Training in progress, step 4000
Browse files- checkpoint-2000/optimizer.pt +1 -1
- checkpoint-2000/pytorch_model.bin +1 -1
- checkpoint-2000/rng_state.pth +1 -1
- checkpoint-2000/scaler.pt +1 -1
- checkpoint-2000/scheduler.pt +1 -1
- checkpoint-2000/trainer_state.json +212 -212
- checkpoint-2000/training_args.bin +1 -1
- checkpoint-3000/optimizer.pt +1 -1
- checkpoint-3000/pytorch_model.bin +1 -1
- checkpoint-3000/rng_state.pth +2 -2
- checkpoint-3000/scaler.pt +1 -1
- checkpoint-3000/scheduler.pt +1 -1
- checkpoint-3000/trainer_state.json +338 -338
- checkpoint-3000/training_args.bin +1 -1
- checkpoint-4000/optimizer.pt +1 -1
- checkpoint-4000/pytorch_model.bin +1 -1
- checkpoint-4000/rng_state.pth +1 -1
- checkpoint-4000/scaler.pt +1 -1
- checkpoint-4000/scheduler.pt +1 -1
- checkpoint-4000/trainer_state.json +462 -462
- checkpoint-4000/training_args.bin +1 -1
- nohup.out +0 -0
- pytorch_model.bin +1 -1
- runs/Jul23_08-44-43_tknadmin-System-Product-Name/events.out.tfevents.1690094707.tknadmin-System-Product-Name.1702135.0 +2 -2
checkpoint-2000/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 6111428695
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a159b02fa01e4405076f9d759e53e7edd6d412b19b53a0743c660e28339c1008
|
| 3 |
size 6111428695
|
checkpoint-2000/pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3055754841
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fb0822dc45eb083f947413e68d9023acdfbe2dc7d6aa96a85b5e714d22ce5be
|
| 3 |
size 3055754841
|
checkpoint-2000/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a1cbefc4297bef083a245f24c738abc29e75fca1b725711059d7702eeb3e5557
|
| 3 |
size 14575
|
checkpoint-2000/scaler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 557
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be1e74deedb43759bd15ec14e78ac0bbfcb7fd82b5ef5c8669deb1c4ee4fd1f2
|
| 3 |
size 557
|
checkpoint-2000/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1f112c28e35ae8caf36bbfc9dd4dacc3130f175782da609fb5266a28a082e3cf
|
| 3 |
size 627
|
checkpoint-2000/trainer_state.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
"best_model_checkpoint": "./checkpoint-2000",
|
| 4 |
-
"epoch": 0.
|
| 5 |
"global_step": 2000,
|
| 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.
|
| 12 |
"learning_rate": 4.6000000000000004e-07,
|
| 13 |
"loss": 1.4182,
|
| 14 |
"step": 25
|
|
@@ -26,487 +26,487 @@
|
|
| 26 |
"step": 75
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch": 0.
|
| 30 |
"learning_rate": 1.94e-06,
|
| 31 |
"loss": 0.7765,
|
| 32 |
"step": 100
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch": 0.
|
| 36 |
"learning_rate": 2.4400000000000004e-06,
|
| 37 |
"loss": 0.7103,
|
| 38 |
"step": 125
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch": 0.
|
| 42 |
"learning_rate": 2.9400000000000002e-06,
|
| 43 |
"loss": 0.6597,
|
| 44 |
"step": 150
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
"learning_rate": 3.44e-06,
|
| 49 |
"loss": 0.6657,
|
| 50 |
"step": 175
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"epoch": 0.
|
| 54 |
"learning_rate": 3.94e-06,
|
| 55 |
"loss": 0.5853,
|
| 56 |
"step": 200
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch": 0.
|
| 60 |
"learning_rate": 4.440000000000001e-06,
|
| 61 |
"loss": 0.5273,
|
| 62 |
"step": 225
|
| 63 |
},
|
| 64 |
{
|
| 65 |
-
"epoch": 0.
|
| 66 |
"learning_rate": 4.94e-06,
|
| 67 |
"loss": 0.5979,
|
| 68 |
"step": 250
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch": 0.
|
| 72 |
"learning_rate": 5.4400000000000004e-06,
|
| 73 |
"loss": 0.5861,
|
| 74 |
"step": 275
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch": 0.
|
| 78 |
"learning_rate": 5.94e-06,
|
| 79 |
"loss": 0.5085,
|
| 80 |
"step": 300
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch": 0.
|
| 84 |
"learning_rate": 6.440000000000001e-06,
|
| 85 |
"loss": 0.4827,
|
| 86 |
"step": 325
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
"learning_rate": 6.9400000000000005e-06,
|
| 91 |
"loss": 0.4909,
|
| 92 |
"step": 350
|
| 93 |
},
|
| 94 |
{
|
| 95 |
-
"epoch": 0.
|
| 96 |
"learning_rate": 7.440000000000001e-06,
|
| 97 |
"loss": 0.4651,
|
| 98 |
"step": 375
|
| 99 |
},
|
| 100 |
{
|
| 101 |
-
"epoch": 0.
|
| 102 |
"learning_rate": 7.94e-06,
|
| 103 |
"loss": 0.494,
|
| 104 |
"step": 400
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch": 0.
|
| 108 |
"learning_rate": 8.44e-06,
|
| 109 |
"loss": 0.4188,
|
| 110 |
"step": 425
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"epoch": 0.
|
| 114 |
"learning_rate": 8.94e-06,
|
| 115 |
"loss": 0.3849,
|
| 116 |
"step": 450
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch": 0.
|
| 120 |
"learning_rate": 9.440000000000001e-06,
|
| 121 |
"loss": 0.4577,
|
| 122 |
"step": 475
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch": 0.
|
| 126 |
"learning_rate": 9.940000000000001e-06,
|
| 127 |
"loss": 0.4415,
|
| 128 |
"step": 500
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"learning_rate": 9.
|
| 133 |
-
"loss": 0.
|
| 134 |
"step": 525
|
| 135 |
},
|
| 136 |
{
|
| 137 |
-
"epoch": 0.
|
| 138 |
-
"learning_rate": 9.
|
| 139 |
-
"loss": 0.
|
| 140 |
"step": 550
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"epoch": 0.
|
| 144 |
-
"learning_rate": 9.
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 575
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch": 0.
|
| 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": 0.
|
| 168 |
-
"learning_rate": 9.
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 675
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"learning_rate": 9.
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 700
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch": 0.
|
| 180 |
-
"learning_rate": 9.
|
| 181 |
-
"loss": 0.
|
| 182 |
"step": 725
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch": 0.
|
| 186 |
-
"learning_rate": 9.
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 750
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch": 0.
|
| 192 |
-
"learning_rate": 9.
|
| 193 |
-
"loss": 0.
|
| 194 |
"step": 775
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch": 0.
|
| 198 |
-
"learning_rate": 9.
|
| 199 |
-
"loss": 0.
|
| 200 |
"step": 800
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch": 0.
|
| 204 |
-
"learning_rate": 9.
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 825
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch": 0.
|
| 210 |
-
"learning_rate": 9.
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 850
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"learning_rate": 9.
|
| 217 |
-
"loss": 0.
|
| 218 |
"step": 875
|
| 219 |
},
|
| 220 |
{
|
| 221 |
-
"epoch": 0.
|
| 222 |
-
"learning_rate": 9.
|
| 223 |
-
"loss": 0.
|
| 224 |
"step": 900
|
| 225 |
},
|
| 226 |
{
|
| 227 |
-
"epoch": 0.
|
| 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": 0.
|
| 240 |
-
"learning_rate":
|
| 241 |
-
"loss": 0.
|
| 242 |
"step": 975
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch": 0.
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 1000
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch": 0.
|
| 252 |
-
"eval_loss": 0.
|
| 253 |
-
"eval_runtime":
|
| 254 |
-
"eval_samples_per_second": 3.
|
| 255 |
-
"eval_steps_per_second": 0.
|
| 256 |
-
"eval_wer": 28.
|
| 257 |
"step": 1000
|
| 258 |
},
|
| 259 |
{
|
| 260 |
-
"epoch": 0.
|
| 261 |
-
"learning_rate":
|
| 262 |
-
"loss": 0.
|
| 263 |
"step": 1025
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch": 0.
|
| 267 |
-
"learning_rate":
|
| 268 |
"loss": 0.3674,
|
| 269 |
"step": 1050
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch": 0.
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 1075
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 0.
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 1100
|
| 282 |
},
|
| 283 |
{
|
| 284 |
-
"epoch": 0.
|
| 285 |
-
"learning_rate":
|
| 286 |
-
"loss": 0.
|
| 287 |
"step": 1125
|
| 288 |
},
|
| 289 |
{
|
| 290 |
-
"epoch": 0.
|
| 291 |
-
"learning_rate":
|
| 292 |
-
"loss": 0.
|
| 293 |
"step": 1150
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"epoch": 0.
|
| 297 |
-
"learning_rate": 8.
|
| 298 |
-
"loss": 0.
|
| 299 |
"step": 1175
|
| 300 |
},
|
| 301 |
{
|
| 302 |
-
"epoch": 0.
|
| 303 |
-
"learning_rate": 8.
|
| 304 |
-
"loss": 0.
|
| 305 |
"step": 1200
|
| 306 |
},
|
| 307 |
{
|
| 308 |
-
"epoch": 0.
|
| 309 |
-
"learning_rate": 8.
|
| 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": 0.
|
| 327 |
-
"learning_rate": 8.
|
| 328 |
-
"loss": 0.
|
| 329 |
"step": 1300
|
| 330 |
},
|
| 331 |
{
|
| 332 |
-
"epoch": 0.
|
| 333 |
-
"learning_rate": 8.
|
| 334 |
-
"loss": 0.
|
| 335 |
"step": 1325
|
| 336 |
},
|
| 337 |
{
|
| 338 |
-
"epoch": 0.
|
| 339 |
-
"learning_rate": 8.
|
| 340 |
-
"loss": 0.
|
| 341 |
"step": 1350
|
| 342 |
},
|
| 343 |
{
|
| 344 |
-
"epoch": 0.
|
| 345 |
-
"learning_rate": 8.
|
| 346 |
-
"loss": 0.
|
| 347 |
"step": 1375
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch": 0.
|
| 351 |
-
"learning_rate": 8.
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 1400
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch": 0.
|
| 357 |
-
"learning_rate":
|
| 358 |
-
"loss": 0.
|
| 359 |
"step": 1425
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 0.
|
| 363 |
-
"learning_rate":
|
| 364 |
-
"loss": 0.
|
| 365 |
"step": 1450
|
| 366 |
},
|
| 367 |
{
|
| 368 |
-
"epoch": 0.
|
| 369 |
-
"learning_rate":
|
| 370 |
-
"loss": 0.
|
| 371 |
"step": 1475
|
| 372 |
},
|
| 373 |
{
|
| 374 |
-
"epoch": 0.
|
| 375 |
-
"learning_rate":
|
| 376 |
-
"loss": 0.
|
| 377 |
"step": 1500
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch": 0.
|
| 381 |
-
"learning_rate":
|
| 382 |
-
"loss": 0.
|
| 383 |
"step": 1525
|
| 384 |
},
|
| 385 |
{
|
| 386 |
-
"epoch": 0.
|
| 387 |
-
"learning_rate":
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 1550
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch": 0.
|
| 393 |
-
"learning_rate":
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 1575
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch": 0.
|
| 399 |
-
"learning_rate":
|
| 400 |
-
"loss": 0.
|
| 401 |
"step": 1600
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 0.
|
| 405 |
-
"learning_rate":
|
| 406 |
-
"loss": 0.
|
| 407 |
"step": 1625
|
| 408 |
},
|
| 409 |
{
|
| 410 |
-
"epoch": 0.
|
| 411 |
-
"learning_rate":
|
| 412 |
-
"loss": 0.
|
| 413 |
"step": 1650
|
| 414 |
},
|
| 415 |
{
|
| 416 |
-
"epoch": 0.
|
| 417 |
-
"learning_rate":
|
| 418 |
-
"loss": 0.
|
| 419 |
"step": 1675
|
| 420 |
},
|
| 421 |
{
|
| 422 |
-
"epoch": 0.
|
| 423 |
-
"learning_rate":
|
| 424 |
-
"loss": 0.
|
| 425 |
"step": 1700
|
| 426 |
},
|
| 427 |
{
|
| 428 |
-
"epoch": 0.
|
| 429 |
-
"learning_rate":
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 1725
|
| 432 |
},
|
| 433 |
{
|
| 434 |
-
"epoch": 0.
|
| 435 |
-
"learning_rate":
|
| 436 |
-
"loss": 0.
|
| 437 |
"step": 1750
|
| 438 |
},
|
| 439 |
{
|
| 440 |
-
"epoch": 0.
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss": 0.
|
| 443 |
"step": 1775
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 0.
|
| 447 |
-
"learning_rate":
|
| 448 |
-
"loss": 0.
|
| 449 |
"step": 1800
|
| 450 |
},
|
| 451 |
{
|
| 452 |
-
"epoch": 0.
|
| 453 |
-
"learning_rate": 7.
|
| 454 |
-
"loss": 0.
|
| 455 |
"step": 1825
|
| 456 |
},
|
| 457 |
{
|
| 458 |
-
"epoch": 0.
|
| 459 |
-
"learning_rate": 7.
|
| 460 |
-
"loss": 0.
|
| 461 |
"step": 1850
|
| 462 |
},
|
| 463 |
{
|
| 464 |
-
"epoch": 0.
|
| 465 |
-
"learning_rate":
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 1875
|
| 468 |
},
|
| 469 |
{
|
| 470 |
-
"epoch": 0.
|
| 471 |
-
"learning_rate":
|
| 472 |
-
"loss": 0.
|
| 473 |
"step": 1900
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"epoch": 0.
|
| 477 |
-
"learning_rate":
|
| 478 |
-
"loss": 0.
|
| 479 |
"step": 1925
|
| 480 |
},
|
| 481 |
{
|
| 482 |
-
"epoch": 0.
|
| 483 |
-
"learning_rate":
|
| 484 |
-
"loss": 0.
|
| 485 |
"step": 1950
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 0.
|
| 489 |
-
"learning_rate":
|
| 490 |
-
"loss": 0.
|
| 491 |
"step": 1975
|
| 492 |
},
|
| 493 |
{
|
| 494 |
-
"epoch": 0.
|
| 495 |
-
"learning_rate":
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 2000
|
| 498 |
},
|
| 499 |
{
|
| 500 |
-
"epoch": 0.
|
| 501 |
-
"eval_loss": 0.
|
| 502 |
-
"eval_runtime":
|
| 503 |
-
"eval_samples_per_second": 3.
|
| 504 |
-
"eval_steps_per_second": 0.
