
| 259 |
{
|
| 260 |
+
"epoch": 3.01,
|
| 261 |
+
"learning_rate": 9.283159053384701e-06,
|
| 262 |
+
"loss": 0.1103,
|
| 263 |
"step": 1025
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"epoch": 3.02,
|
| 267 |
+
"learning_rate": 9.248761695101817e-06,
|
| 268 |
+
"loss": 0.1075,
|
| 269 |
"step": 1050
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"epoch": 3.02,
|
| 273 |
+
"learning_rate": 9.214364336818932e-06,
|
| 274 |
+
"loss": 0.1121,
|
| 275 |
"step": 1075
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 3.02,
|
| 279 |
+
"learning_rate": 9.17996697853605e-06,
|
| 280 |
+
"loss": 0.1189,
|
| 281 |
"step": 1100
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 3.03,
|
| 285 |
+
"learning_rate": 9.145569620253165e-06,
|
| 286 |
+
"loss": 0.1147,
|
| 287 |
"step": 1125
|
| 288 |
},
|
| 289 |
{
|
| 290 |
+
"epoch": 3.03,
|
| 291 |
+
"learning_rate": 9.11117226197028e-06,
|
| 292 |
+
"loss": 0.126,
|
| 293 |
"step": 1150
|
| 294 |
},
|
| 295 |
{
|
| 296 |
+
"epoch": 3.03,
|
| 297 |
+
"learning_rate": 9.076774903687398e-06,
|
| 298 |
+
"loss": 0.1177,
|
| 299 |
"step": 1175
|
| 300 |
},
|
| 301 |
{
|
| 302 |
+
"epoch": 3.03,
|
| 303 |
+
"learning_rate": 9.042377545404513e-06,
|
| 304 |
+
"loss": 0.1294,
|
| 305 |
"step": 1200
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"epoch": 3.04,
|
| 309 |
+
"learning_rate": 9.00798018712163e-06,
|
| 310 |
+
"loss": 0.1222,
|
| 311 |
"step": 1225
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"epoch": 4.0,
|
| 315 |
+
"learning_rate": 8.973582828838746e-06,
|
| 316 |
+
"loss": 0.0985,
|
| 317 |
"step": 1250
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 4.0,
|
| 321 |
+
"learning_rate": 8.939185470555862e-06,
|
| 322 |
+
"loss": 0.0479,
|
| 323 |
"step": 1275
|
| 324 |
},
|
| 325 |
{
|
| 326 |
+
"epoch": 4.01,
|
| 327 |
+
"learning_rate": 8.90478811227298e-06,
|
| 328 |
+
"loss": 0.059,
|
| 329 |
"step": 1300
|
| 330 |
},
|
| 331 |
{
|
| 332 |
+
"epoch": 4.01,
|
| 333 |
+
"learning_rate": 8.870390753990095e-06,
|
| 334 |
+
"loss": 0.0533,
|
| 335 |
"step": 1325
|
| 336 |
},
|
| 337 |
{
|
| 338 |
+
"epoch": 4.01,
|
| 339 |
+
"learning_rate": 8.83599339570721e-06,
|
| 340 |
+
"loss": 0.0564,
|
| 341 |
"step": 1350
|
| 342 |
},
|
| 343 |
{
|
| 344 |
+
"epoch": 4.02,
|
| 345 |
+
"learning_rate": 8.801596037424328e-06,
|
| 346 |
+
"loss": 0.0574,
|
| 347 |
"step": 1375
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"epoch": 4.02,
|
| 351 |
+
"learning_rate": 8.767198679141443e-06,
|
| 352 |
+
"loss": 0.0623,
|
| 353 |
"step": 1400
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 4.02,
|
| 357 |
+
"learning_rate": 8.732801320858559e-06,
|
| 358 |
+
"loss": 0.0628,
|
| 359 |
"step": 1425
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 4.03,
|
| 363 |
+
"learning_rate": 8.698403962575674e-06,
|
| 364 |
+
"loss": 0.0579,
|
| 365 |
"step": 1450
|
| 366 |
},
|
| 367 |
{
|
| 368 |
+
"epoch": 4.03,
|
| 369 |
+
"learning_rate": 8.66400660429279e-06,
|
| 370 |
+
"loss": 0.0579,
|
| 371 |
"step": 1475
|
| 372 |
},
|
| 373 |
{
|
| 374 |
+
"epoch": 4.03,
|
| 375 |
+
"learning_rate": 8.629609246009907e-06,
|
| 376 |
+
"loss": 0.0613,
|
| 377 |
"step": 1500
|
| 378 |
},
|
| 379 |
{
|
| 380 |
+
"epoch": 4.04,
|
| 381 |
+
"learning_rate": 8.595211887727023e-06,
|
| 382 |
+
"loss": 0.0675,
|
| 383 |
"step": 1525
|
| 384 |
},
|
| 385 |
{
|
| 386 |
+
"epoch": 4.04,
|
| 387 |
+
"learning_rate": 8.56081452944414e-06,
|
| 388 |
+
"loss": 0.0606,
|
| 389 |
"step": 1550
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 5.0,
|
| 393 |
+
"learning_rate": 8.526417171161256e-06,
|
| 394 |
+
"loss": 0.0299,
|
| 395 |
"step": 1575
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 5.01,
|
| 399 |
+
"learning_rate": 8.492019812878371e-06,
|
| 400 |
+
"loss": 0.0246,
|
| 401 |
"step": 1600
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 5.01,
|
| 405 |
+
"learning_rate": 8.457622454595488e-06,
|
| 406 |
+
"loss": 0.0287,
|
| 407 |
"step": 1625
|
| 408 |
},
|
| 409 |
{
|
| 410 |
+
"epoch": 5.01,
|
| 411 |
+
"learning_rate": 8.423225096312604e-06,
|
| 412 |
+
"loss": 0.0307,
|
| 413 |
"step": 1650
|
| 414 |
},
|
| 415 |
{
|
| 416 |
+
"epoch": 5.02,
|
| 417 |
+
"learning_rate": 8.38882773802972e-06,
|
| 418 |
+
"loss": 0.0326,
|
| 419 |
"step": 1675
|
| 420 |
},
|
| 421 |
{
|
| 422 |
+
"epoch": 5.02,
|
| 423 |
+
"learning_rate": 8.354430379746837e-06,
|
| 424 |
+
"loss": 0.0312,
|
| 425 |
"step": 1700
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"epoch": 5.02,
|
| 429 |
+
"learning_rate": 8.320033021463952e-06,
|
| 430 |
+
"loss": 0.0326,
|
| 431 |
"step": 1725
|
| 432 |
},
|
| 433 |
{
|
| 434 |
+
"epoch": 5.03,
|
| 435 |
+
"learning_rate": 8.285635663181068e-06,
|
| 436 |
+
"loss": 0.0393,
|
| 437 |
"step": 1750
|
| 438 |
},
|
| 439 |
{
|
| 440 |
+
"epoch": 5.03,
|
| 441 |
+
"learning_rate": 8.251238304898185e-06,
|
| 442 |
+
"loss": 0.0381,
|
| 443 |
"step": 1775
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 5.03,
|
| 447 |
+
"learning_rate": 8.2168409466153e-06,
|
| 448 |
+
"loss": 0.0297,
|
| 449 |
"step": 1800
|
| 450 |
},
|
| 451 |
{
|
| 452 |
+
"epoch": 5.04,
|
| 453 |
+
"learning_rate": 8.182443588332416e-06,
|
| 454 |
+
"loss": 0.0327,
|
| 455 |
"step": 1825
|
| 456 |
},
|
| 457 |
{
|
| 458 |
+
"epoch": 5.04,
|
| 459 |
+
"learning_rate": 8.148046230049534e-06,
|
| 460 |
+
"loss": 0.0324,
|
| 461 |
"step": 1850
|
| 462 |
},
|
| 463 |
{
|
| 464 |
+
"epoch": 6.0,
|
| 465 |
+
"learning_rate": 8.11364887176665e-06,
|
| 466 |
+
"loss": 0.0233,
|
| 467 |
"step": 1875
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 6.01,
|
| 471 |
+
"learning_rate": 8.079251513483765e-06,
|
| 472 |
+
"loss": 0.0194,
|
| 473 |
"step": 1900
|
| 474 |
},
|
| 475 |
{
|
| 476 |
+