|
| 505 |
-
"eval_wer":
|
| 506 |
"step": 2000
|
| 507 |
}
|
| 508 |
],
|
| 509 |
-
"max_steps":
|
| 510 |
"num_train_epochs": 9223372036854775807,
|
| 511 |
"total_flos": 8.16483926016e+18,
|
| 512 |
"trial_name": null,
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 22.052330997610273,
|
| 3 |
"best_model_checkpoint": "./checkpoint-2000",
|
| 4 |
+
"epoch": 0.2857142857142857,
|
| 5 |
"global_step": 2000,
|
| 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": 4.6000000000000004e-07,
|
| 13 |
"loss": 1.4182,
|
| 14 |
"step": 25
|
|
|
|
| 26 |
"step": 75
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.01,
|
| 30 |
"learning_rate": 1.94e-06,
|
| 31 |
"loss": 0.7765,
|
| 32 |
"step": 100
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 0.02,
|
| 36 |
"learning_rate": 2.4400000000000004e-06,
|
| 37 |
"loss": 0.7103,
|
| 38 |
"step": 125
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.02,
|
| 42 |
"learning_rate": 2.9400000000000002e-06,
|
| 43 |
"loss": 0.6597,
|
| 44 |
"step": 150
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.03,
|
| 48 |
"learning_rate": 3.44e-06,
|
| 49 |
"loss": 0.6657,
|
| 50 |
"step": 175
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 0.03,
|
| 54 |
"learning_rate": 3.94e-06,
|
| 55 |
"loss": 0.5853,
|
| 56 |
"step": 200
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 0.03,
|
| 60 |
"learning_rate": 4.440000000000001e-06,
|
| 61 |
"loss": 0.5273,
|
| 62 |
"step": 225
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 0.04,
|
| 66 |
"learning_rate": 4.94e-06,
|
| 67 |
"loss": 0.5979,
|
| 68 |
"step": 250
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 0.04,
|
| 72 |
"learning_rate": 5.4400000000000004e-06,
|
| 73 |
"loss": 0.5861,
|
| 74 |
"step": 275
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 0.04,
|
| 78 |
"learning_rate": 5.94e-06,
|
| 79 |
"loss": 0.5085,
|
| 80 |
"step": 300
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 0.05,
|
| 84 |
"learning_rate": 6.440000000000001e-06,
|
| 85 |
"loss": 0.4827,
|
| 86 |
"step": 325
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.05,
|
| 90 |
"learning_rate": 6.9400000000000005e-06,
|
| 91 |
"loss": 0.4909,
|
| 92 |
"step": 350
|
| 93 |
},
|
| 94 |
{
|
| 95 |
+
"epoch": 0.05,
|
| 96 |
"learning_rate": 7.440000000000001e-06,
|
| 97 |
"loss": 0.4651,
|
| 98 |
"step": 375
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 0.06,
|
| 102 |
"learning_rate": 7.94e-06,
|
| 103 |
"loss": 0.494,
|
| 104 |
"step": 400
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 0.06,
|
| 108 |
"learning_rate": 8.44e-06,
|
| 109 |
"loss": 0.4188,
|
| 110 |
"step": 425
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"epoch": 0.06,
|
| 114 |
"learning_rate": 8.94e-06,
|
| 115 |
"loss": 0.3849,
|
| 116 |
"step": 450
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 0.07,
|
| 120 |
"learning_rate": 9.440000000000001e-06,
|
| 121 |
"loss": 0.4577,
|
| 122 |
"step": 475
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 0.07,
|
| 126 |
"learning_rate": 9.940000000000001e-06,
|
| 127 |
"loss": 0.4415,
|
| 128 |
"step": 500
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.07,
|
| 132 |
+
"learning_rate": 9.966153846153847e-06,
|
| 133 |
+
"loss": 0.4614,
|
| 134 |
"step": 525
|
| 135 |
},
|
| 136 |
{
|
| 137 |
+
"epoch": 0.08,
|
| 138 |
+
"learning_rate": 9.927692307692309e-06,
|
| 139 |
+
"loss": 0.4283,
|
| 140 |
"step": 550
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"epoch": 0.08,
|
| 144 |
+
"learning_rate": 9.88923076923077e-06,
|
| 145 |
+
"loss": 0.4486,
|
| 146 |
"step": 575
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 0.09,
|
| 150 |
+
"learning_rate": 9.850769230769231e-06,
|
| 151 |
+
"loss": 0.4434,
|
| 152 |
"step": 600
|
| 153 |
},
|
| 154 |
{
|
| 155 |
+
"epoch": 0.09,
|
| 156 |
+
"learning_rate": 9.812307692307694e-06,
|
| 157 |
+
"loss": 0.4245,
|
| 158 |
"step": 625
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 0.09,
|
| 162 |
+
"learning_rate": 9.773846153846154e-06,
|
| 163 |
+
"loss": 0.4254,
|
| 164 |
"step": 650
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 0.1,
|
| 168 |
+
"learning_rate": 9.735384615384616e-06,
|
| 169 |
+
"loss": 0.4143,
|
| 170 |
"step": 675
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.1,
|
| 174 |
+
"learning_rate": 9.696923076923078e-06,
|
| 175 |
+
"loss": 0.4115,
|
| 176 |
"step": 700
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 0.1,
|
| 180 |
+
"learning_rate": 9.658461538461539e-06,
|
| 181 |
+
"loss": 0.3656,
|
| 182 |
"step": 725
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"epoch": 0.11,
|
| 186 |
+
"learning_rate": 9.620000000000001e-06,
|
| 187 |
+
"loss": 0.4004,
|
| 188 |
"step": 750
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 0.11,
|
| 192 |
+
"learning_rate": 9.581538461538462e-06,
|
| 193 |
+
"loss": 0.388,
|
| 194 |
"step": 775
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 0.11,
|
| 198 |
+
"learning_rate": 9.543076923076924e-06,
|
| 199 |
+
"loss": 0.3665,
|
| 200 |
"step": 800
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 0.12,
|
| 204 |
+
"learning_rate": 9.504615384615386e-06,
|
| 205 |
+
"loss": 0.3777,
|
| 206 |
"step": 825
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 0.12,
|
| 210 |
+
"learning_rate": 9.466153846153846e-06,
|
| 211 |
+
"loss": 0.4338,
|
| 212 |
"step": 850
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.12,
|
| 216 |
+
"learning_rate": 9.427692307692309e-06,
|
| 217 |
+
"loss": 0.4017,
|
| 218 |
"step": 875
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 0.13,
|
| 222 |
+
"learning_rate": 9.38923076923077e-06,
|
| 223 |
+
"loss": 0.3757,
|
| 224 |
"step": 900
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 0.13,
|
| 228 |
+
"learning_rate": 9.350769230769231e-06,
|
| 229 |
+
"loss": 0.4084,
|
| 230 |
"step": 925
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 0.14,
|
| 234 |
+
"learning_rate": 9.312307692307693e-06,
|
| 235 |
+
"loss": 0.3821,
|
| 236 |
"step": 950
|
| 237 |
},
|
| 238 |
{
|
| 239 |
+
"epoch": 0.14,
|
| 240 |
+
"learning_rate": 9.273846153846154e-06,
|
| 241 |
+
"loss": 0.421,
|
| 242 |
"step": 975
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 0.14,
|
| 246 |
+
"learning_rate": 9.235384615384616e-06,
|
| 247 |
+
"loss": 0.4203,
|
| 248 |
"step": 1000
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 0.14,
|
| 252 |
+
"eval_loss": 0.4127572178840637,
|
| 253 |
+
"eval_runtime": 1802.5775,
|
| 254 |
+
"eval_samples_per_second": 3.656,
|
| 255 |
+
"eval_steps_per_second": 0.457,
|
| 256 |
+
"eval_wer": 28.265624367127064,
|
| 257 |
"step": 1000
|
| 258 |
},
|
| 259 |
{
|
| 260 |
+
"epoch": 0.15,
|
| 261 |
+
"learning_rate": 9.196923076923078e-06,
|
| 262 |
+
"loss": 0.4094,
|
| 263 |
"step": 1025
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"epoch": 0.15,
|
| 267 |
+
"learning_rate": 9.158461538461539e-06,
|
| 268 |
"loss": 0.3674,
|
| 269 |
"step": 1050
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"epoch": 0.15,
|
| 273 |
+
"learning_rate": 9.12e-06,
|
| 274 |
+
"loss": 0.3613,
|
| 275 |
"step": 1075
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.16,
|
| 279 |
+
"learning_rate": 9.081538461538463e-06,
|
| 280 |
+
"loss": 0.3904,
|
| 281 |
"step": 1100
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 0.16,
|
| 285 |
+
"learning_rate": 9.043076923076923e-06,
|
| 286 |
+
"loss": 0.3865,
|
| 287 |
"step": 1125
|
| 288 |
},
|
| 289 |
{
|
| 290 |
+
"epoch": 0.16,
|
| 291 |
+
"learning_rate": 9.004615384615386e-06,
|
| 292 |
+
"loss": 0.3677,
|
| 293 |
"step": 1150
|
| 294 |
},
|
| 295 |
{
|
| 296 |
+
"epoch": 0.17,
|
| 297 |
+
"learning_rate": 8.966153846153846e-06,
|
| 298 |
+
"loss": 0.3295,
|
| 299 |
"step": 1175
|
| 300 |
},
|
| 301 |
{
|
| 302 |
+
"epoch": 0.17,
|
| 303 |
+
"learning_rate": 8.927692307692308e-06,
|
| 304 |
+
"loss": 0.3207,
|
| 305 |
"step": 1200
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"epoch": 0.17,
|
| 309 |
+
"learning_rate": 8.88923076923077e-06,
|
| 310 |
+
"loss": 0.2817,
|
| 311 |
"step": 1225
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"epoch": 0.18,
|
| 315 |
+
"learning_rate": 8.850769230769231e-06,
|
| 316 |
+
"loss": 0.3218,
|
| 317 |
"step": 1250
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.18,
|
| 321 |
+
"learning_rate": 8.812307692307693e-06,
|
| 322 |
+
"loss": 0.3502,
|
| 323 |
"step": 1275
|
| 324 |
},
|
| 325 |
{
|
| 326 |
+
"epoch": 0.19,
|
| 327 |
+
"learning_rate": 8.773846153846155e-06,
|
| 328 |
+
"loss": 0.3432,
|
| 329 |
"step": 1300
|
| 330 |
},
|
| 331 |
{
|
| 332 |
+
"epoch": 0.19,
|
| 333 |
+
"learning_rate": 8.735384615384616e-06,
|
| 334 |
+
"loss": 0.3244,
|
| 335 |
"step": 1325
|
| 336 |
},
|
| 337 |
{
|
| 338 |
+
"epoch": 0.19,
|
| 339 |
+
"learning_rate": 8.696923076923078e-06,
|
| 340 |
+
"loss": 0.2607,
|
| 341 |
"step": 1350
|
| 342 |
},
|
| 343 |
{
|
| 344 |
+
"epoch": 0.2,
|
| 345 |
+
"learning_rate": 8.658461538461538e-06,
|
| 346 |
+
"loss": 0.3469,
|
| 347 |
"step": 1375
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"epoch": 0.2,
|
| 351 |
+
"learning_rate": 8.62e-06,
|
| 352 |
+
"loss": 0.2756,
|
| 353 |
"step": 1400
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 0.2,
|
| 357 |
+
"learning_rate": 8.581538461538463e-06,
|
| 358 |
+
"loss": 0.294,
|
| 359 |
"step": 1425
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.21,
|
| 363 |
+
"learning_rate": 8.543076923076923e-06,
|
| 364 |
+
"loss": 0.3915,
|
| 365 |
"step": 1450
|
| 366 |
},
|
| 367 |
{
|
| 368 |
+
"epoch": 0.21,
|
| 369 |
+
"learning_rate": 8.504615384615385e-06,
|
| 370 |
+
"loss": 0.3237,
|
| 371 |
"step": 1475
|
| 372 |
},
|
| 373 |
{
|
| 374 |
+
"epoch": 0.21,
|
| 375 |
+
"learning_rate": 8.466153846153847e-06,
|
| 376 |
+
"loss": 0.2723,
|
| 377 |
"step": 1500
|
| 378 |
},
|
| 379 |
{
|
| 380 |
+
"epoch": 0.22,
|
| 381 |
+
"learning_rate": 8.427692307692308e-06,
|
| 382 |
+
"loss": 0.2917,
|
| 383 |
"step": 1525
|
| 384 |
},
|
| 385 |
{
|
| 386 |
+
"epoch": 0.22,
|
| 387 |
+
"learning_rate": 8.38923076923077e-06,
|
| 388 |
+
"loss": 0.2921,
|
| 389 |
"step": 1550
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 0.23,
|
| 393 |
+
"learning_rate": 8.35076923076923e-06,
|
| 394 |
+
"loss": 0.3724,
|
| 395 |
"step": 1575
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 0.23,
|
| 399 |
+
"learning_rate": 8.312307692307693e-06,
|
| 400 |
+
"loss": 0.3364,
|
| 401 |
"step": 1600
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.23,
|
| 405 |
+
"learning_rate": 8.273846153846155e-06,
|
| 406 |
+
"loss": 0.3661,
|
| 407 |
"step": 1625
|
| 408 |
},
|
| 409 |
{
|
| 410 |
+
"epoch": 0.24,
|
| 411 |
+
"learning_rate": 8.235384615384615e-06,
|
| 412 |
+
"loss": 0.3307,
|
| 413 |
"step": 1650
|
| 414 |
},
|
| 415 |
{
|
| 416 |
+
"epoch": 0.24,
|
| 417 |
+
"learning_rate": 8.196923076923078e-06,
|
| 418 |
+
"loss": 0.3201,
|
| 419 |
"step": 1675
|
| 420 |
},
|
| 421 |
{
|
| 422 |
+
"epoch": 0.24,
|
| 423 |
+
"learning_rate": 8.15846153846154e-06,
|
| 424 |
+
"loss": 0.2973,
|
| 425 |
"step": 1700
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"epoch": 0.25,
|
| 429 |
+
"learning_rate": 8.120000000000002e-06,
|
| 430 |
+
"loss": 0.3013,
|
| 431 |
"step": 1725
|
| 432 |
},
|
| 433 |
{
|
| 434 |
+
"epoch": 0.25,
|
| 435 |
+
"learning_rate": 8.081538461538462e-06,
|
| 436 |
+
"loss": 0.3349,
|
| 437 |
"step": 1750
|
| 438 |
},
|
| 439 |
{
|
| 440 |
+
"epoch": 0.25,
|
| 441 |
+
"learning_rate": 8.043076923076923e-06,
|
| 442 |
+
"loss": 0.3244,
|
| 443 |
"step": 1775
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.26,
|
| 447 |
+
"learning_rate": 8.004615384615385e-06,
|
| 448 |
+
"loss": 0.3331,
|
| 449 |
"step": 1800
|
| 450 |
},
|
| 451 |
{
|
| 452 |
+
"epoch": 0.26,
|
| 453 |
+
"learning_rate": 7.966153846153847e-06,
|
| 454 |
+
"loss": 0.2723,
|
| 455 |
"step": 1825
|
| 456 |
},
|
| 457 |
{
|
| 458 |
+
"epoch": 0.26,
|
| 459 |
+
"learning_rate": 7.927692307692308e-06,
|
| 460 |
+
"loss": 0.316,
|
| 461 |
"step": 1850
|
| 462 |
},
|
| 463 |
{
|
| 464 |
+
"epoch": 0.27,
|
| 465 |
+
"learning_rate": 7.88923076923077e-06,
|
| 466 |
+
"loss": 0.2648,
|
| 467 |
"step": 1875
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 0.27,
|
| 471 |
+
"learning_rate": 7.850769230769232e-06,
|
| 472 |
+
"loss": 0.2965,
|
| 473 |
"step": 1900
|
| 474 |
},
|
| 475 |
{
|
| 476 |
+
"epoch": 0.28,
|
| 477 |
+
"learning_rate": 7.812307692307694e-06,
|
| 478 |
+
"loss": 0.256,
|
| 479 |
"step": 1925
|
| 480 |
},
|
| 481 |
{
|
| 482 |
+
"epoch": 0.28,
|
| 483 |
+
"learning_rate": 7.773846153846155e-06,
|
| 484 |
+
"loss": 0.2156,
|
| 485 |
"step": 1950
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.28,
|
| 489 |
+
"learning_rate": 7.735384615384615e-06,
|
| 490 |
+
"loss": 0.2418,
|
| 491 |
"step": 1975
|
| 492 |
},
|
| 493 |
{
|
| 494 |
+
"epoch": 0.29,
|
| 495 |
+
"learning_rate": 7.696923076923077e-06,
|
| 496 |
+
"loss": 0.2693,
|
| 497 |
"step": 2000
|
| 498 |
},
|
| 499 |
{
|
| 500 |
+
"epoch": 0.29,
|
| 501 |
+
"eval_loss": 0.3239877223968506,
|
| 502 |
+
"eval_runtime": 1817.7748,
|
| 503 |
+
"eval_samples_per_second": 3.626,
|
| 504 |
+
"eval_steps_per_second": 0.453,
|
| 505 |
+
"eval_wer": 22.052330997610273,
|
| 506 |
"step": 2000
|
| 507 |
}
|
| 508 |
],
|
| 509 |
+
"max_steps": 7000,
|
| 510 |
"num_train_epochs": 9223372036854775807,
|
| 511 |
"total_flos": 8.16483926016e+18,
|
| 512 |
"trial_name": null,
|
checkpoint-2000/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3643
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1ca27d9d1bef820c87cfd52d5adb15af5250e1ac4555cd1703a088ad0dad1a4
|
| 3 |
size 3643
|
checkpoint-3000/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 6111428695
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:204df11caa5a4cc0c8bc1ea18995b04587f1d457c412b5cf195a5a895f83074c
|
| 3 |
size 6111428695
|
checkpoint-3000/pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3055754841
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e8631f8c8b760ece8b67999d400c1670f308714ab9ab1de0f37457e497096b3
|
| 3 |
size 3055754841
|
checkpoint-3000/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3be0ae51fe09627683807728545588affaec3370727df148b33392ba8055e58f
|
| 3 |
+
size 14575
|
checkpoint-3000/scaler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 557
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eed3cdc5e39a2cfa8de4e0045b0e1a83c149d3d1d12677c2a9824efb5d14e59c
|
| 3 |
size 557
|
checkpoint-3000/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dd7319c75e8411b4e2ae5a8b7fcdd7bb253d178b5a5770d7926097398cb526f8
|
| 3 |
size 627
|
checkpoint-3000/trainer_state.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
"best_model_checkpoint": "./checkpoint-3000",
|
| 4 |
-
"epoch": 0.
|
| 5 |
"global_step": 3000,
|
| 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.
|
| 12 |
"learning_rate": 4.6000000000000004e-07,
|
| 13 |
"loss": 1.4182,
|
| 14 |
"step": 25
|
|
@@ -26,736 +26,736 @@
|
|
| 26 |
"step": 75
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch": 0.
|
| 30 |
"learning_rate": 1.94e-06,
|
| 31 |
"loss": 0.7765,
|
| 32 |
"step": 100
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch": 0.
|
| 36 |
"learning_rate": 2.4400000000000004e-06,
|
| 37 |
"loss": 0.7103,
|
| 38 |
"step": 125
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch": 0.
|
| 42 |
"learning_rate": 2.9400000000000002e-06,
|
| 43 |
"loss": 0.6597,
|
| 44 |
"step": 150
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
"learning_rate": 3.44e-06,
|
| 49 |
"loss": 0.6657,
|
| 50 |
"step": 175
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"epoch": 0.
|
| 54 |
"learning_rate": 3.94e-06,
|
| 55 |
"loss": 0.5853,
|
| 56 |
"step": 200
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch": 0.
|
| 60 |
"learning_rate": 4.440000000000001e-06,
|
| 61 |
"loss": 0.5273,
|
| 62 |
"step": 225
|
| 63 |
},
|
| 64 |
{
|
| 65 |
-
"epoch": 0.
|
| 66 |
"learning_rate": 4.94e-06,
|
| 67 |
"loss": 0.5979,
|
| 68 |
"step": 250
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch": 0.
|
| 72 |
"learning_rate": 5.4400000000000004e-06,
|
| 73 |
"loss": 0.5861,
|
| 74 |
"step": 275
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch": 0.
|
| 78 |
"learning_rate": 5.94e-06,
|
| 79 |
"loss": 0.5085,
|
| 80 |
"step": 300
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch": 0.
|
| 84 |
"learning_rate": 6.440000000000001e-06,
|
| 85 |
"loss": 0.4827,
|
| 86 |
"step": 325
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
"learning_rate": 6.9400000000000005e-06,
|
| 91 |
"loss": 0.4909,
|
| 92 |
"step": 350
|
| 93 |
},
|
| 94 |
{
|
| 95 |
-
"epoch": 0.
|
| 96 |
"learning_rate": 7.440000000000001e-06,
|
| 97 |
"loss": 0.4651,
|
| 98 |
"step": 375
|
| 99 |
},
|
| 100 |
{
|
| 101 |
-
"epoch": 0.
|
| 102 |
"learning_rate": 7.94e-06,
|
| 103 |
"loss": 0.494,
|
| 104 |
"step": 400
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch": 0.
|
| 108 |
"learning_rate": 8.44e-06,
|
| 109 |
"loss": 0.4188,
|
| 110 |
"step": 425
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"epoch": 0.
|
| 114 |
"learning_rate": 8.94e-06,
|
| 115 |
"loss": 0.3849,
|
| 116 |
"step": 450
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch": 0.
|
| 120 |
"learning_rate": 9.440000000000001e-06,
|
| 121 |
"loss": 0.4577,
|
| 122 |
"step": 475
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch": 0.
|
| 126 |
"learning_rate": 9.940000000000001e-06,
|
| 127 |
"loss": 0.4415,
|
| 128 |
"step": 500
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"learning_rate": 9.
|
| 133 |
-
"loss": 0.
|
| 134 |
"step": 525
|
| 135 |
},
|
| 136 |
{
|
| 137 |
-
"epoch": 0.
|
| 138 |
-
"learning_rate": 9.
|
| 139 |
-
"loss": 0.
|
| 140 |
"step": 550
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"epoch": 0.
|
| 144 |
-
"learning_rate": 9.
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 575
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch": 0.
|
| 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": 0.
|
| 168 |
-
"learning_rate": 9.
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 675
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"learning_rate": 9.
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 700
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch": 0.
|
| 180 |
-
"learning_rate": 9.
|
| 181 |
-
"loss": 0.
|
| 182 |
"step": 725
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch": 0.
|
| 186 |
-
"learning_rate": 9.
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 750
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch": 0.
|
| 192 |
-
"learning_rate": 9.
|
| 193 |
-
"loss": 0.
|
| 194 |
"step": 775
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch": 0.
|
| 198 |
-
"learning_rate": 9.
|
| 199 |
-
"loss": 0.
|
| 200 |
"step": 800
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch": 0.
|
| 204 |
-
"learning_rate": 9.
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 825
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch": 0.
|
| 210 |
-
"learning_rate": 9.
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 850
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"learning_rate": 9.
|
| 217 |
-
"loss": 0.
|
| 218 |
"step": 875
|
| 219 |
},
|
| 220 |
{
|
| 221 |
-
"epoch": 0.
|
| 222 |
-
"learning_rate": 9.
|
| 223 |
-
"loss": 0.
|
| 224 |
"step": 900
|
| 225 |
},
|
| 226 |
{
|
| 227 |
-
"epoch": 0.
|
| 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": 0.
|
| 240 |
-
"learning_rate":
|
| 241 |
-
"loss": 0.
|
| 242 |
"step": 975
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch": 0.
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 1000
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch": 0.
|
| 252 |
-
"eval_loss": 0.
|
| 253 |
-
"eval_runtime":
|
| 254 |
-
"eval_samples_per_second": 3.
|
| 255 |
-
"eval_steps_per_second": 0.
|
| 256 |
-
"eval_wer": 28.
|
| 257 |
"step": 1000
|
| 258 |
},
|
| 259 |
{
|
| 260 |
-
"epoch": 0.
|
| 261 |
-
"learning_rate":
|
| 262 |
-
"loss": 0.
|
| 263 |
"step": 1025
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch": 0.
|
| 267 |
-
"learning_rate":
|
| 268 |
"loss": 0.3674,
|
| 269 |
"step": 1050
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch": 0.
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 1075
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 0.
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 1100
|
| 282 |
},
|
| 283 |
{
|
| 284 |
-
"epoch": 0.
|
| 285 |
-
"learning_rate":
|
| 286 |
-
"loss": 0.
|
| 287 |
"step": 1125
|
| 288 |
},
|
| 289 |
{
|
| 290 |
-
"epoch": 0.
|
| 291 |
-
"learning_rate":
|
| 292 |
-
"loss": 0.
|
| 293 |
"step": 1150
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"epoch": 0.
|
| 297 |
-
"learning_rate": 8.
|
| 298 |
-
"loss": 0.
|
| 299 |
"step": 1175
|
| 300 |
},
|
| 301 |
{
|
| 302 |
-
"epoch": 0.
|
| 303 |
-
"learning_rate": 8.
|
| 304 |
-
"loss": 0.
|
| 305 |
"step": 1200
|
| 306 |
},
|
| 307 |
{
|
| 308 |
-
"epoch": 0.
|
| 309 |
-
"learning_rate": 8.
|
| 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": 0.
|
| 327 |
-
"learning_rate": 8.
|
| 328 |
-
"loss": 0.
|
| 329 |
"step": 1300
|
| 330 |
},
|
| 331 |
{
|
| 332 |
-
"epoch": 0.
|
| 333 |
-
"learning_rate": 8.
|
| 334 |
-
"loss": 0.
|
| 335 |
"step": 1325
|
| 336 |
},
|
| 337 |
{
|
| 338 |
-
"epoch": 0.
|
| 339 |
-
"learning_rate": 8.
|
| 340 |
-
"loss": 0.
|
| 341 |
"step": 1350
|
| 342 |
},
|
| 343 |
{
|
| 344 |
-
"epoch": 0.
|
| 345 |
-
"learning_rate": 8.
|
| 346 |
-
"loss": 0.
|
| 347 |
"step": 1375
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch": 0.
|
| 351 |
-
"learning_rate": 8.
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 1400
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch": 0.
|
| 357 |
-
"learning_rate":
|
| 358 |
-
"loss": 0.
|
| 359 |
"step": 1425
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 0.
|
| 363 |
-
"learning_rate":
|
| 364 |
-
"loss": 0.
|
| 365 |
"step": 1450
|
| 366 |
},
|
| 367 |
{
|
| 368 |
-
"epoch": 0.
|
| 369 |
-
"learning_rate":
|
| 370 |
-
"loss": 0.
|
| 371 |
"step": 1475
|
| 372 |
},
|
| 373 |
{
|
| 374 |
-
"epoch": 0.
|
| 375 |
-
"learning_rate":
|
| 376 |
-
"loss": 0.
|
| 377 |
"step": 1500
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch": 0.
|
| 381 |
-
"learning_rate":
|
| 382 |
-
"loss": 0.
|
| 383 |
"step": 1525
|
| 384 |
},
|
| 385 |
{
|
| 386 |
-
"epoch": 0.
|
| 387 |
-
"learning_rate":
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 1550
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch": 0.
|
| 393 |
-
"learning_rate":
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 1575
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch": 0.
|
| 399 |
-
"learning_rate":
|
| 400 |
-
"loss": 0.
|
| 401 |
"step": 1600
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 0.
|
| 405 |
-
"learning_rate":
|
| 406 |
-
"loss": 0.
|
| 407 |
"step": 1625
|
| 408 |
},
|
| 409 |
{
|
| 410 |
-
"epoch": 0.
|
| 411 |
-
"learning_rate":
|
| 412 |
-
"loss": 0.
|
| 413 |
"step": 1650
|
| 414 |
},
|
| 415 |
{
|
| 416 |
-
"epoch": 0.
|
| 417 |
-
"learning_rate":
|
| 418 |
-
"loss": 0.
|
| 419 |
"step": 1675
|
| 420 |
},
|
| 421 |
{
|
| 422 |
-
"epoch": 0.
|
| 423 |
-
"learning_rate":
|
| 424 |
-
"loss": 0.
|
| 425 |
"step": 1700
|
| 426 |
},
|
| 427 |
{
|
| 428 |
-
"epoch": 0.
|
| 429 |
-
"learning_rate":
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 1725
|
| 432 |
},
|
| 433 |
{
|
| 434 |
-
"epoch": 0.
|
| 435 |
-
"learning_rate":
|
| 436 |
-
"loss": 0.
|
| 437 |
"step": 1750
|
| 438 |
},
|
| 439 |
{
|
| 440 |
-
"epoch": 0.
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss": 0.
|
| 443 |
"step": 1775
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 0.
|
| 447 |
-
"learning_rate":
|
| 448 |
-
"loss": 0.
|
| 449 |
"step": 1800
|
| 450 |
},
|
| 451 |
{
|
| 452 |
-
"epoch": 0.
|
| 453 |
-
"learning_rate": 7.
|
| 454 |
-
"loss": 0.
|
| 455 |
"step": 1825
|
| 456 |
},
|
| 457 |
{
|
| 458 |
-
"epoch": 0.
|
| 459 |
-
"learning_rate": 7.
|
| 460 |
-
"loss": 0.
|
| 461 |
"step": 1850
|
| 462 |
},
|
| 463 |
{
|
| 464 |
-
"epoch": 0.
|
| 465 |
-
"learning_rate":
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 1875
|
| 468 |
},
|
| 469 |
{
|
| 470 |
-
"epoch": 0.
|
| 471 |
-
"learning_rate":
|
| 472 |
-
"loss": 0.
|
| 473 |
"step": 1900
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"epoch": 0.
|
| 477 |
-
"learning_rate":
|
| 478 |
-
"loss": 0.
|
| 479 |
"step": 1925
|
| 480 |
},
|
| 481 |
{
|
| 482 |
-
"epoch": 0.
|
| 483 |
-
"learning_rate":
|
| 484 |
-
"loss": 0.
|
| 485 |
"step": 1950
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 0.
|
| 489 |
-
"learning_rate":
|
| 490 |
-
"loss": 0.
|
| 491 |
"step": 1975
|
| 492 |
},
|
| 493 |
{
|
| 494 |
-
"epoch": 0.
|
| 495 |
-
"learning_rate":
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 2000
|
| 498 |
},
|
| 499 |
{
|
| 500 |
-
"epoch": 0.
|
| 501 |
-
"eval_loss": 0.
|
| 502 |
-
"eval_runtime":
|
| 503 |
-
"eval_samples_per_second": 3.
|
| 504 |
-
"eval_steps_per_second": 0.
|
| 505 |
-
"eval_wer":
|
| 506 |
"step": 2000
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch": 0.
|
| 510 |
-
"learning_rate":
|
| 511 |
-
"loss": 0.
|
| 512 |
"step": 2025
|
| 513 |
},
|
| 514 |
{
|
| 515 |
-
"epoch": 0.
|
| 516 |
-
"learning_rate":
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 2050
|
| 519 |
},
|
| 520 |
{
|
| 521 |
-
"epoch": 0.
|
| 522 |
-
"learning_rate":
|
| 523 |
-
"loss": 0.
|
| 524 |
"step": 2075
|
| 525 |
},
|
| 526 |
{
|
| 527 |
-
"epoch": 0.
|
| 528 |
-
"learning_rate":
|
| 529 |
-
"loss": 0.
|
| 530 |
"step": 2100
|
| 531 |
},
|
| 532 |
{
|
| 533 |
-
"epoch": 0.
|
| 534 |
-
"learning_rate":
|
| 535 |
-
"loss": 0.
|
| 536 |
"step": 2125
|
| 537 |
},
|
| 538 |
{
|
| 539 |
-
"epoch": 0.
|
| 540 |
-
"learning_rate":
|
| 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": 0.
|
| 558 |
-
"learning_rate":
|
| 559 |
-
"loss": 0.
|
| 560 |
"step": 2225
|
| 561 |
},
|
| 562 |
{
|
| 563 |
-
"epoch": 0.
|
| 564 |
-
"learning_rate":
|
| 565 |
-
"loss": 0.
|
| 566 |
"step": 2250
|
| 567 |
},
|
| 568 |
{
|
| 569 |
-
"epoch": 0.
|
| 570 |
-
"learning_rate":
|
| 571 |
-
"loss": 0.
|
| 572 |
"step": 2275
|
| 573 |
},
|
| 574 |
{
|
| 575 |
-
"epoch": 0.
|
| 576 |
-
"learning_rate":
|
| 577 |
-
"loss": 0.
|
| 578 |
"step": 2300
|
| 579 |
},
|
| 580 |
{
|
| 581 |
-
"epoch": 0.
|
| 582 |
-
"learning_rate":
|
| 583 |
-
"loss": 0.
|
| 584 |
"step": 2325
|
| 585 |
},
|
| 586 |
{
|
| 587 |
-
"epoch": 0.