"epoch": 6.01,
|
| 477 |
+
"learning_rate": 8.04485415520088e-06,
|
| 478 |
+
"loss": 0.0206,
|
| 479 |
"step": 1925
|
| 480 |
},
|
| 481 |
{
|
| 482 |
+
"epoch": 6.01,
|
| 483 |
+
"learning_rate": 8.010456796917998e-06,
|
| 484 |
+
"loss": 0.0165,
|
| 485 |
"step": 1950
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 6.01,
|
| 489 |
+
"learning_rate": 7.976059438635113e-06,
|
| 490 |
+
"loss": 0.0217,
|
| 491 |
"step": 1975
|
| 492 |
},
|
| 493 |
{
|
| 494 |
+
"epoch": 6.02,
|
| 495 |
+
"learning_rate": 7.941662080352229e-06,
|
| 496 |
+
"loss": 0.0185,
|
| 497 |
"step": 2000
|
| 498 |
},
|
| 499 |
{
|
| 500 |
+
"epoch": 6.02,
|
| 501 |
+
"eval_loss": 0.47960373759269714,
|
| 502 |
+
"eval_runtime": 2219.3816,
|
| 503 |
+
"eval_samples_per_second": 0.581,
|
| 504 |
+
"eval_steps_per_second": 0.291,
|
| 505 |
+
"eval_wer": 30.16603445059387,
|
| 506 |
"step": 2000
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 6.02,
|
| 510 |
+
"learning_rate": 7.907264722069346e-06,
|
| 511 |
+
"loss": 0.0185,
|
| 512 |
"step": 2025
|
| 513 |
},
|
| 514 |
{
|
| 515 |
+
"epoch": 6.02,
|
| 516 |
+
"learning_rate": 7.872867363786462e-06,
|
| 517 |
+
"loss": 0.0158,
|
| 518 |
"step": 2050
|
| 519 |
},
|
| 520 |
{
|
| 521 |
+
"epoch": 6.03,
|
| 522 |
+
"learning_rate": 7.838470005503577e-06,
|
| 523 |
+
"loss": 0.0196,
|
| 524 |
"step": 2075
|
| 525 |
},
|
| 526 |
{
|
| 527 |
+
"epoch": 6.03,
|
| 528 |
+
"learning_rate": 7.804072647220694e-06,
|
| 529 |
+
"loss": 0.0224,
|
| 530 |
"step": 2100
|
| 531 |
},
|
| 532 |
{
|
| 533 |
+
"epoch": 6.03,
|
| 534 |
+
"learning_rate": 7.76967528893781e-06,
|
| 535 |
+
"loss": 0.0173,
|
| 536 |
"step": 2125
|
| 537 |
},
|
| 538 |
{
|
| 539 |
+
"epoch": 6.04,
|
| 540 |
+
"learning_rate": 7.735277930654927e-06,
|
| 541 |
+
"loss": 0.0208,
|
| 542 |
"step": 2150
|
| 543 |
},
|
| 544 |
{
|
| 545 |
+
"epoch": 7.0,
|
| 546 |
+
"learning_rate": 7.700880572372043e-06,
|
| 547 |
+
"loss": 0.018,
|
| 548 |
"step": 2175
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 7.0,
|
| 552 |
+
"learning_rate": 7.666483214089158e-06,
|
| 553 |
+
"loss": 0.0116,
|
| 554 |
"step": 2200
|
| 555 |
},
|
| 556 |
{
|
| 557 |
+
"epoch": 7.01,
|
| 558 |
+
"learning_rate": 7.632085855806276e-06,
|
| 559 |
+
"loss": 0.0141,
|
| 560 |
"step": 2225
|
| 561 |
},
|
| 562 |
{
|
| 563 |
+
"epoch": 7.01,
|
| 564 |
+
"learning_rate": 7.597688497523391e-06,
|
| 565 |
+
"loss": 0.013,
|
| 566 |
"step": 2250
|
| 567 |
},
|
| 568 |
{
|
| 569 |
+
"epoch": 7.01,
|
| 570 |
+
"learning_rate": 7.563291139240508e-06,
|
| 571 |
+
"loss": 0.0151,
|
| 572 |
"step": 2275
|
| 573 |
},
|
| 574 |
{
|
| 575 |
+
"epoch": 7.02,
|
| 576 |
+
"learning_rate": 7.528893780957623e-06,
|
| 577 |
+
"loss": 0.0129,
|
| 578 |
"step": 2300
|
| 579 |
},
|
| 580 |
{
|
| 581 |
+
"epoch": 7.02,
|
| 582 |
+
"learning_rate": 7.49449642267474e-06,
|
| 583 |
+
"loss": 0.0156,
|
| 584 |
"step": 2325
|
| 585 |
},
|
| 586 |
{
|
| 587 |
+
"epoch": 7.02,
|
| 588 |
+
"learning_rate": 7.460099064391856e-06,
|
| 589 |
+
"loss": 0.0122,
|
| 590 |
"step": 2350
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 7.03,
|
| 594 |
+
"learning_rate": 7.425701706108971e-06,
|
| 595 |
+
"loss": 0.0141,
|
| 596 |
"step": 2375
|
| 597 |
},
|
| 598 |
{
|
| 599 |
+
"epoch": 7.03,
|
| 600 |
+
"learning_rate": 7.391304347826087e-06,
|
| 601 |
+
"loss": 0.0164,
|
| 602 |
"step": 2400
|
| 603 |
},
|
| 604 |
{
|
| 605 |
+
"epoch": 7.03,
|
| 606 |
+
"learning_rate": 7.3569069895432036e-06,
|
| 607 |
+
"loss": 0.0134,
|
| 608 |
"step": 2425
|
| 609 |
},
|
| 610 |
{
|
| 611 |
+
"epoch": 7.04,
|
| 612 |
+
"learning_rate": 7.322509631260319e-06,
|
| 613 |
+
"loss": 0.0135,
|
| 614 |
"step": 2450
|
| 615 |
},
|
| 616 |
{
|
| 617 |
+
"epoch": 7.04,
|
| 618 |
+
"learning_rate": 7.2881122729774356e-06,
|
| 619 |
+
"loss": 0.0135,
|
| 620 |
"step": 2475
|
| 621 |
},
|
| 622 |
{
|
| 623 |
+
"epoch": 8.0,
|
| 624 |
+
"learning_rate": 7.253714914694552e-06,
|
| 625 |
+
"loss": 0.0082,
|
| 626 |
"step": 2500
|
| 627 |
},
|
| 628 |
{
|
| 629 |
+
"epoch": 8.01,
|
| 630 |
+
"learning_rate": 7.219317556411668e-06,
|
| 631 |
+
"loss": 0.0083,
|
| 632 |
"step": 2525
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 8.01,
|
| 636 |
+
"learning_rate": 7.184920198128784e-06,
|
| 637 |
+
"loss": 0.0096,
|
| 638 |
"step": 2550
|
| 639 |
},
|
| 640 |
{
|
| 641 |
+
"epoch": 8.01,
|
| 642 |
+
"learning_rate": 7.1505228398459e-06,
|
| 643 |
+
"loss": 0.0091,
|
| 644 |
"step": 2575
|
| 645 |
},
|
| 646 |
{
|
| 647 |
+
"epoch": 8.02,
|
| 648 |
+
"learning_rate": 7.116125481563017e-06,
|
| 649 |
+
"loss": 0.0136,
|
| 650 |
"step": 2600
|
| 651 |
},
|
| 652 |
{
|
| 653 |
+
"epoch": 8.02,
|
| 654 |
+
"learning_rate": 7.081728123280132e-06,
|
| 655 |
+
"loss": 0.0088,
|
| 656 |
"step": 2625
|
| 657 |
},
|
| 658 |
{
|
| 659 |
+
"epoch": 8.02,
|
| 660 |
+
"learning_rate": 7.047330764997249e-06,
|
| 661 |
+
"loss": 0.0121,
|
| 662 |
"step": 2650
|
| 663 |
},
|
| 664 |
{
|
| 665 |
+
"epoch": 8.03,
|
| 666 |
+
"learning_rate": 7.012933406714365e-06,
|
| 667 |
+
"loss": 0.0084,
|
| 668 |
"step": 2675
|
| 669 |
},
|
| 670 |
{
|
| 671 |
+
"epoch": 8.03,
|
| 672 |
+
"learning_rate": 6.978536048431482e-06,
|
| 673 |
+
"loss": 0.0072,
|
| 674 |
"step": 2700
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 8.03,
|
| 678 |
+
"learning_rate": 6.944138690148597e-06,
|
| 679 |
+
"loss": 0.0117,
|
| 680 |
"step": 2725
|
| 681 |
},
|
| 682 |
{
|
| 683 |
+
"epoch": 8.03,
|
| 684 |
+
"learning_rate": 6.909741331865714e-06,
|
| 685 |
+
"loss": 0.0062,
|
| 686 |
"step": 2750
|
| 687 |
},
|
| 688 |
{
|
| 689 |
+
"epoch": 8.04,
|
| 690 |
+
"learning_rate": 6.87534397358283e-06,
|
| 691 |
+
"loss": 0.0102,
|
| 692 |
"step": 2775
|
| 693 |
},
|
| 694 |
{
|
| 695 |
+
"epoch": 9.0,
|
| 696 |
+
"learning_rate": 6.840946615299946e-06,
|
| 697 |
+
"loss": 0.0088,
|
| 698 |
"step": 2800
|
| 699 |
},
|
| 700 |
{
|
| 701 |
+
"epoch": 9.0,
|
| 702 |
+
"learning_rate": 6.806549257017062e-06,
|
| 703 |
+
"loss": 0.0071,
|
| 704 |
"step": 2825
|
| 705 |
},
|
| 706 |
{
|
| 707 |
+
"epoch": 9.01,
|
| 708 |
+
"learning_rate": 6.772151898734178e-06,
|
| 709 |
+
"loss": 0.0063,
|
| 710 |
"step": 2850
|
| 711 |
},
|
| 712 |
{
|
| 713 |
+
"epoch": 9.01,
|
| 714 |
+
"learning_rate": 6.737754540451293e-06,
|
| 715 |
+
"loss": 0.0076,
|
| 716 |
"step": 2875
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 9.01,
|
| 720 |
+
"learning_rate": 6.70335718216841e-06,
|
| 721 |
+
"loss": 0.0094,
|
| 722 |
"step": 2900
|
| 723 |
},
|
| 724 |
{
|
| 725 |
+
"epoch": 9.02,
|
| 726 |
+
"learning_rate": 6.668959823885526e-06,
|
| 727 |
+
"loss": 0.0053,
|
| 728 |
"step": 2925
|
| 729 |
},
|
| 730 |
{
|
| 731 |
+
"epoch": 9.02,
|
| 732 |
+
"learning_rate": 6.6345624656026416e-06,
|
| 733 |
+
"loss": 0.0078,
|
| 734 |
"step": 2950
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 9.02,
|
| 738 |
+
"learning_rate": 6.600165107319758e-06,
|
| 739 |
+
"loss": 0.0078,
|
| 740 |
"step": 2975
|
| 741 |
},
|
| 742 |
{
|
| 743 |
+
"epoch": 9.03,
|
| 744 |
+
"learning_rate": 6.565767749036874e-06,
|
| 745 |
+
"loss": 0.0062,
|
| 746 |
"step": 3000
|
| 747 |
},
|
| 748 |
{
|
| 749 |
+
"epoch": 9.03,
|
| 750 |
+
"eval_loss": 0.5142830014228821,
|
| 751 |
+
"eval_runtime": 2235.4063,
|
| 752 |
+
"eval_samples_per_second": 0.577,
|
| 753 |
+
"eval_steps_per_second": 0.289,
|
| 754 |
+
"eval_wer": 30.041664075617184,
|
| 755 |
"step": 3000
|
| 756 |
},
|
| 757 |
{
|
| 758 |
+
"epoch": 9.03,
|
| 759 |
+
"learning_rate": 6.531370390753991e-06,
|
| 760 |
+
"loss": 0.0044,
|
| 761 |
"step": 3025
|
| 762 |
},
|
| 763 |
{
|
| 764 |
+
"epoch": 9.03,
|
| 765 |
+
"learning_rate": 6.496973032471106e-06,
|
| 766 |
+
"loss": 0.0084,
|
| 767 |
"step": 3050
|
| 768 |
},
|
| 769 |
{
|
| 770 |
+
"epoch": 9.04,
|
| 771 |
+
"learning_rate": 6.462575674188223e-06,
|
| 772 |
+
"loss": 0.0068,
|
| 773 |
"step": 3075
|
| 774 |
},
|
| 775 |
{
|
| 776 |
+
"epoch": 9.04,
|
| 777 |
+
"learning_rate": 6.428178315905339e-06,
|
| 778 |
+
"loss": 0.008,
|
| 779 |
"step": 3100
|
| 780 |
},
|
| 781 |
{
|
| 782 |
+
"epoch": 10.0,
|
| 783 |
+
"learning_rate": 6.393780957622455e-06,
|
| 784 |
+
"loss": 0.0047,
|
| 785 |
"step": 3125
|
| 786 |
},
|
| 787 |
{
|
| 788 |
+
"epoch": 10.01,
|
| 789 |
+
"learning_rate": 6.359383599339571e-06,
|
| 790 |
+
"loss": 0.0049,
|
| 791 |
"step": 3150
|
| 792 |
},
|
| 793 |
{
|
| 794 |
+
"epoch": 10.01,
|
| 795 |
+
"learning_rate": 6.324986241056688e-06,
|
| 796 |
+
"loss": 0.0048,
|
| 797 |
"step": 3175
|
| 798 |
},
|
| 799 |
{
|
| 800 |
+
"epoch": 10.01,
|
| 801 |
+
"learning_rate": 6.290588882773804e-06,
|
| 802 |
+
"loss": 0.0036,
|
| 803 |
"step": 3200
|
| 804 |
},
|
| 805 |
{
|
| 806 |
+
"epoch": 10.02,
|
| 807 |
+
"learning_rate": 6.25619152449092e-06,
|
| 808 |
+
"loss": 0.0045,
|
| 809 |
"step": 3225
|
| 810 |
},
|
| 811 |
{
|
| 812 |
+
"epoch": 10.02,
|
| 813 |
+
"learning_rate": 6.221794166208036e-06,
|
| 814 |
+
"loss": 0.006,
|
| 815 |
"step": 3250
|
| 816 |
},
|
| 817 |
{
|
| 818 |
+
"epoch": 10.02,
|
| 819 |
+
"learning_rate": 6.1873968079251525e-06,
|
| 820 |
+
"loss": 0.0045,
|
| 821 |
"step": 3275
|
| 822 |
},
|
| 823 |
{
|
| 824 |
+
"epoch": 10.03,
|
| 825 |
+
"learning_rate": 6.152999449642269e-06,
|
| 826 |
+
"loss": 0.0046,
|
| 827 |
"step": 3300
|
| 828 |
},
|
| 829 |
{
|
| 830 |
+
"epoch": 10.03,
|
| 831 |
+
"learning_rate": 6.1186020913593845e-06,
|
| 832 |
+
"loss": 0.0052,
|
| 833 |
"step": 3325
|
| 834 |
},
|
| 835 |
{
|
| 836 |
+
"epoch": 10.03,
|
| 837 |
+
"learning_rate": 6.0842047330765e-06,
|
| 838 |
+
"loss": 0.0036,
|
| 839 |
"step": 3350
|
| 840 |
},
|
| 841 |
{
|
| 842 |
+
"epoch": 10.04,
|
| 843 |
+
"learning_rate": 6.049807374793616e-06,
|
| 844 |
+
"loss": 0.0064,
|
| 845 |
"step": 3375
|
| 846 |
},
|
| 847 |
{
|
| 848 |
+
"epoch": 10.04,
|
| 849 |
+
"learning_rate": 6.015410016510732e-06,
|
| 850 |
+
"loss": 0.0062,
|
| 851 |
"step": 3400
|
| 852 |
},
|
| 853 |
{
|
| 854 |
+
"epoch": 11.0,
|
| 855 |
+
"learning_rate": 5.9810126582278484e-06,
|
| 856 |
+
"loss": 0.0058,
|
| 857 |
"step": 3425
|
| 858 |
},
|
| 859 |
{
|
| 860 |
+
"epoch": 11.01,
|
| 861 |
+
"learning_rate": 5.946615299944965e-06,
|
| 862 |
+
"loss": 0.0042,
|
| 863 |
"step": 3450
|
| 864 |
},
|
| 865 |
{
|
| 866 |
+
"epoch": 11.01,
|
| 867 |
+
"learning_rate": 5.9122179416620804e-06,
|
| 868 |
+
"loss": 0.0038,
|
| 869 |
"step": 3475
|
| 870 |
},
|
| 871 |
{
|
| 872 |
+
"epoch": 11.01,
|
| 873 |
+
"learning_rate": 5.877820583379197e-06,
|
| 874 |
+
"loss": 0.0036,
|
| 875 |
"step": 3500
|
| 876 |
},
|
| 877 |
{
|
| 878 |
+
"epoch": 11.01,
|
| 879 |
+
"learning_rate": 5.843423225096313e-06,
|
| 880 |
+
"loss": 0.0067,
|
| 881 |
"step": 3525
|
| 882 |
},
|
| 883 |
{
|
| 884 |
+
"epoch": 11.02,
|
| 885 |
+
"learning_rate": 5.809025866813429e-06,
|
| 886 |
+
"loss": 0.0063,
|
| 887 |
"step": 3550
|
| 888 |
},
|
| 889 |
{
|
| 890 |
+
"epoch": 11.02,
|
| 891 |
+
"learning_rate": 5.774628508530545e-06,
|
| 892 |
+
"loss": 0.0045,
|
| 893 |
"step": 3575
|
| 894 |
},
|
| 895 |
{
|
| 896 |
+
"epoch": 11.02,
|
| 897 |
+
"learning_rate": 5.740231150247662e-06,
|
| 898 |
+
"loss": 0.0032,
|
| 899 |
"step": 3600
|
| 900 |
},
|
| 901 |
{
|
| 902 |
+
"epoch": 11.03,
|
| 903 |
+
"learning_rate": 5.705833791964778e-06,
|
| 904 |
+
"loss": 0.0046,
|
| 905 |
"step": 3625
|
| 906 |
},
|
| 907 |
{
|
| 908 |
+
"epoch": 11.03,
|
| 909 |
+
"learning_rate": 5.671436433681894e-06,
|
| 910 |
+
"loss": 0.005,
|
| 911 |
"step": 3650
|
| 912 |
},
|
| 913 |
{
|
| 914 |
+
"epoch": 11.03,
|
| 915 |
+