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 2350
|
| 591 |
},
|
| 592 |
{
|
| 593 |
-
"epoch": 0.
|
| 594 |
-
"learning_rate":
|
| 595 |
-
"loss": 0.
|
| 596 |
"step": 2375
|
| 597 |
},
|
| 598 |
{
|
| 599 |
-
"epoch": 0.
|
| 600 |
-
"learning_rate":
|
| 601 |
-
"loss": 0.
|
| 602 |
"step": 2400
|
| 603 |
},
|
| 604 |
{
|
| 605 |
-
"epoch": 0.
|
| 606 |
-
"learning_rate":
|
| 607 |
-
"loss": 0.
|
| 608 |
"step": 2425
|
| 609 |
},
|
| 610 |
{
|
| 611 |
-
"epoch": 0.
|
| 612 |
-
"learning_rate":
|
| 613 |
-
"loss": 0.
|
| 614 |
"step": 2450
|
| 615 |
},
|
| 616 |
{
|
| 617 |
-
"epoch": 0.
|
| 618 |
-
"learning_rate":
|
| 619 |
-
"loss": 0.
|
| 620 |
"step": 2475
|
| 621 |
},
|
| 622 |
{
|
| 623 |
-
"epoch": 0.
|
| 624 |
-
"learning_rate":
|
| 625 |
-
"loss": 0.
|
| 626 |
"step": 2500
|
| 627 |
},
|
| 628 |
{
|
| 629 |
-
"epoch": 0.
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 2525
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch": 0.
|
| 636 |
-
"learning_rate":
|
| 637 |
-
"loss": 0.
|
| 638 |
"step": 2550
|
| 639 |
},
|
| 640 |
{
|
| 641 |
-
"epoch": 0.
|
| 642 |
-
"learning_rate":
|
| 643 |
-
"loss": 0.
|
| 644 |
"step": 2575
|
| 645 |
},
|
| 646 |
{
|
| 647 |
-
"epoch": 0.
|
| 648 |
-
"learning_rate":
|
| 649 |
-
"loss": 0.
|
| 650 |
"step": 2600
|
| 651 |
},
|
| 652 |
{
|
| 653 |
-
"epoch": 0.
|
| 654 |
-
"learning_rate":
|
| 655 |
-
"loss": 0.
|
| 656 |
"step": 2625
|
| 657 |
},
|
| 658 |
{
|
| 659 |
-
"epoch": 0.
|
| 660 |
-
"learning_rate":
|
| 661 |
-
"loss": 0.
|
| 662 |
"step": 2650
|
| 663 |
},
|
| 664 |
{
|
| 665 |
-
"epoch": 0.
|
| 666 |
-
"learning_rate":
|
| 667 |
-
"loss": 0.
|
| 668 |
"step": 2675
|
| 669 |
},
|
| 670 |
{
|
| 671 |
-
"epoch": 0.
|
| 672 |
-
"learning_rate":
|
| 673 |
-
"loss": 0.
|
| 674 |
"step": 2700
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch": 0.
|
| 678 |
-
"learning_rate":
|
| 679 |
-
"loss": 0.
|
| 680 |
"step": 2725
|
| 681 |
},
|
| 682 |
{
|
| 683 |
-
"epoch": 0.
|
| 684 |
-
"learning_rate":
|
| 685 |
-
"loss": 0.
|
| 686 |
"step": 2750
|
| 687 |
},
|
| 688 |
{
|
| 689 |
-
"epoch": 0.
|
| 690 |
-
"learning_rate":
|
| 691 |
-
"loss": 0.
|
| 692 |
"step": 2775
|
| 693 |
},
|
| 694 |
{
|
| 695 |
-
"epoch": 0.
|
| 696 |
-
"learning_rate":
|
| 697 |
-
"loss": 0.
|
| 698 |
"step": 2800
|
| 699 |
},
|
| 700 |
{
|
| 701 |
-
"epoch": 0.
|
| 702 |
-
"learning_rate":
|
| 703 |
-
"loss": 0.
|
| 704 |
"step": 2825
|
| 705 |
},
|
| 706 |
{
|
| 707 |
-
"epoch": 0.
|
| 708 |
-
"learning_rate":
|
| 709 |
-
"loss": 0.
|
| 710 |
"step": 2850
|
| 711 |
},
|
| 712 |
{
|
| 713 |
-
"epoch": 0.
|
| 714 |
-
"learning_rate":
|
| 715 |
-
"loss": 0.
|
| 716 |
"step": 2875
|
| 717 |
},
|
| 718 |
{
|
| 719 |
-
"epoch": 0.
|
| 720 |
-
"learning_rate":
|
| 721 |
-
"loss": 0.
|
| 722 |
"step": 2900
|
| 723 |
},
|
| 724 |
{
|
| 725 |
-
"epoch": 0.
|
| 726 |
-
"learning_rate":
|
| 727 |
-
"loss": 0.
|
| 728 |
"step": 2925
|
| 729 |
},
|
| 730 |
{
|
| 731 |
-
"epoch": 0.
|
| 732 |
-
"learning_rate":
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 2950
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"epoch": 0.
|
| 738 |
-
"learning_rate":
|
| 739 |
-
"loss": 0.
|
| 740 |
"step": 2975
|
| 741 |
},
|
| 742 |
{
|
| 743 |
-
"epoch": 0.
|
| 744 |
-
"learning_rate":
|
| 745 |
-
"loss": 0.
|
| 746 |
"step": 3000
|
| 747 |
},
|
| 748 |
{
|
| 749 |
-
"epoch": 0.
|
| 750 |
-
"eval_loss": 0.
|
| 751 |
-
"eval_runtime":
|
| 752 |
-
"eval_samples_per_second": 3.
|
| 753 |
-
"eval_steps_per_second": 0.
|
| 754 |
-
"eval_wer":
|
| 755 |
"step": 3000
|
| 756 |
}
|
| 757 |
],
|
| 758 |
-
"max_steps":
|
| 759 |
"num_train_epochs": 9223372036854775807,
|
| 760 |
"total_flos": 1.224725889024e+19,
|
| 761 |
"trial_name": null,
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 18.14370772408765,
|
| 3 |
"best_model_checkpoint": "./checkpoint-3000",
|
| 4 |
+
"epoch": 0.42857142857142855,
|
| 5 |
"global_step": 3000,
|
| 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": 4.6000000000000004e-07,
|
| 13 |
"loss": 1.4182,
|
| 14 |
"step": 25
|
|
|
|
| 26 |
"step": 75
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.01,
|
| 30 |
"learning_rate": 1.94e-06,
|
| 31 |
"loss": 0.7765,
|
| 32 |
"step": 100
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 0.02,
|
| 36 |
"learning_rate": 2.4400000000000004e-06,
|
| 37 |
"loss": 0.7103,
|
| 38 |
"step": 125
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.02,
|
| 42 |
"learning_rate": 2.9400000000000002e-06,
|
| 43 |
"loss": 0.6597,
|
| 44 |
"step": 150
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.03,
|
| 48 |
"learning_rate": 3.44e-06,
|
| 49 |
"loss": 0.6657,
|
| 50 |
"step": 175
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 0.03,
|
| 54 |
"learning_rate": 3.94e-06,
|
| 55 |
"loss": 0.5853,
|
| 56 |
"step": 200
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 0.03,
|
| 60 |
"learning_rate": 4.440000000000001e-06,
|
| 61 |
"loss": 0.5273,
|
| 62 |
"step": 225
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 0.04,
|
| 66 |
"learning_rate": 4.94e-06,
|
| 67 |
"loss": 0.5979,
|
| 68 |
"step": 250
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 0.04,
|
| 72 |
"learning_rate": 5.4400000000000004e-06,
|
| 73 |
"loss": 0.5861,
|
| 74 |
"step": 275
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 0.04,
|
| 78 |
"learning_rate": 5.94e-06,
|
| 79 |
"loss": 0.5085,
|
| 80 |
"step": 300
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 0.05,
|
| 84 |
"learning_rate": 6.440000000000001e-06,
|
| 85 |
"loss": 0.4827,
|
| 86 |
"step": 325
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.05,
|
| 90 |
"learning_rate": 6.9400000000000005e-06,
|
| 91 |
"loss": 0.4909,
|
| 92 |
"step": 350
|
| 93 |
},
|
| 94 |
{
|
| 95 |
+
"epoch": 0.05,
|
| 96 |
"learning_rate": 7.440000000000001e-06,
|
| 97 |
"loss": 0.4651,
|
| 98 |
"step": 375
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 0.06,
|
| 102 |
"learning_rate": 7.94e-06,
|
| 103 |
"loss": 0.494,
|
| 104 |
"step": 400
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 0.06,
|
| 108 |
"learning_rate": 8.44e-06,
|
| 109 |
"loss": 0.4188,
|
| 110 |
"step": 425
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"epoch": 0.06,
|
| 114 |
"learning_rate": 8.94e-06,
|
| 115 |
"loss": 0.3849,
|
| 116 |
"step": 450
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 0.07,
|
| 120 |
"learning_rate": 9.440000000000001e-06,
|
| 121 |
"loss": 0.4577,
|
| 122 |
"step": 475
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 0.07,
|
| 126 |
"learning_rate": 9.940000000000001e-06,
|
| 127 |
"loss": 0.4415,
|
| 128 |
"step": 500
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.07,
|
| 132 |
+
"learning_rate": 9.966153846153847e-06,
|
| 133 |
+
"loss": 0.4614,
|
| 134 |
"step": 525
|
| 135 |
},
|
| 136 |
{
|
| 137 |
+
"epoch": 0.08,
|
| 138 |
+
"learning_rate": 9.927692307692309e-06,
|
| 139 |
+
"loss": 0.4283,
|
| 140 |
"step": 550
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"epoch": 0.08,
|
| 144 |
+
"learning_rate": 9.88923076923077e-06,
|
| 145 |
+
"loss": 0.4486,
|
| 146 |
"step": 575
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 0.09,
|
| 150 |
+
"learning_rate": 9.850769230769231e-06,
|
| 151 |
+
"loss": 0.4434,
|
| 152 |
"step": 600
|
| 153 |
},
|
| 154 |
{
|
| 155 |
+
"epoch": 0.09,
|
| 156 |
+
"learning_rate": 9.812307692307694e-06,
|
| 157 |
+
"loss": 0.4245,
|
| 158 |
"step": 625
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 0.09,
|
| 162 |
+
"learning_rate": 9.773846153846154e-06,
|
| 163 |
+
"loss": 0.4254,
|
| 164 |
"step": 650
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 0.1,
|
| 168 |
+
"learning_rate": 9.735384615384616e-06,
|
| 169 |
+
"loss": 0.4143,
|
| 170 |
"step": 675
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.1,
|
| 174 |
+
"learning_rate": 9.696923076923078e-06,
|
| 175 |
+
"loss": 0.4115,
|
| 176 |
"step": 700
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 0.1,
|
| 180 |
+
"learning_rate": 9.658461538461539e-06,
|
| 181 |
+
"loss": 0.3656,
|
| 182 |
"step": 725
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"epoch": 0.11,
|
| 186 |
+
"learning_rate": 9.620000000000001e-06,
|
| 187 |
+
"loss": 0.4004,
|
| 188 |
"step": 750
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 0.11,
|
| 192 |
+
"learning_rate": 9.581538461538462e-06,
|
| 193 |
+
"loss": 0.388,
|
| 194 |
"step": 775
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 0.11,
|
| 198 |
+
"learning_rate": 9.543076923076924e-06,
|
| 199 |
+
"loss": 0.3665,
|
| 200 |
"step": 800
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 0.12,
|
| 204 |
+
"learning_rate": 9.504615384615386e-06,
|
| 205 |
+
"loss": 0.3777,
|
| 206 |
"step": 825
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 0.12,
|
| 210 |
+
"learning_rate": 9.466153846153846e-06,
|
| 211 |
+
"loss": 0.4338,
|
| 212 |
"step": 850
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.12,
|
| 216 |
+
"learning_rate": 9.427692307692309e-06,
|
| 217 |
+
"loss": 0.4017,
|
| 218 |
"step": 875
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 0.13,
|
| 222 |
+
"learning_rate": 9.38923076923077e-06,
|
| 223 |
+
"loss": 0.3757,
|
| 224 |
"step": 900
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 0.13,
|
| 228 |
+
"learning_rate": 9.350769230769231e-06,
|
| 229 |
+
"loss": 0.4084,
|
| 230 |
"step": 925
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 0.14,
|
| 234 |
+
"learning_rate": 9.312307692307693e-06,
|
| 235 |
+
"loss": 0.3821,
|
| 236 |
"step": 950
|
| 237 |
},
|
| 238 |
{
|
| 239 |
+
"epoch": 0.14,
|
| 240 |
+
"learning_rate": 9.273846153846154e-06,
|
| 241 |
+
"loss": 0.421,
|
| 242 |
"step": 975
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 0.14,
|
| 246 |
+
"learning_rate": 9.235384615384616e-06,
|
| 247 |
+
"loss": 0.4203,
|
| 248 |
"step": 1000
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 0.14,
|
| 252 |
+
"eval_loss": 0.4127572178840637,
|
| 253 |
+
"eval_runtime": 1802.5775,
|
| 254 |
+
"eval_samples_per_second": 3.656,
|
| 255 |
+
"eval_steps_per_second": 0.457,
|
| 256 |
+
"eval_wer": 28.265624367127064,
|
| 257 |
"step": 1000
|
| 258 |
},
|
| 259 |
{
|
| 260 |
+
"epoch": 0.15,
|
| 261 |
+
"learning_rate": 9.196923076923078e-06,
|
| 262 |
+
"loss": 0.4094,
|
| 263 |
"step": 1025
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"epoch": 0.15,
|
| 267 |
+
"learning_rate": 9.158461538461539e-06,
|
| 268 |
"loss": 0.3674,
|
| 269 |
"step": 1050
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"epoch": 0.15,
|
| 273 |
+
"learning_rate": 9.12e-06,
|
| 274 |
+
"loss": 0.3613,
|
| 275 |
"step": 1075
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.16,
|
| 279 |
+
"learning_rate": 9.081538461538463e-06,
|
| 280 |
+
"loss": 0.3904,
|
| 281 |
"step": 1100
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 0.16,
|
| 285 |
+
"learning_rate": 9.043076923076923e-06,
|
| 286 |
+
"loss": 0.3865,
|
| 287 |
"step": 1125
|
| 288 |
},
|
| 289 |
{
|
| 290 |
+
"epoch": 0.16,
|
| 291 |
+
"learning_rate": 9.004615384615386e-06,
|
| 292 |
+
"loss": 0.3677,
|
| 293 |
"step": 1150
|
| 294 |
},
|
| 295 |
{
|
| 296 |
+
"epoch": 0.17,
|
| 297 |
+
"learning_rate": 8.966153846153846e-06,
|
| 298 |
+
"loss": 0.3295,
|
| 299 |
"step": 1175
|
| 300 |
},
|
| 301 |
{
|
| 302 |
+
"epoch": 0.17,
|
| 303 |
+
"learning_rate": 8.927692307692308e-06,
|
| 304 |
+
"loss": 0.3207,
|
| 305 |
"step": 1200
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"epoch": 0.17,
|
| 309 |
+
"learning_rate": 8.88923076923077e-06,
|
| 310 |
+
"loss": 0.2817,
|
| 311 |
"step": 1225
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"epoch": 0.18,
|
| 315 |
+
"learning_rate": 8.850769230769231e-06,
|
| 316 |
+
"loss": 0.3218,
|
| 317 |
"step": 1250
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.18,
|
| 321 |
+
"learning_rate": 8.812307692307693e-06,
|
| 322 |
+
"loss": 0.3502,
|
| 323 |
"step": 1275
|
| 324 |
},
|
| 325 |
{
|
| 326 |
+
"epoch": 0.19,
|
| 327 |
+
"learning_rate": 8.773846153846155e-06,
|
| 328 |
+
"loss": 0.3432,
|
| 329 |
"step": 1300
|
| 330 |
},
|
| 331 |
{
|
| 332 |
+
"epoch": 0.19,
|
| 333 |
+
"learning_rate": 8.735384615384616e-06,
|
| 334 |
+
"loss": 0.3244,
|
| 335 |
"step": 1325
|
| 336 |
},
|
| 337 |
{
|
| 338 |
+
"epoch": 0.19,
|
| 339 |
+
"learning_rate": 8.696923076923078e-06,
|
| 340 |
+
"loss": 0.2607,
|
| 341 |
"step": 1350
|
| 342 |
},
|
| 343 |
{