"learning_rate": 5.63703907539901e-06,
|
| 916 |
+
"loss": 0.004,
|
| 917 |
"step": 3675
|
| 918 |
},
|
| 919 |
{
|
| 920 |
+
"epoch": 11.04,
|
| 921 |
+
"learning_rate": 5.6026417171161265e-06,
|
| 922 |
+
"loss": 0.0037,
|
| 923 |
"step": 3700
|
| 924 |
},
|
| 925 |
{
|
| 926 |
+
"epoch": 12.0,
|
| 927 |
+
"learning_rate": 5.568244358833242e-06,
|
| 928 |
+
"loss": 0.0024,
|
| 929 |
"step": 3725
|
| 930 |
},
|
| 931 |
{
|
| 932 |
+
"epoch": 12.0,
|
| 933 |
+
"learning_rate": 5.5338470005503585e-06,
|
| 934 |
+
"loss": 0.002,
|
| 935 |
"step": 3750
|
| 936 |
},
|
| 937 |
{
|
| 938 |
+
"epoch": 12.01,
|
| 939 |
+
"learning_rate": 5.499449642267475e-06,
|
| 940 |
+
"loss": 0.0014,
|
| 941 |
"step": 3775
|
| 942 |
},
|
| 943 |
{
|
| 944 |
+
"epoch": 12.01,
|
| 945 |
+
"learning_rate": 5.465052283984591e-06,
|
| 946 |
+
"loss": 0.0039,
|
| 947 |
"step": 3800
|
| 948 |
},
|
| 949 |
{
|
| 950 |
+
"epoch": 12.01,
|
| 951 |
+
"learning_rate": 5.430654925701706e-06,
|
| 952 |
+
"loss": 0.0025,
|
| 953 |
"step": 3825
|
| 954 |
},
|
| 955 |
{
|
| 956 |
+
"epoch": 12.02,
|
| 957 |
+
"learning_rate": 5.3962575674188225e-06,
|
| 958 |
+
"loss": 0.0026,
|
| 959 |
"step": 3850
|
| 960 |
},
|
| 961 |
{
|
| 962 |
+
"epoch": 12.02,
|
| 963 |
+
"learning_rate": 5.361860209135938e-06,
|
| 964 |
+
"loss": 0.0013,
|
| 965 |
"step": 3875
|
| 966 |
},
|
| 967 |
{
|
| 968 |
+
"epoch": 12.02,
|
| 969 |
+
"learning_rate": 5.3274628508530544e-06,
|
| 970 |
+
"loss": 0.0027,
|
| 971 |
"step": 3900
|
| 972 |
},
|
| 973 |
{
|
| 974 |
+
"epoch": 12.03,
|
| 975 |
+
"learning_rate": 5.293065492570171e-06,
|
| 976 |
+
"loss": 0.0022,
|
| 977 |
"step": 3925
|
| 978 |
},
|
| 979 |
{
|
| 980 |
+
"epoch": 12.03,
|
| 981 |
+
"learning_rate": 5.258668134287287e-06,
|
| 982 |
+
"loss": 0.0024,
|
| 983 |
"step": 3950
|
| 984 |
},
|
| 985 |
{
|
| 986 |
+
"epoch": 12.03,
|
| 987 |
+
"learning_rate": 5.224270776004403e-06,
|
| 988 |
+
"loss": 0.0034,
|
| 989 |
"step": 3975
|
| 990 |
},
|
| 991 |
{
|
| 992 |
+
"epoch": 12.04,
|
| 993 |
+
"learning_rate": 5.189873417721519e-06,
|
| 994 |
+
"loss": 0.0022,
|
| 995 |
"step": 4000
|
| 996 |
},
|
| 997 |
{
|
| 998 |
+
"epoch": 12.04,
|
| 999 |
+
"eval_loss": 0.5468949675559998,
|
| 1000 |
+
"eval_runtime": 2225.9485,
|
| 1001 |
+
"eval_samples_per_second": 0.579,
|
| 1002 |
+
"eval_steps_per_second": 0.29,
|
| 1003 |
+
"eval_wer": 28.630060319631863,
|
| 1004 |
"step": 4000
|
| 1005 |
},
|
| 1006 |
{
|
| 1007 |
+
"epoch": 12.04,
|
| 1008 |
+
"learning_rate": 5.155476059438636e-06,
|
| 1009 |
+
"loss": 0.0028,
|
| 1010 |
"step": 4025
|
| 1011 |
},
|
| 1012 |
{
|
| 1013 |
+
"epoch": 13.0,
|
| 1014 |
+
"learning_rate": 5.121078701155751e-06,
|
| 1015 |
+
"loss": 0.0019,
|
| 1016 |
"step": 4050
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
+
"epoch": 13.01,
|
| 1020 |
+
"learning_rate": 5.086681342872868e-06,
|
| 1021 |
+
"loss": 0.0022,
|
| 1022 |
"step": 4075
|
| 1023 |
},
|
| 1024 |
{
|
| 1025 |
+
"epoch": 13.01,
|
| 1026 |
+
"learning_rate": 5.052283984589984e-06,
|
| 1027 |
+
"loss": 0.0015,
|
| 1028 |
"step": 4100
|
| 1029 |
},
|
| 1030 |
{
|
| 1031 |
+
"epoch": 13.01,
|
| 1032 |
+
"learning_rate": 5.0178866263071005e-06,
|
| 1033 |
+
"loss": 0.0015,
|
| 1034 |
"step": 4125
|
| 1035 |
},
|
| 1036 |
{
|
| 1037 |
+
"epoch": 13.02,
|
| 1038 |
+
"learning_rate": 4.983489268024216e-06,
|
| 1039 |
+
"loss": 0.0027,
|
| 1040 |
"step": 4150
|
| 1041 |
},
|
| 1042 |
{
|
| 1043 |
+
"epoch": 13.02,
|
| 1044 |
+
"learning_rate": 4.949091909741332e-06,
|
| 1045 |
+
"loss": 0.0024,
|
| 1046 |
"step": 4175
|
| 1047 |
},
|
| 1048 |
{
|
| 1049 |
+
"epoch": 13.02,
|
| 1050 |
+
"learning_rate": 4.914694551458448e-06,
|
| 1051 |
+
"loss": 0.0016,
|
| 1052 |
"step": 4200
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
+
"epoch": 13.03,
|
| 1056 |
+
"learning_rate": 4.8802971931755645e-06,
|
| 1057 |
+
"loss": 0.0027,
|
| 1058 |
"step": 4225
|
| 1059 |
},
|
| 1060 |
{
|
| 1061 |
+
"epoch": 13.03,
|
| 1062 |
+
"learning_rate": 4.845899834892681e-06,
|
| 1063 |
+
"loss": 0.0023,
|
| 1064 |
"step": 4250
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
+
"epoch": 13.03,
|
| 1068 |
+
"learning_rate": 4.8115024766097965e-06,
|
| 1069 |
+
"loss": 0.0023,
|
| 1070 |
"step": 4275
|
| 1071 |
},
|
| 1072 |
{
|
| 1073 |
+
"epoch": 13.03,
|
| 1074 |
+
"learning_rate": 4.777105118326913e-06,
|
| 1075 |
+
"loss": 0.0026,
|
| 1076 |
"step": 4300
|
| 1077 |
},
|
| 1078 |
{
|
| 1079 |
+
"epoch": 13.04,
|
| 1080 |
+
"learning_rate": 4.742707760044029e-06,
|
| 1081 |
+
"loss": 0.002,
|
| 1082 |
"step": 4325
|
| 1083 |
},
|
| 1084 |
{
|
| 1085 |
+
"epoch": 14.0,
|
| 1086 |
+
"learning_rate": 4.708310401761145e-06,
|
| 1087 |
+
"loss": 0.0009,
|
| 1088 |
"step": 4350
|
| 1089 |
},
|
| 1090 |
{
|
| 1091 |
+
"epoch": 14.0,
|
| 1092 |
+
"learning_rate": 4.673913043478261e-06,
|
| 1093 |
+
"loss": 0.0034,
|
| 1094 |
"step": 4375
|
| 1095 |
},
|
| 1096 |
{
|
| 1097 |
+
"epoch": 14.01,
|
| 1098 |
+
"learning_rate": 4.639515685195378e-06,
|
| 1099 |
+
"loss": 0.0012,
|
| 1100 |
"step": 4400
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
+
"epoch": 14.01,
|
| 1104 |
+
"learning_rate": 4.605118326912493e-06,
|
| 1105 |
+
"loss": 0.002,
|
| 1106 |
"step": 4425
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
+
"epoch": 14.01,
|
| 1110 |
+
"learning_rate": 4.57072096862961e-06,
|
| 1111 |
+
"loss": 0.0007,
|
| 1112 |
"step": 4450
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
+
"epoch": 14.02,
|
| 1116 |
+
"learning_rate": 4.536323610346725e-06,
|
| 1117 |
+
"loss": 0.0015,
|
| 1118 |
"step": 4475
|
| 1119 |
},
|
| 1120 |
{
|
| 1121 |
+
"epoch": 14.02,
|
| 1122 |
+
"learning_rate": 4.501926252063842e-06,
|
| 1123 |
+
"loss": 0.0008,
|
| 1124 |
"step": 4500
|
| 1125 |
},
|
| 1126 |
{
|