|
| 344 |
+
"epoch": 0.2,
|
| 345 |
+
"learning_rate": 8.658461538461538e-06,
|
| 346 |
+
"loss": 0.3469,
|
| 347 |
"step": 1375
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"epoch": 0.2,
|
| 351 |
+
"learning_rate": 8.62e-06,
|
| 352 |
+
"loss": 0.2756,
|
| 353 |
"step": 1400
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 0.2,
|
| 357 |
+
"learning_rate": 8.581538461538463e-06,
|
| 358 |
+
"loss": 0.294,
|
| 359 |
"step": 1425
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.21,
|
| 363 |
+
"learning_rate": 8.543076923076923e-06,
|
| 364 |
+
"loss": 0.3915,
|
| 365 |
"step": 1450
|
| 366 |
},
|
| 367 |
{
|
| 368 |
+
"epoch": 0.21,
|
| 369 |
+
"learning_rate": 8.504615384615385e-06,
|
| 370 |
+
"loss": 0.3237,
|
| 371 |
"step": 1475
|
| 372 |
},
|
| 373 |
{
|
| 374 |
+
"epoch": 0.21,
|
| 375 |
+
"learning_rate": 8.466153846153847e-06,
|
| 376 |
+
"loss": 0.2723,
|
| 377 |
"step": 1500
|
| 378 |
},
|
| 379 |
{
|
| 380 |
+
"epoch": 0.22,
|
| 381 |
+
"learning_rate": 8.427692307692308e-06,
|
| 382 |
+
"loss": 0.2917,
|
| 383 |
"step": 1525
|
| 384 |
},
|
| 385 |
{
|
| 386 |
+
"epoch": 0.22,
|
| 387 |
+
"learning_rate": 8.38923076923077e-06,
|
| 388 |
+
"loss": 0.2921,
|
| 389 |
"step": 1550
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 0.23,
|
| 393 |
+
"learning_rate": 8.35076923076923e-06,
|
| 394 |
+
"loss": 0.3724,
|
| 395 |
"step": 1575
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 0.23,
|
| 399 |
+
"learning_rate": 8.312307692307693e-06,
|
| 400 |
+
"loss": 0.3364,
|
| 401 |
"step": 1600
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.23,
|
| 405 |
+
"learning_rate": 8.273846153846155e-06,
|
| 406 |
+
"loss": 0.3661,
|
| 407 |
"step": 1625
|
| 408 |
},
|
| 409 |
{
|
| 410 |
+
"epoch": 0.24,
|
| 411 |
+
"learning_rate": 8.235384615384615e-06,
|
| 412 |
+
"loss": 0.3307,
|
| 413 |
"step": 1650
|
| 414 |
},
|
| 415 |
{
|
| 416 |
+
"epoch": 0.24,
|
| 417 |
+
"learning_rate": 8.196923076923078e-06,
|
| 418 |
+
"loss": 0.3201,
|
| 419 |
"step": 1675
|
| 420 |
},
|
| 421 |
{
|
| 422 |
+
"epoch": 0.24,
|
| 423 |
+
"learning_rate": 8.15846153846154e-06,
|
| 424 |
+
"loss": 0.2973,
|
| 425 |
"step": 1700
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"epoch": 0.25,
|
| 429 |
+
"learning_rate": 8.120000000000002e-06,
|
| 430 |
+
"loss": 0.3013,
|
| 431 |
"step": 1725
|
| 432 |
},
|
| 433 |
{
|
| 434 |
+
"epoch": 0.25,
|
| 435 |
+
"learning_rate": 8.081538461538462e-06,
|
| 436 |
+
"loss": 0.3349,
|
| 437 |
"step": 1750
|
| 438 |
},
|
| 439 |
{
|
| 440 |
+
"epoch": 0.25,
|
| 441 |
+
"learning_rate": 8.043076923076923e-06,
|
| 442 |
+
"loss": 0.3244,
|
| 443 |
"step": 1775
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.26,
|
| 447 |
+
"learning_rate": 8.004615384615385e-06,
|
| 448 |
+
"loss": 0.3331,
|
| 449 |
"step": 1800
|
| 450 |
},
|
| 451 |
{
|
| 452 |
+
"epoch": 0.26,
|
| 453 |
+
"learning_rate": 7.966153846153847e-06,
|
| 454 |
+
"loss": 0.2723,
|
| 455 |
"step": 1825
|
| 456 |
},
|
| 457 |
{
|
| 458 |
+
"epoch": 0.26,
|
| 459 |
+
"learning_rate": 7.927692307692308e-06,
|
| 460 |
+
"loss": 0.316,
|
| 461 |
"step": 1850
|
| 462 |
},
|
| 463 |
{
|
| 464 |
+
"epoch": 0.27,
|
| 465 |
+
"learning_rate": 7.88923076923077e-06,
|
| 466 |
+
"loss": 0.2648,
|
| 467 |
"step": 1875
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 0.27,
|
| 471 |
+
"learning_rate": 7.850769230769232e-06,
|
| 472 |
+
"loss": 0.2965,
|
| 473 |
"step": 1900
|
| 474 |
},
|
| 475 |
{
|
| 476 |
+
"epoch": 0.28,
|
| 477 |
+
"learning_rate": 7.812307692307694e-06,
|
| 478 |
+
"loss": 0.256,
|
| 479 |
"step": 1925
|
| 480 |
},
|
| 481 |
{
|
| 482 |
+
"epoch": 0.28,
|
| 483 |
+
"learning_rate": 7.773846153846155e-06,
|
| 484 |
+
"loss": 0.2156,
|
| 485 |
"step": 1950
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.28,
|
| 489 |
+
"learning_rate": 7.735384615384615e-06,
|
| 490 |
+
"loss": 0.2418,
|
| 491 |
"step": 1975
|
| 492 |
},
|
| 493 |
{
|
| 494 |
+
"epoch": 0.29,
|
| 495 |
+
"learning_rate": 7.696923076923077e-06,
|
| 496 |
+
"loss": 0.2693,
|
| 497 |
"step": 2000
|
| 498 |
},
|
| 499 |
{
|
| 500 |
+
"epoch": 0.29,
|
| 501 |
+
"eval_loss": 0.3239877223968506,
|
| 502 |
+
"eval_runtime": 1817.7748,
|
| 503 |
+
"eval_samples_per_second": 3.626,
|
| 504 |
+
"eval_steps_per_second": 0.453,
|
| 505 |
+
"eval_wer": 22.052330997610273,
|
| 506 |
"step": 2000
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 0.29,
|
| 510 |
+
"learning_rate": 7.65846153846154e-06,
|
| 511 |
+
"loss": 0.2191,
|
| 512 |
"step": 2025
|
| 513 |
},
|
| 514 |
{
|
| 515 |
+
"epoch": 0.29,
|
| 516 |
+
"learning_rate": 7.620000000000001e-06,
|
| 517 |
+
"loss": 0.2437,
|
| 518 |
"step": 2050
|
| 519 |
},
|
| 520 |
{
|
| 521 |
+
"epoch": 0.3,
|
| 522 |
+
"learning_rate": 7.581538461538462e-06,
|
| 523 |
+
"loss": 0.2519,
|
| 524 |
"step": 2075
|
| 525 |
},
|
| 526 |
{
|
| 527 |
+
"epoch": 0.3,
|
| 528 |
+
"learning_rate": 7.544615384615386e-06,
|
| 529 |
+
"loss": 0.3198,
|
| 530 |
"step": 2100
|
| 531 |
},
|
| 532 |
{
|
| 533 |
+
"epoch": 0.3,
|
| 534 |
+
"learning_rate": 7.5061538461538465e-06,
|
| 535 |
+
"loss": 0.2708,
|
| 536 |
"step": 2125
|
| 537 |
},
|
| 538 |
{
|
| 539 |
+
"epoch": 0.31,
|
| 540 |
+
"learning_rate": 7.467692307692308e-06,
|
| 541 |
+
"loss": 0.265,
|
| 542 |
"step": 2150
|
| 543 |
},
|
| 544 |
{
|
| 545 |
+
"epoch": 0.31,
|
| 546 |
+
"learning_rate": 7.42923076923077e-06,
|
| 547 |
+
"loss": 0.2935,
|
| 548 |
"step": 2175
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 0.31,
|
| 552 |
+
"learning_rate": 7.390769230769231e-06,
|
| 553 |
+
"loss": 0.2595,
|
| 554 |
"step": 2200
|
| 555 |
},
|
| 556 |
{
|
| 557 |
+
"epoch": 0.32,
|
| 558 |
+
"learning_rate": 7.3523076923076935e-06,
|
| 559 |
+
"loss": 0.2613,
|
| 560 |
"step": 2225
|
| 561 |
},
|
| 562 |
{
|
| 563 |
+
"epoch": 0.32,
|
| 564 |
+
"learning_rate": 7.313846153846155e-06,
|
| 565 |
+
"loss": 0.2229,
|
| 566 |
"step": 2250
|
| 567 |
},
|
| 568 |
{
|
| 569 |
+
"epoch": 0.33,
|
| 570 |
+
"learning_rate": 7.275384615384616e-06,
|
| 571 |
+
"loss": 0.2814,
|
| 572 |
"step": 2275
|
| 573 |
},
|
| 574 |
{
|
| 575 |
+
"epoch": 0.33,
|
| 576 |
+
"learning_rate": 7.236923076923078e-06,
|
| 577 |
+
"loss": 0.2681,
|
| 578 |
"step": 2300
|
| 579 |
},
|
| 580 |
{
|
| 581 |
+
"epoch": 0.33,
|
| 582 |
+
"learning_rate": 7.198461538461539e-06,
|
| 583 |
+
"loss": 0.2531,
|
| 584 |
"step": 2325
|
| 585 |
},
|
| 586 |
{
|
| 587 |
+
"epoch": 0.34,
|
| 588 |
+
"learning_rate": 7.16e-06,
|
| 589 |
+
"loss": 0.2104,
|
| 590 |
"step": 2350
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 0.34,
|
| 594 |
+
"learning_rate": 7.121538461538462e-06,
|
| 595 |
+
"loss": 0.2294,
|
| 596 |
"step": 2375
|
| 597 |
},
|
| 598 |
{
|
| 599 |
+
"epoch": 0.34,
|
| 600 |
+
"learning_rate": 7.0830769230769235e-06,
|
| 601 |
+
"loss": 0.2556,
|
| 602 |
"step": 2400
|
| 603 |
},
|
| 604 |
{
|
| 605 |
+
"epoch": 0.35,
|
| 606 |
+
"learning_rate": 7.044615384615386e-06,
|
| 607 |
+
"loss": 0.2185,
|
| 608 |
"step": 2425
|
| 609 |
},
|
| 610 |
{
|
| 611 |
+
"epoch": 0.35,
|
| 612 |
+
"learning_rate": 7.006153846153847e-06,
|
| 613 |
+
"loss": 0.2585,
|
| 614 |
"step": 2450
|
| 615 |
},
|
| 616 |
{
|
| 617 |
+
"epoch": 0.35,
|
| 618 |
+
"learning_rate": 6.967692307692308e-06,
|
| 619 |
+
"loss": 0.2325,
|
| 620 |
"step": 2475
|
| 621 |
},
|
| 622 |
{
|
| 623 |
+
"epoch": 0.36,
|
| 624 |
+
"learning_rate": 6.9292307692307705e-06,
|
| 625 |
+
"loss": 0.2068,
|
| 626 |
"step": 2500
|
| 627 |
},
|
| 628 |
{
|
| 629 |
+
"epoch": 0.36,
|
| 630 |
+
"learning_rate": 6.890769230769231e-06,
|
| 631 |
+
"loss": 0.2233,
|
| 632 |
"step": 2525
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 0.36,
|
| 636 |
+
"learning_rate": 6.852307692307692e-06,
|
| 637 |
+
"loss": 0.2649,
|
| 638 |
"step": 2550
|
| 639 |
},
|
| 640 |
{
|
| 641 |
+
"epoch": 0.37,
|
| 642 |
+
"learning_rate": 6.8138461538461545e-06,
|
| 643 |
+
"loss": 0.2524,
|
| 644 |
"step": 2575
|
| 645 |
},
|
| 646 |
{
|
| 647 |
+
"epoch": 0.37,
|
| 648 |
+
"learning_rate": 6.775384615384616e-06,
|
| 649 |
+
"loss": 0.2847,
|
| 650 |
"step": 2600
|
| 651 |
},
|
| 652 |
{
|
| 653 |
+
"epoch": 0.38,
|
| 654 |
+
"learning_rate": 6.736923076923078e-06,
|
| 655 |
+
"loss": 0.1954,
|
| 656 |
"step": 2625
|
| 657 |
},
|
| 658 |
{
|
| 659 |
+
"epoch": 0.38,
|
| 660 |
+
"learning_rate": 6.698461538461539e-06,
|
| 661 |
+
"loss": 0.2553,
|
| 662 |
"step": 2650
|
| 663 |
},
|
| 664 |
{
|
| 665 |
+
"epoch": 0.38,
|
| 666 |
+
"learning_rate": 6.660000000000001e-06,
|
| 667 |
+
"loss": 0.2138,
|
| 668 |
"step": 2675
|
| 669 |
},
|
| 670 |
{
|
| 671 |
+
"epoch": 0.39,
|
| 672 |
+
"learning_rate": 6.621538461538463e-06,
|
| 673 |
+
"loss": 0.2245,
|
| 674 |
"step": 2700
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 0.39,
|
| 678 |
+
"learning_rate": 6.583076923076923e-06,
|
| 679 |
+
"loss": 0.2137,
|
| 680 |
"step": 2725
|
| 681 |
},
|
| 682 |
{
|
| 683 |
+
"epoch": 0.39,
|
| 684 |
+
"learning_rate": 6.5446153846153846e-06,
|
| 685 |
+
"loss": 0.2558,
|
| 686 |
"step": 2750
|
| 687 |
},
|
| 688 |
{
|
| 689 |
+
"epoch": 0.4,
|
| 690 |
+
"learning_rate": 6.506153846153847e-06,
|
| 691 |
+
"loss": 0.1981,
|
| 692 |
"step": 2775
|
| 693 |
},
|
| 694 |
{
|
| 695 |
+
"epoch": 0.4,
|
| 696 |
+
"learning_rate": 6.467692307692308e-06,
|
| 697 |
+
"loss": 0.2327,
|
| 698 |
"step": 2800
|
| 699 |
},
|
| 700 |
{
|
| 701 |
+
"epoch": 0.4,
|
| 702 |
+
"learning_rate": 6.42923076923077e-06,
|
| 703 |
+
"loss": 0.219,
|
| 704 |
"step": 2825
|
| 705 |
},
|
| 706 |
{
|
| 707 |
+
"epoch": 0.41,
|
| 708 |
+
"learning_rate": 6.3907692307692315e-06,
|
| 709 |
+
"loss": 0.226,
|
| 710 |
"step": 2850
|
| 711 |
},
|
| 712 |
{
|
| 713 |
+
"epoch": 0.41,
|
| 714 |
+
"learning_rate": 6.352307692307693e-06,
|
| 715 |
+
"loss": 0.2192,
|
| 716 |
"step": 2875
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 0.41,
|
| 720 |
+
"learning_rate": 6.313846153846155e-06,
|
| 721 |
+
"loss": 0.2146,
|
| 722 |
"step": 2900
|
| 723 |
},
|
| 724 |
{
|
| 725 |
+
"epoch": 0.42,
|
| 726 |
+
"learning_rate": 6.2753846153846155e-06,
|
| 727 |
+
"loss": 0.2058,
|
| 728 |
"step": 2925
|
| 729 |
},
|
| 730 |
{
|
| 731 |
+
"epoch": 0.42,
|
| 732 |
+
"learning_rate": 6.236923076923077e-06,
|
| 733 |
+
"loss": 0.2369,
|
| 734 |
"step": 2950
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 0.42,
|
| 738 |
+
"learning_rate": 6.198461538461539e-06,
|
| 739 |
+
"loss": 0.2202,
|
| 740 |
"step": 2975
|
| 741 |
},
|
| 742 |
{
|
| 743 |
+
"epoch": 0.43,
|
| 744 |
+
"learning_rate": 6.16e-06,
|
| 745 |
+
"loss": 0.2228,
|
| 746 |
"step": 3000
|
| 747 |
},
|
| 748 |
{
|
| 749 |
+
"epoch": 0.43,
|
| 750 |
+
"eval_loss": 0.27374517917633057,
|
| 751 |
+
"eval_runtime": 1805.7724,
|
| 752 |
+
"eval_samples_per_second": 3.65,
|
| 753 |
+
"eval_steps_per_second": 0.456,
|
| 754 |
+
"eval_wer": 18.14370772408765,
|
| 755 |
"step": 3000
|
| 756 |
}
|
| 757 |
],
|
| 758 |
+
"max_steps": 7000,
|
| 759 |
"num_train_epochs": 9223372036854775807,
|
| 760 |
"total_flos": 1.224725889024e+19,
|
| 761 |
"trial_name": null,
|
checkpoint-3000/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3643
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1ca27d9d1bef820c87cfd52d5adb15af5250e1ac4555cd1703a088ad0dad1a4
|
| 3 |
size 3643
|
checkpoint-4000/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 6111428695
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1d10d5e2e5766ead361dc8ecf77781538635af5fdd98db487bfde99823ce4a07
|
| 3 |
size 6111428695
|
checkpoint-4000/pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3055754841
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:def26fdaf433678a4a7ce858c8c045bcb0081f1a9d18211fd996f65174b4429e
|
| 3 |
size 3055754841
|
checkpoint-4000/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:139a67ed18200d5b28ad0b83b97b2bbd7bf683a055720be06a088fa3bb073509
|
| 3 |
size 14575
|
checkpoint-4000/scaler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 557
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8b27cbf906faccb76347b161e471af6eaccfe3470862a556cf92db01b1d1b35b
|
| 3 |
size 557
|
checkpoint-4000/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8c8724f1b8669fb50a95fa5a9cd53bb72dbb12e135946c3302c5f49b3cc9f276
|
| 3 |
size 627
|
checkpoint-4000/trainer_state.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
"best_model_checkpoint": "./checkpoint-4000",
|
| 4 |
-
"epoch": 1.