| 1127 |
+
"epoch": 14.02,
|
| 1128 |
+
"learning_rate": 4.467528893780958e-06,
|
| 1129 |
+
"loss": 0.0019,
|
| 1130 |
"step": 4525
|
| 1131 |
},
|
| 1132 |
{
|
| 1133 |
+
"epoch": 14.03,
|
| 1134 |
+
"learning_rate": 4.4331315354980745e-06,
|
| 1135 |
+
"loss": 0.0013,
|
| 1136 |
"step": 4550
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
+
"epoch": 14.03,
|
| 1140 |
+
"learning_rate": 4.39873417721519e-06,
|
| 1141 |
+
"loss": 0.0017,
|
| 1142 |
"step": 4575
|
| 1143 |
},
|
| 1144 |
{
|
| 1145 |
+
"epoch": 14.03,
|
| 1146 |
+
"learning_rate": 4.3643368189323065e-06,
|
| 1147 |
+
"loss": 0.0031,
|
| 1148 |
"step": 4600
|
| 1149 |
},
|
| 1150 |
{
|
| 1151 |
+
"epoch": 14.04,
|
| 1152 |
+
"learning_rate": 4.329939460649423e-06,
|
| 1153 |
+
"loss": 0.0014,
|
| 1154 |
"step": 4625
|
| 1155 |
},
|
| 1156 |
{
|
| 1157 |
+
"epoch": 14.04,
|
| 1158 |
+
"learning_rate": 4.2955421023665385e-06,
|
| 1159 |
+
"loss": 0.001,
|
| 1160 |
"step": 4650
|
| 1161 |
},
|
| 1162 |
{
|
| 1163 |
+
"epoch": 15.0,
|
| 1164 |
+
"learning_rate": 4.261144744083655e-06,
|
| 1165 |
+
"loss": 0.0006,
|
| 1166 |
"step": 4675
|
| 1167 |
},
|
| 1168 |
{
|
| 1169 |
+
"epoch": 15.01,
|
| 1170 |
+
"learning_rate": 4.2267473858007705e-06,
|
| 1171 |
+
"loss": 0.0004,
|
| 1172 |
"step": 4700
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
+
"epoch": 15.01,
|
| 1176 |
+
"learning_rate": 4.192350027517887e-06,
|
| 1177 |
+
"loss": 0.0014,
|
| 1178 |
"step": 4725
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
+
"epoch": 15.01,
|
| 1182 |
+
"learning_rate": 4.157952669235003e-06,
|
| 1183 |
+
"loss": 0.0009,
|
| 1184 |
"step": 4750
|
| 1185 |
},
|
| 1186 |
{
|
| 1187 |
+
"epoch": 15.02,
|
| 1188 |
+
"learning_rate": 4.123555310952119e-06,
|
| 1189 |
+
"loss": 0.0006,
|
| 1190 |
"step": 4775
|
| 1191 |
},
|
| 1192 |
{
|
| 1193 |
+
"epoch": 15.02,
|
| 1194 |
+
"learning_rate": 4.089157952669235e-06,
|
| 1195 |
+
"loss": 0.001,
|
| 1196 |
"step": 4800
|
| 1197 |
},
|
| 1198 |
{
|
| 1199 |
+
"epoch": 15.02,
|
| 1200 |
+
"learning_rate": 4.054760594386352e-06,
|
| 1201 |
+
"loss": 0.0005,
|
| 1202 |
"step": 4825
|
| 1203 |
},
|
| 1204 |
{
|
| 1205 |
+
"epoch": 15.03,
|
| 1206 |
+
"learning_rate": 4.020363236103468e-06,
|
| 1207 |
+
"loss": 0.0005,
|
| 1208 |
"step": 4850
|
| 1209 |
},
|
| 1210 |
{
|
| 1211 |
+
"epoch": 15.03,
|
| 1212 |
+
"learning_rate": 3.985965877820584e-06,
|
| 1213 |
+
"loss": 0.0014,
|
| 1214 |
"step": 4875
|
| 1215 |
},
|
| 1216 |
{
|
| 1217 |
+
"epoch": 15.03,
|
| 1218 |
+
"learning_rate": 3.951568519537699e-06,
|
| 1219 |
+
"loss": 0.0004,
|
| 1220 |
"step": 4900
|
| 1221 |
},
|
| 1222 |
{
|
| 1223 |
+
"epoch": 15.04,
|
| 1224 |
+
"learning_rate": 3.917171161254816e-06,
|
| 1225 |
+
"loss": 0.0005,
|
| 1226 |
"step": 4925
|
| 1227 |
},
|
| 1228 |
{
|
| 1229 |
+
"epoch": 15.04,
|
| 1230 |
+
"learning_rate": 3.882773802971932e-06,
|
| 1231 |
+
"loss": 0.0007,
|
| 1232 |
"step": 4950
|
| 1233 |
},
|
| 1234 |
{
|
| 1235 |
+
"epoch": 16.0,
|
| 1236 |
+
"learning_rate": 3.8483764446890486e-06,
|
| 1237 |
+
"loss": 0.0005,
|
| 1238 |
"step": 4975
|
| 1239 |
},
|
| 1240 |
{
|
| 1241 |
+
"epoch": 16.01,
|
| 1242 |
+
"learning_rate": 3.813979086406164e-06,
|
| 1243 |
+
"loss": 0.0004,
|
| 1244 |
"step": 5000
|
| 1245 |
},
|
| 1246 |
{
|
| 1247 |
+
"epoch": 16.01,
|
| 1248 |
+
"eval_loss": 0.569535493850708,
|
| 1249 |
+
"eval_runtime": 2212.0775,
|
| 1250 |
"eval_samples_per_second": 0.583,
|
| 1251 |
"eval_steps_per_second": 0.292,
|
| 1252 |
+
"eval_wer": 27.952241776008957,
|
| 1253 |
"step": 5000
|
| 1254 |
},
|
| 1255 |
{
|
| 1256 |
+
"epoch": 16.01,
|
| 1257 |
+
"learning_rate": 3.7795817281232806e-06,
|
| 1258 |
+
"loss": 0.0006,
|
| 1259 |
"step": 5025
|
| 1260 |
},
|
| 1261 |
{
|
| 1262 |
+
"epoch": 16.01,
|
| 1263 |
+
"learning_rate": 3.7451843698403965e-06,
|
| 1264 |
+
"loss": 0.0012,
|
| 1265 |
"step": 5050
|
| 1266 |
},
|
| 1267 |
{
|
| 1268 |
+
"epoch": 16.01,
|
| 1269 |
+
"learning_rate": 3.710787011557513e-06,
|
| 1270 |
+
"loss": 0.0005,
|
| 1271 |
"step": 5075
|
| 1272 |
},
|
| 1273 |
{
|
| 1274 |
+
"epoch": 16.02,
|
| 1275 |
+
"learning_rate": 3.676389653274629e-06,
|
| 1276 |
+
"loss": 0.0004,
|
| 1277 |
"step": 5100
|
| 1278 |
},
|
| 1279 |
{
|
| 1280 |
+
"epoch": 16.02,
|
| 1281 |
+
"learning_rate": 3.6419922949917445e-06,
|
| 1282 |
+
"loss": 0.0007,
|
| 1283 |
"step": 5125
|
| 1284 |
},
|
| 1285 |
{
|
| 1286 |
+
"epoch": 16.02,
|
| 1287 |
+
"learning_rate": 3.607594936708861e-06,
|
| 1288 |
+
"loss": 0.0007,
|
| 1289 |
"step": 5150
|
| 1290 |
},
|
| 1291 |
{
|
| 1292 |
+
"epoch": 16.03,
|
| 1293 |
+
"learning_rate": 3.573197578425977e-06,
|
| 1294 |
+
"loss": 0.0004,
|
| 1295 |
"step": 5175
|
| 1296 |
},
|
| 1297 |
{
|
| 1298 |
+
"epoch": 16.03,
|
| 1299 |
+
"learning_rate": 3.5388002201430934e-06,
|
| 1300 |
+
"loss": 0.0003,
|
| 1301 |
"step": 5200
|
| 1302 |
},
|
| 1303 |
{
|
| 1304 |
+
"epoch": 16.03,
|
| 1305 |
+
"learning_rate": 3.5044028618602094e-06,
|
| 1306 |
+
"loss": 0.0004,
|
| 1307 |
"step": 5225
|
| 1308 |
},
|
| 1309 |
{
|
| 1310 |
+
"epoch": 16.04,
|
| 1311 |
+
"learning_rate": 3.4700055035773258e-06,
|
| 1312 |
+
"loss": 0.0004,
|
| 1313 |
"step": 5250
|
| 1314 |
},
|
| 1315 |
{
|
| 1316 |
+
"epoch": 17.0,
|
| 1317 |
+
"learning_rate": 3.4356081452944418e-06,
|
| 1318 |
+
"loss": 0.0002,
|
| 1319 |
"step": 5275
|
| 1320 |
},
|
| 1321 |
{
|
| 1322 |
+
"epoch": 17.0,
|
| 1323 |
+
"learning_rate": 3.4012107870115578e-06,
|
| 1324 |
+
"loss": 0.0002,
|
| 1325 |
"step": 5300
|
| 1326 |
},
|
| 1327 |
{
|
| 1328 |
+
"epoch": 17.01,
|
| 1329 |
+
"learning_rate": 3.366813428728674e-06,
|
| 1330 |
+
"loss": 0.0002,
|
| 1331 |
"step": 5325
|
| 1332 |
},
|
| 1333 |
{
|
| 1334 |
+
"epoch": 17.01,
|
| 1335 |
+
"learning_rate": 3.33241607044579e-06,
|
| 1336 |
+
"loss": 0.0001,
|
| 1337 |
"step": 5350
|
| 1338 |
},
|
| 1339 |
{
|
| 1340 |
+
"epoch": 17.01,
|
| 1341 |
+
"learning_rate": 3.2980187121629057e-06,
|
| 1342 |
+
"loss": 0.0001,
|
| 1343 |
"step": 5375
|
| 1344 |
},
|
| 1345 |
{
|
| 1346 |
+
"epoch": 17.02,
|
| 1347 |
+
"learning_rate": 3.263621353880022e-06,
|
| 1348 |
+
"loss": 0.0002,
|
| 1349 |
"step": 5400
|
| 1350 |
},
|
| 1351 |
{
|
| 1352 |
+
"epoch": 17.02,
|
| 1353 |
+
"learning_rate": 3.229223995597138e-06,
|
| 1354 |
+
"loss": 0.0001,
|
| 1355 |
"step": 5425
|
| 1356 |
},
|
| 1357 |
{
|
| 1358 |
+
"epoch": 17.02,
|
| 1359 |
+
"learning_rate": 3.1948266373142546e-06,
|
| 1360 |
+
"loss": 0.0004,
|
| 1361 |
"step": 5450
|
| 1362 |
},
|
| 1363 |
{
|
| 1364 |
+
"epoch": 17.03,
|
| 1365 |
+
"learning_rate": 3.1604292790313706e-06,
|
| 1366 |
+
"loss": 0.0002,
|
| 1367 |
"step": 5475
|
| 1368 |
},
|
| 1369 |
{
|
| 1370 |
+
"epoch": 17.03,
|
| 1371 |
+
"learning_rate": 3.126031920748487e-06,
|
| 1372 |
+
"loss": 0.0002,
|
| 1373 |
"step": 5500
|
| 1374 |
},
|
| 1375 |
{
|
| 1376 |
+
"epoch": 17.03,
|
| 1377 |
+
"learning_rate": 3.091634562465603e-06,
|
| 1378 |
+
"loss": 0.0002,
|
| 1379 |
"step": 5525
|
| 1380 |
},
|
| 1381 |
{
|
| 1382 |
+
"epoch": 17.04,
|
| 1383 |
+
"learning_rate": 3.057237204182719e-06,
|
| 1384 |
+
"loss": 0.0017,
|
| 1385 |
"step": 5550
|
| 1386 |
},
|
| 1387 |
{
|
| 1388 |
+
"epoch": 17.04,
|
| 1389 |
+
"learning_rate": 3.0228398458998354e-06,
|
| 1390 |
+
"loss": 0.0002,
|
| 1391 |
"step": 5575
|
| 1392 |
},
|
| 1393 |
{
|
| 1394 |
+
"epoch": 18.0,
|
| 1395 |
+
"learning_rate": 2.988442487616951e-06,
|
| 1396 |
+
"loss": 0.0004,
|
| 1397 |
"step": 5600
|
| 1398 |
},
|
| 1399 |
{
|
| 1400 |
+
"epoch": 18.01,
|
| 1401 |
+
"learning_rate": 2.9540451293340674e-06,
|
| 1402 |
+
"loss": 0.0001,
|
| 1403 |
"step": 5625
|
| 1404 |
},
|
| 1405 |
{
|
| 1406 |
+
"epoch": 18.01,
|
| 1407 |
+
"learning_rate": 2.9196477710511834e-06,
|
| 1408 |
+
"loss": 0.0002,
|
| 1409 |
"step": 5650
|
| 1410 |
},
|
| 1411 |
{
|
| 1412 |
+
"epoch": 18.01,
|
| 1413 |
+
"learning_rate": 2.8852504127682994e-06,
|
| 1414 |
+
"loss": 0.0001,
|
| 1415 |
"step": 5675
|
| 1416 |
},
|
| 1417 |
{
|
| 1418 |
+
"epoch": 18.02,
|
| 1419 |
+
"learning_rate": 2.850853054485416e-06,
|
| 1420 |
+
"loss": 0.0001,
|
| 1421 |
"step": 5700
|
| 1422 |
},
|
| 1423 |
{
|
| 1424 |
+
"epoch": 18.02,
|
| 1425 |
+
"learning_rate": 2.8164556962025318e-06,
|
| 1426 |
+
"loss": 0.0002,
|
| 1427 |
"step": 5725
|
| 1428 |
},
|
| 1429 |
{
|
| 1430 |
+
"epoch": 18.02,
|
| 1431 |
+
"learning_rate": 2.782058337919648e-06,
|
| 1432 |
+
"loss": 0.0003,
|
| 1433 |
"step": 5750
|
| 1434 |
},
|
| 1435 |
{
|
| 1436 |
+
"epoch": 18.03,
|
| 1437 |
+
"learning_rate": 2.747660979636764e-06,
|
| 1438 |
+
"loss": 0.0001,
|
| 1439 |
"step": 5775
|
| 1440 |
},
|
| 1441 |
{
|
| 1442 |
+
"epoch": 18.03,
|
| 1443 |
+
"learning_rate": 2.7132636213538806e-06,
|
| 1444 |
+
"loss": 0.0001,
|
| 1445 |
"step": 5800
|
| 1446 |
},
|
| 1447 |
{
|
| 1448 |
+
"epoch": 18.03,
|
| 1449 |
+
"learning_rate": 2.6788662630709966e-06,
|
| 1450 |
+
"loss": 0.0001,
|
| 1451 |
"step": 5825
|
| 1452 |
},
|
| 1453 |
{
|
| 1454 |
+
"epoch": 18.03,
|
| 1455 |
+
"learning_rate": 2.644468904788112e-06,
|
| 1456 |
+
"loss": 0.0001,
|
| 1457 |
"step": 5850
|
| 1458 |
},
|
| 1459 |
{
|
| 1460 |
+
"epoch": 18.04,
|
| 1461 |
+
"learning_rate": 2.6100715465052286e-06,
|
| 1462 |
+
"loss": 0.0001,
|
| 1463 |
"step": 5875
|
| 1464 |
},
|
| 1465 |
{
|
| 1466 |
+
"epoch": 19.0,
|
| 1467 |
+
"learning_rate": 2.5756741882223446e-06,
|
| 1468 |
+
"loss": 0.0001,
|
| 1469 |
"step": 5900
|
| 1470 |
},
|
| 1471 |
{
|
| 1472 |
+
"epoch": 19.0,
|
| 1473 |
+
"learning_rate": 2.5412768299394606e-06,
|
| 1474 |
+
"loss": 0.0005,
|
| 1475 |
"step": 5925
|
| 1476 |
},
|
| 1477 |
{
|
| 1478 |
+
"epoch": 19.01,
|
| 1479 |
+
"learning_rate": 2.506879471656577e-06,
|
| 1480 |
+
"loss": 0.0002,
|
| 1481 |
"step": 5950
|
| 1482 |
},
|
| 1483 |
{
|
| 1484 |
+
"epoch": 19.01,
|
| 1485 |
+
"learning_rate": 2.472482113373693e-06,
|
| 1486 |
+
"loss": 0.0001,
|
| 1487 |
"step": 5975
|
| 1488 |
},
|
| 1489 |
{
|
| 1490 |
+
"epoch": 19.01,
|
| 1491 |
+
"learning_rate": 2.4380847550908094e-06,
|
| 1492 |
+
"loss": 0.0001,
|
| 1493 |
"step": 6000
|
| 1494 |
},
|
| 1495 |
{
|
| 1496 |
+
"epoch": 19.01,
|
| 1497 |
+
"eval_loss": 0.5891119837760925,
|
| 1498 |
+
"eval_runtime": 2229.5631,
|
| 1499 |
+
"eval_samples_per_second": 0.578,
|
| 1500 |
"eval_steps_per_second": 0.289,
|
| 1501 |
+
"eval_wer": 27.52938250108824,
|
| 1502 |
"step": 6000
|
| 1503 |
},
|
| 1504 |
{
|
| 1505 |
+
"epoch": 19.02,
|
| 1506 |
+
"learning_rate": 2.4036873968079254e-06,
|
| 1507 |
+
"loss": 0.0001,
|
| 1508 |
+
"step": 6025
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 19.02,
|
| 1512 |
+
"learning_rate": 2.3692900385250414e-06,
|
| 1513 |
+
"loss": 0.0001,
|
| 1514 |
+
"step": 6050
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 19.02,
|
| 1518 |
+
"learning_rate": 2.3348926802421574e-06,
|
| 1519 |
+
"loss": 0.0001,
|
| 1520 |
+
"step": 6075
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 19.03,
|
| 1524 |
+
"learning_rate": 2.300495321959274e-06,
|
| 1525 |
+
"loss": 0.0003,
|
| 1526 |
+
"step": 6100
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 19.03,
|
| 1530 |
+
"learning_rate": 2.26609796367639e-06,
|
| 1531 |
+
"loss": 0.0001,
|
| 1532 |
+
"step": 6125
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 19.03,
|
| 1536 |
+
"learning_rate": 2.231700605393506e-06,
|
| 1537 |
+
"loss": 0.0001,
|
| 1538 |
+
"step": 6150
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 19.04,
|
| 1542 |
+
"learning_rate": 2.1973032471106222e-06,
|
| 1543 |
+
"loss": 0.0001,
|
| 1544 |
+
"step": 6175