|
| 5 |
"global_step": 4000,
|
| 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.
|
| 12 |
"learning_rate": 4.6000000000000004e-07,
|
| 13 |
"loss": 1.4182,
|
| 14 |
"step": 25
|
|
@@ -26,985 +26,985 @@
|
|
| 26 |
"step": 75
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch": 0.
|
| 30 |
"learning_rate": 1.94e-06,
|
| 31 |
"loss": 0.7765,
|
| 32 |
"step": 100
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch": 0.
|
| 36 |
"learning_rate": 2.4400000000000004e-06,
|
| 37 |
"loss": 0.7103,
|
| 38 |
"step": 125
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch": 0.
|
| 42 |
"learning_rate": 2.9400000000000002e-06,
|
| 43 |
"loss": 0.6597,
|
| 44 |
"step": 150
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
"learning_rate": 3.44e-06,
|
| 49 |
"loss": 0.6657,
|
| 50 |
"step": 175
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"epoch": 0.
|
| 54 |
"learning_rate": 3.94e-06,
|
| 55 |
"loss": 0.5853,
|
| 56 |
"step": 200
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch": 0.
|
| 60 |
"learning_rate": 4.440000000000001e-06,
|
| 61 |
"loss": 0.5273,
|
| 62 |
"step": 225
|
| 63 |
},
|
| 64 |
{
|
| 65 |
-
"epoch": 0.
|
| 66 |
"learning_rate": 4.94e-06,
|
| 67 |
"loss": 0.5979,
|
| 68 |
"step": 250
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch": 0.
|
| 72 |
"learning_rate": 5.4400000000000004e-06,
|
| 73 |
"loss": 0.5861,
|
| 74 |
"step": 275
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch": 0.
|
| 78 |
"learning_rate": 5.94e-06,
|
| 79 |
"loss": 0.5085,
|
| 80 |
"step": 300
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch": 0.
|
| 84 |
"learning_rate": 6.440000000000001e-06,
|
| 85 |
"loss": 0.4827,
|
| 86 |
"step": 325
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
"learning_rate": 6.9400000000000005e-06,
|
| 91 |
"loss": 0.4909,
|
| 92 |
"step": 350
|
| 93 |
},
|
| 94 |
{
|
| 95 |
-
"epoch": 0.
|
| 96 |
"learning_rate": 7.440000000000001e-06,
|
| 97 |
"loss": 0.4651,
|
| 98 |
"step": 375
|
| 99 |
},
|
| 100 |
{
|
| 101 |
-
"epoch": 0.
|
| 102 |
"learning_rate": 7.94e-06,
|
| 103 |
"loss": 0.494,
|
| 104 |
"step": 400
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch": 0.
|
| 108 |
"learning_rate": 8.44e-06,
|
| 109 |
"loss": 0.4188,
|
| 110 |
"step": 425
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"epoch": 0.
|
| 114 |
"learning_rate": 8.94e-06,
|
| 115 |
"loss": 0.3849,
|
| 116 |
"step": 450
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch": 0.
|
| 120 |
"learning_rate": 9.440000000000001e-06,
|
| 121 |
"loss": 0.4577,
|
| 122 |
"step": 475
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch": 0.
|
| 126 |
"learning_rate": 9.940000000000001e-06,
|
| 127 |
"loss": 0.4415,
|
| 128 |
"step": 500
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"learning_rate": 9.
|
| 133 |
-
"loss": 0.
|
| 134 |
"step": 525
|
| 135 |
},
|
| 136 |
{
|
| 137 |
-
"epoch": 0.
|
| 138 |
-
"learning_rate": 9.
|
| 139 |
-
"loss": 0.
|
| 140 |
"step": 550
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"epoch": 0.
|
| 144 |
-
"learning_rate": 9.
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 575
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch": 0.
|
| 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": 0.
|
| 168 |
-
"learning_rate": 9.
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 675
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"learning_rate": 9.
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 700
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch": 0.
|
| 180 |
-
"learning_rate": 9.
|
| 181 |
-
"loss": 0.
|
| 182 |
"step": 725
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch": 0.
|
| 186 |
-
"learning_rate": 9.
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 750
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch": 0.
|
| 192 |
-
"learning_rate": 9.
|
| 193 |
-
"loss": 0.
|
| 194 |
"step": 775
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch": 0.
|
| 198 |
-
"learning_rate": 9.
|
| 199 |
-
"loss": 0.
|
| 200 |
"step": 800
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch": 0.
|
| 204 |
-
"learning_rate": 9.
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 825
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch": 0.
|
| 210 |
-
"learning_rate": 9.
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 850
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"learning_rate": 9.
|
| 217 |
-
"loss": 0.
|
| 218 |
"step": 875
|
| 219 |
},
|
| 220 |
{
|
| 221 |
-
"epoch": 0.
|
| 222 |
-
"learning_rate": 9.
|
| 223 |
-
"loss": 0.
|
| 224 |
"step": 900
|
| 225 |
},
|
| 226 |
{
|
| 227 |
-
"epoch": 0.
|
| 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": 0.
|
| 240 |
-
"learning_rate":
|
| 241 |
-
"loss": 0.
|
| 242 |
"step": 975
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch": 0.
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 1000
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch": 0.
|
| 252 |
-
"eval_loss": 0.
|
| 253 |
-
"eval_runtime":
|
| 254 |
-
"eval_samples_per_second": 3.
|
| 255 |
-
"eval_steps_per_second": 0.
|
| 256 |
-
"eval_wer": 28.
|
| 257 |
"step": 1000
|
| 258 |
},
|
| 259 |
{
|
| 260 |
-
"epoch": 0.
|
| 261 |
-
"learning_rate":
|
| 262 |
-
"loss": 0.
|
| 263 |
"step": 1025
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch": 0.
|
| 267 |
-
"learning_rate":
|
| 268 |
"loss": 0.3674,
|
| 269 |
"step": 1050
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch": 0.
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 1075
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 0.
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 1100
|
| 282 |
},
|
| 283 |
{
|
| 284 |
-
"epoch": 0.
|
| 285 |
-
"learning_rate":
|
| 286 |
-
"loss": 0.
|
| 287 |
"step": 1125
|
| 288 |
},
|
| 289 |
{
|
| 290 |
-
"epoch": 0.
|
| 291 |
-
"learning_rate":
|
| 292 |
-
"loss": 0.
|
| 293 |
"step": 1150
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"epoch": 0.
|
| 297 |
-
"learning_rate": 8.
|
| 298 |
-
"loss": 0.
|
| 299 |
"step": 1175
|
| 300 |
},
|
| 301 |
{
|
| 302 |
-
"epoch": 0.
|
| 303 |
-
"learning_rate": 8.
|
| 304 |
-
"loss": 0.
|
| 305 |
"step": 1200
|
| 306 |
},
|
| 307 |
{
|
| 308 |
-
"epoch": 0.
|
| 309 |
-
"learning_rate": 8.
|
| 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": 0.
|
| 327 |
-
"learning_rate": 8.
|
| 328 |
-
"loss": 0.
|
| 329 |
"step": 1300
|
| 330 |
},
|
| 331 |
{
|
| 332 |
-
"epoch": 0.
|
| 333 |
-
"learning_rate": 8.
|
| 334 |
-
"loss": 0.
|
| 335 |
"step": 1325
|
| 336 |
},
|
| 337 |
{
|
| 338 |
-
"epoch": 0.
|
| 339 |
-
"learning_rate": 8.
|
| 340 |
-
"loss": 0.
|
| 341 |
"step": 1350
|
| 342 |
},
|
| 343 |
{
|
| 344 |
-
"epoch": 0.
|
| 345 |
-
"learning_rate": 8.
|
| 346 |
-
"loss": 0.
|
| 347 |
"step": 1375
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch": 0.
|
| 351 |
-
"learning_rate": 8.
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 1400
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch": 0.
|
| 357 |
-
"learning_rate":
|
| 358 |
-
"loss": 0.
|
| 359 |
"step": 1425
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 0.
|
| 363 |
-
"learning_rate":
|
| 364 |
-
"loss": 0.
|
| 365 |
"step": 1450
|
| 366 |
},
|
| 367 |
{
|
| 368 |
-
"epoch": 0.
|
| 369 |
-
"learning_rate":
|
| 370 |
-
"loss": 0.
|
| 371 |
"step": 1475
|
| 372 |
},
|
| 373 |
{
|
| 374 |
-
"epoch": 0.
|
| 375 |
-
"learning_rate":
|
| 376 |
-
"loss": 0.
|
| 377 |
"step": 1500
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch": 0.
|
| 381 |
-
"learning_rate":
|
| 382 |
-
"loss": 0.
|
| 383 |
"step": 1525
|
| 384 |
},
|
| 385 |
{
|
| 386 |
-
"epoch": 0.
|
| 387 |
-
"learning_rate":
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 1550
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch": 0.
|
| 393 |
-
"learning_rate":
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 1575
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch": 0.
|
| 399 |
-
"learning_rate":
|
| 400 |
-
"loss": 0.
|
| 401 |
"step": 1600
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 0.
|
| 405 |
-
"learning_rate":
|
| 406 |
-
"loss": 0.
|
| 407 |
"step": 1625
|
| 408 |
},
|
| 409 |
{
|
| 410 |
-
"epoch": 0.
|
| 411 |
-
"learning_rate":
|
| 412 |
-
"loss": 0.
|
| 413 |
"step": 1650
|
| 414 |
},
|
| 415 |
{
|
| 416 |
-
"epoch": 0.
|
| 417 |
-
"learning_rate":
|
| 418 |
-
"loss": 0.
|
| 419 |
"step": 1675
|
| 420 |
},
|
| 421 |
{
|
| 422 |
-
"epoch": 0.
|
| 423 |
-
"learning_rate":
|
| 424 |
-
"loss": 0.
|
| 425 |
"step": 1700
|
| 426 |
},
|
| 427 |
{
|
| 428 |
-
"epoch": 0.
|
| 429 |
-
"learning_rate":
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 1725
|
| 432 |
},
|
| 433 |
{
|
| 434 |
-
"epoch": 0.
|
| 435 |
-
"learning_rate":
|
| 436 |
-
"loss": 0.
|
| 437 |
"step": 1750
|
| 438 |
},
|
| 439 |
{
|
| 440 |
-
"epoch": 0.
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss": 0.
|
| 443 |
"step": 1775
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 0.
|
| 447 |
-
"learning_rate":
|
| 448 |
-
"loss": 0.
|
| 449 |
"step": 1800
|
| 450 |
},
|
| 451 |
{
|
| 452 |
-
"epoch": 0.
|
| 453 |
-
"learning_rate": 7.
|
| 454 |
-
"loss": 0.
|
| 455 |
"step": 1825
|
| 456 |
},
|
| 457 |
{
|
| 458 |
-
"epoch": 0.
|
| 459 |
-
"learning_rate": 7.
|
| 460 |
-
"loss": 0.
|
| 461 |
"step": 1850
|
| 462 |
},
|
| 463 |
{
|
| 464 |
-
"epoch": 0.
|
| 465 |
-
"learning_rate":
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 1875
|
| 468 |
},
|
| 469 |
{
|
| 470 |
-
"epoch": 0.
|
| 471 |
-
"learning_rate":
|
| 472 |
-
"loss": 0.
|
| 473 |
"step": 1900
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"epoch": 0.
|
| 477 |
-
"learning_rate":
|
| 478 |
-
"loss": 0.
|
| 479 |
"step": 1925
|
| 480 |
},
|
| 481 |
{
|
| 482 |
-
"epoch": 0.
|
| 483 |
-
"learning_rate":
|
| 484 |
-
"loss": 0.
|
| 485 |
"step": 1950
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 0.
|
| 489 |
-
"learning_rate":
|
| 490 |
-
"loss": 0.
|
| 491 |
"step": 1975
|
| 492 |
},
|
| 493 |
{
|
| 494 |
-
"epoch": 0.
|
| 495 |
-
"learning_rate":
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 2000
|
| 498 |
},
|
| 499 |
{
|
| 500 |
-
"epoch": 0.
|
| 501 |
-
"eval_loss": 0.
|
| 502 |
-
"eval_runtime":
|
| 503 |
-
"eval_samples_per_second": 3.
|
| 504 |
-
"eval_steps_per_second": 0.
|
| 505 |
-
"eval_wer":
|
| 506 |
"step": 2000
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch": 0.
|
| 510 |
-
"learning_rate":
|
| 511 |
-
"loss": 0.
|
| 512 |
"step": 2025
|
| 513 |
},
|
| 514 |
{
|
| 515 |
-
"epoch": 0.
|
| 516 |
-
"learning_rate":
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 2050
|
| 519 |
},
|
| 520 |
{
|
| 521 |
-
"epoch": 0.
|
| 522 |
-
"learning_rate":
|
| 523 |
-
"loss": 0.
|
| 524 |
"step": 2075
|
| 525 |
},
|
| 526 |
{
|
| 527 |
-
"epoch": 0.
|
| 528 |
-
"learning_rate":
|
| 529 |
-
"loss": 0.
|
| 530 |
"step": 2100
|
| 531 |
},
|
| 532 |
{
|
| 533 |
-
"epoch": 0.
|
| 534 |
-
"learning_rate":
|
| 535 |
-
"loss": 0.
|
| 536 |
"step": 2125
|
| 537 |
},
|
| 538 |
{
|
| 539 |
-
"epoch": 0.
|
| 540 |
-
"learning_rate":
|
| 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": 0.
|
| 558 |
-
"learning_rate":
|
| 559 |
-
"loss": 0.
|
| 560 |
"step": 2225
|
| 561 |
},
|
| 562 |
{
|
| 563 |
-
"epoch": 0.
|
| 564 |
-
"learning_rate":
|
| 565 |
-
"loss": 0.
|
| 566 |
"step": 2250
|
| 567 |
},
|
| 568 |
{
|
| 569 |
-
"epoch": 0.
|
| 570 |
-
"learning_rate":
|
| 571 |
-
"loss": 0.
|
| 572 |
"step": 2275
|
| 573 |
},
|
| 574 |
{
|
| 575 |
-
"epoch": 0.
|
| 576 |
-
"learning_rate":
|
| 577 |
-
"loss": 0.
|
| 578 |
"step": 2300
|
| 579 |
},
|
| 580 |
{
|
| 581 |
-
"epoch": 0.
|
| 582 |
-
"learning_rate":
|
| 583 |
-
"loss": 0.
|
| 584 |
"step": 2325
|
| 585 |
},
|
| 586 |
{
|
| 587 |
-
"epoch": 0.