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 19.04,
|
| 1548 |
+
"learning_rate": 2.1629058888277382e-06,
|
| 1549 |
+
"loss": 0.0001,
|
| 1550 |
+
"step": 6200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 20.0,
|
| 1554 |
+
"learning_rate": 2.1285085305448542e-06,
|
| 1555 |
+
"loss": 0.0001,
|
| 1556 |
+
"step": 6225
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 20.01,
|
| 1560 |
+
"learning_rate": 2.0941111722619706e-06,
|
| 1561 |
+
"loss": 0.0004,
|
| 1562 |
+
"step": 6250
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 20.01,
|
| 1566 |
+
"learning_rate": 2.0597138139790866e-06,
|
| 1567 |
+
"loss": 0.0001,
|
| 1568 |
+
"step": 6275
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 20.01,
|
| 1572 |
+
"learning_rate": 2.0253164556962026e-06,
|
| 1573 |
+
"loss": 0.0001,
|
| 1574 |
+
"step": 6300
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 20.02,
|
| 1578 |
+
"learning_rate": 1.990919097413319e-06,
|
| 1579 |
+
"loss": 0.0001,
|
| 1580 |
+
"step": 6325
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 20.02,
|
| 1584 |
+
"learning_rate": 1.956521739130435e-06,
|
| 1585 |
+
"loss": 0.0001,
|
| 1586 |
+
"step": 6350
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 20.02,
|
| 1590 |
+
"learning_rate": 1.922124380847551e-06,
|
| 1591 |
+
"loss": 0.0001,
|
| 1592 |
+
"step": 6375
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 20.03,
|
| 1596 |
+
"learning_rate": 1.887727022564667e-06,
|
| 1597 |
+
"loss": 0.0001,
|
| 1598 |
+
"step": 6400
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 20.03,
|
| 1602 |
+
"learning_rate": 1.8533296642817832e-06,
|
| 1603 |
+
"loss": 0.0001,
|
| 1604 |
+
"step": 6425
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 20.03,
|
| 1608 |
+
"learning_rate": 1.8189323059988994e-06,
|
| 1609 |
+
"loss": 0.0001,
|
| 1610 |
+
"step": 6450
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 20.04,
|
| 1614 |
+
"learning_rate": 1.7845349477160156e-06,
|
| 1615 |
+
"loss": 0.0001,
|
| 1616 |
+
"step": 6475
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 20.04,
|
| 1620 |
+
"learning_rate": 1.7501375894331316e-06,
|
| 1621 |
+
"loss": 0.0003,
|
| 1622 |
+
"step": 6500
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 21.0,
|
| 1626 |
+
"learning_rate": 1.7157402311502476e-06,
|
| 1627 |
+
"loss": 0.0001,
|
| 1628 |
+
"step": 6525
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 21.01,
|
| 1632 |
+
"learning_rate": 1.6813428728673638e-06,
|
| 1633 |
+
"loss": 0.0001,
|
| 1634 |
+
"step": 6550
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 21.01,
|
| 1638 |
+
"learning_rate": 1.64694551458448e-06,
|
| 1639 |
+
"loss": 0.0001,
|
| 1640 |
+
"step": 6575
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 21.01,
|
| 1644 |
+
"learning_rate": 1.6125481563015962e-06,
|
| 1645 |
+
"loss": 0.0001,
|
| 1646 |
+
"step": 6600
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 21.01,
|
| 1650 |
+
"learning_rate": 1.5781507980187122e-06,
|
| 1651 |
+
"loss": 0.0001,
|
| 1652 |
+
"step": 6625
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 21.02,
|
| 1656 |
+
"learning_rate": 1.5437534397358285e-06,
|
| 1657 |
+
"loss": 0.0001,
|
| 1658 |
+
"step": 6650
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 21.02,
|
| 1662 |
+
"learning_rate": 1.5093560814529444e-06,
|
| 1663 |
+
"loss": 0.0001,
|
| 1664 |
+
"step": 6675
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 21.02,
|
| 1668 |
+
"learning_rate": 1.4749587231700607e-06,
|
| 1669 |
+
"loss": 0.0005,
|
| 1670 |
+
"step": 6700
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 21.03,
|
| 1674 |
+
"learning_rate": 1.4405613648871769e-06,
|
| 1675 |
+
"loss": 0.0001,
|
| 1676 |
+
"step": 6725
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 21.03,
|
| 1680 |
+
"learning_rate": 1.4061640066042929e-06,
|
| 1681 |
+
"loss": 0.0003,
|
| 1682 |
+
"step": 6750
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 21.03,
|
| 1686 |
+
"learning_rate": 1.371766648321409e-06,
|
| 1687 |
+
"loss": 0.0001,
|
| 1688 |
+
"step": 6775
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 21.04,
|
| 1692 |
+
"learning_rate": 1.3373692900385253e-06,
|
| 1693 |
+
"loss": 0.0001,
|
| 1694 |
+
"step": 6800
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 22.0,
|
| 1698 |
+
"learning_rate": 1.3029719317556413e-06,
|
| 1699 |
+
"loss": 0.0001,
|
| 1700 |
+
"step": 6825
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 22.0,
|
| 1704 |
+
"learning_rate": 1.2685745734727575e-06,
|
| 1705 |
+
"loss": 0.0001,
|
| 1706 |
+
"step": 6850
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 22.01,
|
| 1710 |
+
"learning_rate": 1.2341772151898737e-06,
|
| 1711 |
+
"loss": 0.0001,
|
| 1712 |
+
"step": 6875
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 22.01,
|
| 1716 |
+
"learning_rate": 1.1997798569069897e-06,
|
| 1717 |
+
"loss": 0.0001,
|
| 1718 |
+
"step": 6900
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 22.01,
|
| 1722 |
+
"learning_rate": 1.1653824986241059e-06,
|
| 1723 |
+
"loss": 0.0001,
|
| 1724 |
+
"step": 6925
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 22.02,
|
| 1728 |
+
"learning_rate": 1.1309851403412219e-06,
|
| 1729 |
+
"loss": 0.0002,
|
| 1730 |
+
"step": 6950
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 22.02,
|
| 1734 |
+
"learning_rate": 1.096587782058338e-06,
|
| 1735 |
+
"loss": 0.0001,
|
| 1736 |
+
"step": 6975
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 22.02,
|
| 1740 |
+
"learning_rate": 1.0621904237754543e-06,
|
| 1741 |
+
"loss": 0.0002,
|
| 1742 |
+
"step": 7000
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 22.02,
|
| 1746 |
+
"eval_loss": 0.5983160138130188,
|
| 1747 |
+
"eval_runtime": 2191.4681,
|
| 1748 |
+
"eval_samples_per_second": 0.588,
|
| 1749 |
+
"eval_steps_per_second": 0.294,
|
| 1750 |
+
"eval_wer": 27.49828990734407,
|
| 1751 |
+
"step": 7000
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 22.03,
|
| 1755 |
+
"learning_rate": 1.0277930654925703e-06,
|
| 1756 |
+
"loss": 0.0001,
|
| 1757 |
+
"step": 7025
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 22.03,
|
| 1761 |
+
"learning_rate": 9.933957072096865e-07,
|
| 1762 |
+
"loss": 0.0001,
|
| 1763 |
+
"step": 7050
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 22.03,
|
| 1767 |
+
"learning_rate": 9.589983489268025e-07,
|
| 1768 |
+
"loss": 0.0001,
|
| 1769 |
+
"step": 7075
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 22.04,
|
| 1773 |
+
"learning_rate": 9.246009906439187e-07,
|
| 1774 |
+
"loss": 0.0001,
|
| 1775 |
+
"step": 7100
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 22.04,
|
| 1779 |
+
"learning_rate": 8.902036323610347e-07,
|
| 1780 |
+
"loss": 0.0001,
|
| 1781 |
+
"step": 7125
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 23.0,
|
| 1785 |
+
"learning_rate": 8.558062740781509e-07,
|
| 1786 |
+
"loss": 0.0001,
|
| 1787 |
+
"step": 7150
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 23.01,
|
| 1791 |
+
"learning_rate": 8.214089157952671e-07,
|
| 1792 |
+
"loss": 0.0001,
|
| 1793 |
+
"step": 7175
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 23.01,
|
| 1797 |
+
"learning_rate": 7.870115575123831e-07,
|
| 1798 |
+
"loss": 0.0001,
|
| 1799 |
+
"step": 7200
|
| 1800 |
+
},
|
| 1801 |
+
{
|
| 1802 |
+
"epoch": 23.01,
|
| 1803 |
+
"learning_rate": 7.526141992294993e-07,
|
| 1804 |
+
"loss": 0.0001,
|
| 1805 |
+
"step": 7225
|
| 1806 |
+
},
|
| 1807 |
+
{
|
| 1808 |
+
"epoch": 23.02,
|
| 1809 |
+
"learning_rate": 7.182168409466153e-07,
|
| 1810 |
+
"loss": 0.0001,
|
| 1811 |
+
"step": 7250
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"epoch": 23.02,
|
| 1815 |
+
"learning_rate": 6.838194826637315e-07,
|
| 1816 |
+
"loss": 0.0001,
|
| 1817 |
+
"step": 7275
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 23.02,
|
| 1821 |
+
"learning_rate": 6.494221243808475e-07,
|
| 1822 |
+
"loss": 0.0001,
|
| 1823 |
+
"step": 7300
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 23.03,
|
| 1827 |
+
"learning_rate": 6.150247660979637e-07,
|
| 1828 |
+
"loss": 0.0001,
|
| 1829 |
+
"step": 7325
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 23.03,
|
| 1833 |
+
"learning_rate": 5.806274078150798e-07,
|
| 1834 |
+
"loss": 0.0001,
|
| 1835 |
+
"step": 7350
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 23.03,
|
| 1839 |
+
"learning_rate": 5.462300495321959e-07,
|
| 1840 |
+
"loss": 0.0001,
|
| 1841 |
+
"step": 7375
|
| 1842 |
+
},
|
| 1843 |
+
{
|
| 1844 |
+
"epoch": 23.03,
|
| 1845 |
+
"learning_rate": 5.118326912493121e-07,
|
| 1846 |
+
"loss": 0.0001,
|
| 1847 |
+
"step": 7400
|
| 1848 |
+
},
|
| 1849 |
+
{
|
| 1850 |
+
"epoch": 23.04,
|
| 1851 |
+
"learning_rate": 4.774353329664282e-07,
|
| 1852 |
+
"loss": 0.0001,
|
| 1853 |
+
"step": 7425
|
| 1854 |
+
},
|
| 1855 |
+
{
|
| 1856 |
+
"epoch": 24.0,
|
| 1857 |
+
"learning_rate": 4.4303797468354435e-07,
|
| 1858 |
+
"loss": 0.0001,
|
| 1859 |
+
"step": 7450
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 24.0,
|
| 1863 |
+
"learning_rate": 4.0864061640066045e-07,
|
| 1864 |
+
"loss": 0.0001,
|
| 1865 |
+
"step": 7475
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 24.01,
|
| 1869 |
+
"learning_rate": 3.7424325811777655e-07,
|
| 1870 |
+
"loss": 0.0001,
|
| 1871 |
+
"step": 7500
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 24.01,
|
| 1875 |
+
"learning_rate": 3.3984589983489265e-07,
|
| 1876 |
+
"loss": 0.0001,
|
| 1877 |
+
"step": 7525
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 24.01,
|
| 1881 |
+
"learning_rate": 3.054485415520088e-07,
|
| 1882 |
+
"loss": 0.0001,
|
| 1883 |
+
"step": 7550
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 24.02,
|
| 1887 |
+
"learning_rate": 2.7105118326912496e-07,
|
| 1888 |
+
"loss": 0.0001,
|
| 1889 |
+
"step": 7575
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 24.02,
|
| 1893 |
+
"learning_rate": 2.3665382498624108e-07,
|
| 1894 |
+
"loss": 0.0001,
|
| 1895 |
+
"step": 7600
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 24.02,
|
| 1899 |
+
"learning_rate": 2.0225646670335718e-07,
|
| 1900 |
+
"loss": 0.0001,
|
| 1901 |
+
"step": 7625
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 24.03,
|
| 1905 |
+
"learning_rate": 1.6785910842047334e-07,
|
| 1906 |
+
"loss": 0.0001,
|
| 1907 |
+
"step": 7650
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 24.03,
|
| 1911 |
+
"learning_rate": 1.3346175013758944e-07,
|
| 1912 |
+
"loss": 0.0001,
|
| 1913 |
+
"step": 7675
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 24.03,
|
| 1917 |
+
"learning_rate": 9.906439185470557e-08,
|
| 1918 |
+
"loss": 0.0001,
|
| 1919 |
+
"step": 7700
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 24.04,
|
| 1923 |
+
"learning_rate": 6.466703357182169e-08,
|
| 1924 |
+
"loss": 0.0001,
|
| 1925 |
+
"step": 7725
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 24.04,
|
| 1929 |
+
"learning_rate": 3.026967528893781e-08,
|
| 1930 |
+
"loss": 0.0001,
|
| 1931 |
+
"step": 7750
|
| 1932 |
+
},
|
| 1933 |
+
{
|
| 1934 |
+
"epoch": 25.0,
|
| 1935 |
+
"step": 7768,
|
| 1936 |
+
"total_flos": 1.2664482317815356e+20,
|
| 1937 |
+
"train_loss": 0.0937204750527475,
|
| 1938 |
+
"train_runtime": 49250.3081,
|
| 1939 |
+
"train_samples_per_second": 2.524,
|
| 1940 |
+
"train_steps_per_second": 0.158
|
| 1941 |
}
|
| 1942 |
],
|
| 1943 |
+
"max_steps": 7768,
|
| 1944 |
"num_train_epochs": 9223372036854775807,
|
| 1945 |
+
"total_flos": 1.2664482317815356e+20,
|
| 1946 |
"trial_name": null,
|
| 1947 |
"trial_params": null
|
| 1948 |
}
|