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 2350
|
| 591 |
},
|
| 592 |
{
|
| 593 |
-
"epoch": 0.
|
| 594 |
-
"learning_rate":
|
| 595 |
-
"loss": 0.
|
| 596 |
"step": 2375
|
| 597 |
},
|
| 598 |
{
|
| 599 |
-
"epoch": 0.
|
| 600 |
-
"learning_rate":
|
| 601 |
-
"loss": 0.
|
| 602 |
"step": 2400
|
| 603 |
},
|
| 604 |
{
|
| 605 |
-
"epoch": 0.
|
| 606 |
-
"learning_rate":
|
| 607 |
-
"loss": 0.
|
| 608 |
"step": 2425
|
| 609 |
},
|
| 610 |
{
|
| 611 |
-
"epoch": 0.
|
| 612 |
-
"learning_rate":
|
| 613 |
-
"loss": 0.
|
| 614 |
"step": 2450
|
| 615 |
},
|
| 616 |
{
|
| 617 |
-
"epoch": 0.
|
| 618 |
-
"learning_rate":
|
| 619 |
-
"loss": 0.
|
| 620 |
"step": 2475
|
| 621 |
},
|
| 622 |
{
|
| 623 |
-
"epoch": 0.
|
| 624 |
-
"learning_rate":
|
| 625 |
-
"loss": 0.
|
| 626 |
"step": 2500
|
| 627 |
},
|
| 628 |
{
|
| 629 |
-
"epoch": 0.
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 2525
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch": 0.
|
| 636 |
-
"learning_rate":
|
| 637 |
-
"loss": 0.
|
| 638 |
"step": 2550
|
| 639 |
},
|
| 640 |
{
|
| 641 |
-
"epoch": 0.
|
| 642 |
-
"learning_rate":
|
| 643 |
-
"loss": 0.
|
| 644 |
"step": 2575
|
| 645 |
},
|
| 646 |
{
|
| 647 |
-
"epoch": 0.
|
| 648 |
-
"learning_rate":
|
| 649 |
-
"loss": 0.
|
| 650 |
"step": 2600
|
| 651 |
},
|
| 652 |
{
|
| 653 |
-
"epoch": 0.
|
| 654 |
-
"learning_rate":
|
| 655 |
-
"loss": 0.
|
| 656 |
"step": 2625
|
| 657 |
},
|
| 658 |
{
|
| 659 |
-
"epoch": 0.
|
| 660 |
-
"learning_rate":
|
| 661 |
-
"loss": 0.
|
| 662 |
"step": 2650
|
| 663 |
},
|
| 664 |
{
|
| 665 |
-
"epoch": 0.
|
| 666 |
-
"learning_rate":
|
| 667 |
-
"loss": 0.
|
| 668 |
"step": 2675
|
| 669 |
},
|
| 670 |
{
|
| 671 |
-
"epoch": 0.
|
| 672 |
-
"learning_rate":
|
| 673 |
-
"loss": 0.
|
| 674 |
"step": 2700
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch": 0.
|
| 678 |
-
"learning_rate":
|
| 679 |
-
"loss": 0.
|
| 680 |
"step": 2725
|
| 681 |
},
|
| 682 |
{
|
| 683 |
-
"epoch": 0.
|
| 684 |
-
"learning_rate":
|
| 685 |
-
"loss": 0.
|
| 686 |
"step": 2750
|
| 687 |
},
|
| 688 |
{
|
| 689 |
-
"epoch": 0.
|
| 690 |
-
"learning_rate":
|
| 691 |
-
"loss": 0.
|
| 692 |
"step": 2775
|
| 693 |
},
|
| 694 |
{
|
| 695 |
-
"epoch": 0.
|
| 696 |
-
"learning_rate":
|
| 697 |
-
"loss": 0.
|
| 698 |
"step": 2800
|
| 699 |
},
|
| 700 |
{
|
| 701 |
-
"epoch": 0.
|
| 702 |
-
"learning_rate":
|
| 703 |
-
"loss": 0.
|
| 704 |
"step": 2825
|
| 705 |
},
|
| 706 |
{
|
| 707 |
-
"epoch": 0.
|
| 708 |
-
"learning_rate":
|
| 709 |
-
"loss": 0.
|
| 710 |
"step": 2850
|
| 711 |
},
|
| 712 |
{
|
| 713 |
-
"epoch": 0.
|
| 714 |
-
"learning_rate":
|
| 715 |
-
"loss": 0.
|
| 716 |
"step": 2875
|
| 717 |
},
|
| 718 |
{
|
| 719 |
-
"epoch": 0.
|
| 720 |
-
"learning_rate":
|
| 721 |
-
"loss": 0.
|
| 722 |
"step": 2900
|
| 723 |
},
|
| 724 |
{
|
| 725 |
-
"epoch": 0.
|
| 726 |
-
"learning_rate":
|
| 727 |
-
"loss": 0.
|
| 728 |
"step": 2925
|
| 729 |
},
|
| 730 |
{
|
| 731 |
-
"epoch": 0.
|
| 732 |
-
"learning_rate":
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 2950
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"epoch": 0.
|
| 738 |
-
"learning_rate":
|
| 739 |
-
"loss": 0.
|
| 740 |
"step": 2975
|
| 741 |
},
|
| 742 |
{
|
| 743 |
-
"epoch": 0.
|
| 744 |
-
"learning_rate":
|
| 745 |
-
"loss": 0.
|
| 746 |
"step": 3000
|
| 747 |
},
|
| 748 |
{
|
| 749 |
-
"epoch": 0.
|
| 750 |
-
"eval_loss": 0.
|
| 751 |
-
"eval_runtime":
|
| 752 |
-
"eval_samples_per_second": 3.
|
| 753 |
-
"eval_steps_per_second": 0.
|
| 754 |
-
"eval_wer":
|
| 755 |
"step": 3000
|
| 756 |
},
|
| 757 |
{
|
| 758 |
-
"epoch": 0.
|
| 759 |
-
"learning_rate":
|
| 760 |
-
"loss": 0.
|
| 761 |
"step": 3025
|
| 762 |
},
|
| 763 |
{
|
| 764 |
-
"epoch": 0.
|
| 765 |
-
"learning_rate":
|
| 766 |
-
"loss": 0.
|
| 767 |
"step": 3050
|
| 768 |
},
|
| 769 |
{
|
| 770 |
-
"epoch": 0.
|
| 771 |
-
"learning_rate":
|
| 772 |
-
"loss": 0.
|
| 773 |
"step": 3075
|
| 774 |
},
|
| 775 |
{
|
| 776 |
-
"epoch": 0.
|
| 777 |
-
"learning_rate":
|
| 778 |
-
"loss": 0.
|
| 779 |
"step": 3100
|
| 780 |
},
|
| 781 |
{
|
| 782 |
-
"epoch": 0.
|
| 783 |
-
"learning_rate":
|
| 784 |
-
"loss": 0.
|
| 785 |
"step": 3125
|
| 786 |
},
|
| 787 |
{
|
| 788 |
-
"epoch": 0.
|
| 789 |
-
"learning_rate":
|
| 790 |
-
"loss": 0.
|
| 791 |
"step": 3150
|
| 792 |
},
|
| 793 |
{
|
| 794 |
-
"epoch": 0.
|
| 795 |
-
"learning_rate":
|
| 796 |
-
"loss": 0.
|
| 797 |
"step": 3175
|
| 798 |
},
|
| 799 |
{
|
| 800 |
-
"epoch": 0.
|
| 801 |
-
"learning_rate":
|
| 802 |
-
"loss": 0.
|
| 803 |
"step": 3200
|
| 804 |
},
|
| 805 |
{
|
| 806 |
-
"epoch": 0.
|
| 807 |
-
"learning_rate":
|
| 808 |
-
"loss": 0.
|
| 809 |
"step": 3225
|
| 810 |
},
|
| 811 |
{
|
| 812 |
-
"epoch": 0.
|
| 813 |
-
"learning_rate":
|
| 814 |
-
"loss": 0.
|
| 815 |
"step": 3250
|
| 816 |
},
|
| 817 |
{
|
| 818 |
-
"epoch": 0.
|
| 819 |
-
"learning_rate":
|
| 820 |
-
"loss": 0.
|
| 821 |
"step": 3275
|
| 822 |
},
|
| 823 |
{
|
| 824 |
"epoch": 1.0,
|
| 825 |
-
"learning_rate":
|
| 826 |
-
"loss": 0.
|
| 827 |
"step": 3300
|
| 828 |
},
|
| 829 |
{
|
| 830 |
-
"epoch": 1.
|
| 831 |
-
"learning_rate":
|
| 832 |
-
"loss": 0.
|
| 833 |
"step": 3325
|
| 834 |
},
|
| 835 |
{
|
| 836 |
"epoch": 1.01,
|
| 837 |
-
"learning_rate":
|
| 838 |
-
"loss": 0.
|
| 839 |
"step": 3350
|
| 840 |
},
|
| 841 |
{
|
| 842 |
-
"epoch": 1.
|
| 843 |
-
"learning_rate":
|
| 844 |
-
"loss": 0.
|
| 845 |
"step": 3375
|
| 846 |
},
|
| 847 |
{
|
| 848 |
-
"epoch": 1.
|
| 849 |
-
"learning_rate":
|
| 850 |
-
"loss": 0.
|
| 851 |
"step": 3400
|
| 852 |
},
|
| 853 |
{
|
| 854 |
-
"epoch": 1.
|
| 855 |
-
"learning_rate":
|
| 856 |
-
"loss": 0.
|
| 857 |
"step": 3425
|
| 858 |
},
|
| 859 |
{
|
| 860 |
-
"epoch": 1.
|
| 861 |
-
"learning_rate":
|
| 862 |
-
"loss": 0.
|
| 863 |
"step": 3450
|
| 864 |
},
|
| 865 |
{
|
| 866 |
-
"epoch": 1.
|
| 867 |
-
"learning_rate":
|
| 868 |
-
"loss": 0.
|
| 869 |
"step": 3475
|
| 870 |
},
|
| 871 |
{
|
| 872 |
-
"epoch": 1.
|
| 873 |
-
"learning_rate":
|
| 874 |
-
"loss": 0.
|
| 875 |
"step": 3500
|
| 876 |
},
|
| 877 |
{
|
| 878 |
-
"epoch": 1.
|
| 879 |
-
"learning_rate":
|
| 880 |
-
"loss": 0.
|
| 881 |
"step": 3525
|
| 882 |
},
|
| 883 |
{
|
| 884 |
-
"epoch": 1.
|
| 885 |
-
"learning_rate":
|
| 886 |
-
"loss": 0.
|
| 887 |
"step": 3550
|
| 888 |
},
|
| 889 |
{
|
| 890 |
-
"epoch": 1.
|
| 891 |
-
"learning_rate":
|
| 892 |
-
"loss": 0.
|
| 893 |
"step": 3575
|
| 894 |
},
|
| 895 |
{
|
| 896 |
-
"epoch": 1.
|
| 897 |
-
"learning_rate":
|
| 898 |
-
"loss": 0.
|
| 899 |
"step": 3600
|
| 900 |
},
|
| 901 |
{
|
| 902 |
-
"epoch": 1.
|
| 903 |
-
"learning_rate":
|
| 904 |
-
"loss": 0.
|
| 905 |
"step": 3625
|
| 906 |
},
|
| 907 |
{
|
| 908 |
-
"epoch": 1.
|
| 909 |
-
"learning_rate":
|
| 910 |
-
"loss": 0.
|
| 911 |
"step": 3650
|
| 912 |
},
|
| 913 |
{
|
| 914 |
-
"epoch": 1.
|
| 915 |
-
"learning_rate":
|
| 916 |
-
"loss": 0.
|
| 917 |
"step": 3675
|
| 918 |
},
|
| 919 |
{
|
| 920 |
-
"epoch": 1.
|
| 921 |
-
"learning_rate":
|
| 922 |
-
"loss": 0.
|
| 923 |
"step": 3700
|
| 924 |
},
|
| 925 |
{
|
| 926 |
-
"epoch": 1.
|
| 927 |
-
"learning_rate":
|
| 928 |
-
"loss": 0.
|
| 929 |
"step": 3725
|
| 930 |
},
|
| 931 |
{
|
| 932 |
-
"epoch": 1.
|
| 933 |
-
"learning_rate":
|
| 934 |
-
"loss": 0.
|
| 935 |
"step": 3750
|
| 936 |
},
|
| 937 |
{
|
| 938 |
-
"epoch": 1.
|
| 939 |
-
"learning_rate":
|
| 940 |
-
"loss": 0.
|
| 941 |
"step": 3775
|
| 942 |
},
|
| 943 |
{
|
| 944 |
-
"epoch": 1.
|
| 945 |
-
"learning_rate":
|
| 946 |
-
"loss": 0.
|
| 947 |
"step": 3800
|
| 948 |
},
|
| 949 |
{
|
| 950 |
-
"epoch": 1.
|
| 951 |
-
"learning_rate":
|
| 952 |
-
"loss": 0.
|
| 953 |
"step": 3825
|
| 954 |
},
|
| 955 |
{
|
| 956 |
-
"epoch": 1.
|
| 957 |
-
"learning_rate":
|
| 958 |
-
"loss": 0.
|
| 959 |
"step": 3850
|
| 960 |
},
|
| 961 |
{
|
| 962 |
-
"epoch": 1.
|
| 963 |
-
"learning_rate":
|
| 964 |
-
"loss": 0.
|
| 965 |
"step": 3875
|
| 966 |
},
|
| 967 |
{
|
| 968 |
-
"epoch": 1.
|
| 969 |
-
"learning_rate":
|
| 970 |
-
"loss": 0.
|
| 971 |
"step": 3900
|
| 972 |
},
|
| 973 |
{
|
| 974 |
-
"epoch": 1.
|
| 975 |
-
"learning_rate":
|
| 976 |
-
"loss": 0.
|
| 977 |
"step": 3925
|
| 978 |
},
|
| 979 |
{
|
| 980 |
-
"epoch": 1.
|
| 981 |
-
"learning_rate":
|
| 982 |
-
"loss": 0.
|
| 983 |
"step": 3950
|
| 984 |
},
|
| 985 |
{
|
| 986 |
-
"epoch": 1.
|
| 987 |
-
"learning_rate":
|
| 988 |
-
"loss": 0.
|
| 989 |
"step": 3975
|
| 990 |
},
|
| 991 |
{
|
| 992 |
-
"epoch": 1.
|
| 993 |
-
"learning_rate":
|
| 994 |
-
"loss": 0.
|
| 995 |
"step": 4000
|
| 996 |
},
|
| 997 |
{
|
| 998 |
-
"epoch": 1.
|
| 999 |
-
"eval_loss": 0.
|
| 1000 |
-
"eval_runtime":
|
| 1001 |
-
"eval_samples_per_second": 3.
|
| 1002 |
-
"eval_steps_per_second": 0.
|
| 1003 |
-
"eval_wer":
|
| 1004 |
"step": 4000
|
| 1005 |
}
|
| 1006 |
],
|
| 1007 |
-
"max_steps":
|
| 1008 |
"num_train_epochs": 9223372036854775807,
|
| 1009 |
"total_flos": 1.632763731050496e+19,
|
| 1010 |
"trial_name": null,
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 16.353436753209934,
|
| 3 |
"best_model_checkpoint": "./checkpoint-4000",
|
| 4 |
+
"epoch": 1.1005714285714285,
|
| 5 |
"global_step": 4000,
|
| 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": 4.6000000000000004e-07,
|
| 13 |
"loss": 1.4182,
|
| 14 |
"step": 25
|
|
|
|
| 26 |
"step": 75
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.01,
|
| 30 |
"learning_rate": 1.94e-06,
|
| 31 |
"loss": 0.7765,
|
| 32 |
"step": 100
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 0.02,
|
| 36 |
"learning_rate": 2.4400000000000004e-06,
|
| 37 |
"loss": 0.7103,
|
| 38 |
"step": 125
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.02,
|
| 42 |
"learning_rate": 2.9400000000000002e-06,
|
| 43 |
"loss": 0.6597,
|
| 44 |
"step": 150
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.03,
|
| 48 |
"learning_rate": 3.44e-06,
|
| 49 |
"loss": 0.6657,
|
| 50 |
"step": 175
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 0.03,
|
| 54 |
"learning_rate": 3.94e-06,
|
| 55 |
"loss": 0.5853,
|
| 56 |
"step": 200
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 0.03,
|
| 60 |
"learning_rate": 4.440000000000001e-06,
|
| 61 |
"loss": 0.5273,
|
| 62 |
"step": 225
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 0.04,
|
| 66 |
"learning_rate": 4.94e-06,
|
| 67 |
"loss": 0.5979,
|
| 68 |
"step": 250
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 0.04,
|
| 72 |
"learning_rate": 5.4400000000000004e-06,
|
| 73 |
"loss": 0.5861,
|
| 74 |
"step": 275
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 0.04,
|
| 78 |
"learning_rate": 5.94e-06,
|
| 79 |
"loss": 0.5085,
|
| 80 |
"step": 300
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 0.05,
|
| 84 |
"learning_rate": 6.440000000000001e-06,
|
| 85 |
"loss": 0.4827,
|
| 86 |
"step": 325
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.05,
|
| 90 |
"learning_rate": 6.9400000000000005e-06,
|
| 91 |
"loss": 0.4909,
|
| 92 |
"step": 350
|
| 93 |
},
|
| 94 |
{
|
| 95 |
+
"epoch": 0.05,
|
| 96 |
"learning_rate": 7.440000000000001e-06,
|
| 97 |
"loss": 0.4651,
|
| 98 |
"step": 375
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 0.06,
|
| 102 |
"learning_rate": 7.94e-06,
|
| 103 |
"loss": 0.494,
|
| 104 |
"step": 400
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 0.06,
|
| 108 |
"learning_rate": 8.44e-06,
|
| 109 |
"loss": 0.4188,
|
| 110 |
"step": 425
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"epoch": 0.06,
|
| 114 |
"learning_rate": 8.94e-06,
|
| 115 |
"loss": 0.3849,
|
| 116 |
"step": 450
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 0.07,
|
| 120 |
"learning_rate": 9.440000000000001e-06,
|
| 121 |
"loss": 0.4577,
|
| 122 |
"step": 475
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 0.07,
|
| 126 |
"learning_rate": 9.940000000000001e-06,
|
| 127 |
"loss": 0.4415,
|
| 128 |
"step": 500
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.07,
|
| 132 |
+
"learning_rate": 9.966153846153847e-06,
|
| 133 |
+
"loss": 0.4614,
|
| 134 |
"step": 525
|
| 135 |
},
|
| 136 |
{
|
| 137 |
+
"epoch": 0.08,
|
| 138 |
+
"learning_rate": 9.927692307692309e-06,
|
| 139 |
+
"loss": 0.4283,
|
| 140 |
"step": 550
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"epoch": 0.08,
|
| 144 |
+
"learning_rate": 9.88923076923077e-06,
|
| 145 |
+
"loss": 0.4486,
|
| 146 |
"step": 575
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 0.09,
|
| 150 |
+
"learning_rate": 9.850769230769231e-06,
|
| 151 |
+
"loss": 0.4434,
|
| 152 |
"step": 600
|
| 153 |
},
|
| 154 |
{
|
| 155 |
+
"epoch": 0.09,
|
| 156 |
+
"learning_rate": 9.812307692307694e-06,
|
| 157 |
+
"loss": 0.4245,
|
| 158 |
"step": 625
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 0.09,
|
| 162 |
+
"learning_rate": 9.773846153846154e-06,
|
| 163 |
+
"loss": 0.4254,
|
| 164 |
"step": 650
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 0.1,
|
| 168 |
+
"learning_rate": 9.735384615384616e-06,
|
| 169 |
+
"loss": 0.4143,
|
| 170 |
"step": 675
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.1,
|
| 174 |
+
"learning_rate": 9.696923076923078e-06,
|
| 175 |
+
"loss": 0.4115,
|
| 176 |
"step": 700
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 0.1,
|
| 180 |
+
"learning_rate": 9.658461538461539e-06,
|
| 181 |
+
"loss": 0.3656,
|
| 182 |
"step": 725
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"epoch": 0.11,
|
| 186 |
+
"learning_rate": 9.620000000000001e-06,
|
| 187 |
+
"loss": 0.4004,
|
| 188 |
"step": 750
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 0.11,
|
| 192 |
+
"learning_rate": 9.581538461538462e-06,
|
| 193 |
+
"loss": 0.388,
|
| 194 |
"step": 775
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 0.11,
|
| 198 |
+
"learning_rate": 9.543076923076924e-06,
|
| 199 |
+
"loss": 0.3665,
|
| 200 |
"step": 800
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 0.12,
|
| 204 |
+
"learning_rate": 9.504615384615386e-06,
|
| 205 |
+
"loss": 0.3777,
|
| 206 |
"step": 825
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 0.12,
|
| 210 |
+
"learning_rate": 9.466153846153846e-06,
|
| 211 |
+
"loss": 0.4338,
|
| 212 |
"step": 850
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.12,
|
| 216 |
+
"learning_rate": 9.427692307692309e-06,
|
| 217 |
+
"loss": 0.4017,
|
| 218 |
"step": 875
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 0.13,
|
| 222 |
+
"learning_rate": 9.38923076923077e-06,
|
| 223 |
+
"loss": 0.3757,
|
| 224 |
"step": 900
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 0.13,
|
| 228 |
+
"learning_rate": 9.350769230769231e-06,
|
| 229 |
+
"loss": 0.4084,
|
| 230 |
"step": 925
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 0.14,
|
| 234 |
+
"learning_rate": 9.312307692307693e-06,
|
| 235 |
+
"loss": 0.3821,
|
| 236 |
"step": 950
|
| 237 |
},
|
| 238 |
{
|
| 239 |
+
"epoch": 0.14,
|
| 240 |
+
"learning_rate": 9.273846153846154e-06,
|
| 241 |
+
"loss": 0.421,
|
| 242 |
"step": 975
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 0.14,
|
| 246 |
+
"learning_rate": 9.235384615384616e-06,
|
| 247 |
+
"loss": 0.4203,
|
| 248 |
"step": 1000
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 0.14,
|
| 252 |
+
"eval_loss": 0.4127572178840637,
|
| 253 |
+
"eval_runtime": 1802.5775,
|
| 254 |
+
"eval_samples_per_second": 3.656,
|
| 255 |
+
"eval_steps_per_second": 0.457,
|
| 256 |
+
"eval_wer": 28.265624367127064,
|
| 257 |
"step": 1000
|
| 258 |
},
|
| 259 |
{
|
| 260 |
+
"epoch": 0.15,
|
| 261 |
+
"learning_rate": 9.196923076923078e-06,
|
| 262 |
+
"loss": 0.4094,
|
| 263 |
"step": 1025
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"epoch": 0.15,
|
| 267 |
+
"learning_rate": 9.158461538461539e-06,
|
| 268 |
"loss": 0.3674,
|
| 269 |
"step": 1050
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"epoch": 0.15,
|
| 273 |
+
"learning_rate": 9.12e-06,
|
| 274 |
+
"loss": 0.3613,
|
| 275 |
"step": 1075
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.16,
|
| 279 |
+
"learning_rate": 9.081538461538463e-06,
|
| 280 |
+
"loss": 0.3904,
|
| 281 |
"step": 1100
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 0.16,
|
| 285 |
+
"learning_rate": 9.043076923076923e-06,
|
| 286 |
+
"loss": 0.3865,
|
| 287 |
"step": 1125
|
| 288 |
},
|
| 289 |
{
|
| 290 |
+
"epoch": 0.16,
|
| 291 |
+
"learning_rate": 9.004615384615386e-06,
|
| 292 |
+
"loss": 0.3677,
|
| 293 |
"step": 1150
|
| 294 |
},
|
| 295 |
{
|
| 296 |
+
"epoch": 0.17,
|
| 297 |
+
"learning_rate": 8.966153846153846e-06,
|
| 298 |
+
"loss": 0.3295,
|
| 299 |
"step": 1175
|
| 300 |
},
|
| 301 |
{
|
| 302 |
+
"epoch": 0.17,
|
| 303 |
+
"learning_rate": 8.927692307692308e-06,
|
| 304 |
+
"loss": 0.3207,
|
| 305 |
"step": 1200
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"epoch": 0.17,
|
| 309 |
+
"learning_rate": 8.88923076923077e-06,
|
| 310 |
+
"loss": 0.2817,
|
| 311 |
"step": 1225
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"epoch": 0.18,
|
| 315 |
+
"learning_rate": 8.850769230769231e-06,
|
| 316 |
+
"loss": 0.3218,
|
| 317 |
"step": 1250
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.18,
|
| 321 |
+
"learning_rate": 8.812307692307693e-06,
|
| 322 |
+
"loss": 0.3502,
|
| 323 |
"step": 1275
|
| 324 |
},
|
| 325 |
{
|
| 326 |
+
"epoch": 0.19,
|
| 327 |
+
"learning_rate": 8.773846153846155e-06,
|
| 328 |
+
"loss": 0.3432,
|
| 329 |
"step": 1300
|
| 330 |
},
|
| 331 |
{
|
| 332 |
+
"epoch": 0.19,
|
| 333 |
+
"learning_rate": 8.735384615384616e-06,
|
| 334 |
+
"loss": 0.3244,
|
| 335 |
"step": 1325
|
| 336 |
},
|
| 337 |
{
|
| 338 |
+
"epoch": 0.19,
|
| 339 |
+
"learning_rate": 8.696923076923078e-06,
|
| 340 |
+
"loss": 0.2607,
|
| 341 |
"step": 1350
|
| 342 |
},
|
| 343 |
{
|
| 344 |
+
"epoch": 0.2,
|
| 345 |
+
"learning_rate": 8.658461538461538e-06,
|
| 346 |
+
"loss": 0.3469,
|
| 347 |
"step": 1375
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"epoch": 0.2,
|
| 351 |
+
"learning_rate": 8.62e-06,
|
| 352 |
+
"loss": 0.2756,
|
| 353 |
"step": 1400
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 0.2,
|
| 357 |
+
"learning_rate": 8.581538461538463e-06,
|
| 358 |
+
"loss": 0.294,
|
| 359 |
"step": 1425
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.21,
|
| 363 |
+
"learning_rate": 8.543076923076923e-06,
|
| 364 |
+
"loss": 0.3915,
|
| 365 |
"step": 1450
|
| 366 |
},
|
| 367 |
{
|
| 368 |
+
"epoch": 0.21,
|
| 369 |
+
"learning_rate": 8.504615384615385e-06,
|
| 370 |
+
"loss": 0.3237,
|
| 371 |
"step": 1475
|
| 372 |
},
|
| 373 |
{
|
| 374 |
+
"epoch": 0.21,
|
| 375 |
+
"learning_rate": 8.466153846153847e-06,
|
| 376 |
+
"loss": 0.2723,
|
| 377 |
"step": 1500
|
| 378 |
},
|
| 379 |
{
|
| 380 |
+
"epoch": 0.22,
|
| 381 |
+
"learning_rate": 8.427692307692308e-06,
|
| 382 |
+
"loss": 0.2917,
|
| 383 |
"step": 1525
|
| 384 |
},
|
| 385 |
{
|
| 386 |
+
"epoch": 0.22,
|
| 387 |
+
"learning_rate": 8.38923076923077e-06,
|
| 388 |
+
"loss": 0.2921,
|
| 389 |
"step": 1550
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 0.23,
|
| 393 |
+
"learning_rate": 8.35076923076923e-06,
|
| 394 |
+
"loss": 0.3724,
|
| 395 |
"step": 1575
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 0.23,
|
| 399 |
+
"learning_rate": 8.312307692307693e-06,
|
| 400 |
+
"loss": 0.3364,
|
| 401 |
"step": 1600
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.23,
|
| 405 |
+
"learning_rate": 8.273846153846155e-06,
|
| 406 |
+
"loss": 0.3661,
|
| 407 |
"step": 1625
|
| 408 |
},
|
| 409 |
{
|
| 410 |
+
"epoch": 0.24,
|
| 411 |
+
"learning_rate": 8.235384615384615e-06,
|
| 412 |
+
"loss": 0.3307,
|
| 413 |
"step": 1650
|
| 414 |
},
|
| 415 |
{
|
| 416 |
+
"epoch": 0.24,
|
| 417 |
+
"learning_rate": 8.196923076923078e-06,
|
| 418 |
+
"loss": 0.3201,
|
| 419 |
"step": 1675
|
| 420 |
},
|
| 421 |
{
|
| 422 |
+
"epoch": 0.24,
|
| 423 |
+
"learning_rate": 8.15846153846154e-06,
|
| 424 |
+
"loss": 0.2973,
|
| 425 |
"step": 1700
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"epoch": 0.25,
|
| 429 |
+
"learning_rate": 8.120000000000002e-06,
|
| 430 |
+
"loss": 0.3013,
|
| 431 |
"step": 1725
|
| 432 |
},
|
| 433 |
{
|
| 434 |
+
"epoch": 0.25,
|
| 435 |
+
"learning_rate": 8.081538461538462e-06,
|
| 436 |
+
"loss": 0.3349,
|
| 437 |
"step": 1750
|
| 438 |
},
|
| 439 |
{
|
| 440 |
+
"epoch": 0.25,
|
| 441 |
+
"learning_rate": 8.043076923076923e-06,
|
| 442 |
+
"loss": 0.3244,
|
| 443 |
"step": 1775
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.26,
|
| 447 |
+
"learning_rate": 8.004615384615385e-06,
|
| 448 |
+
"loss": 0.3331,
|
| 449 |
"step": 1800
|
| 450 |
},
|
| 451 |
{
|
| 452 |
+
"epoch": 0.26,
|
| 453 |
+
"learning_rate": 7.966153846153847e-06,
|
| 454 |
+
"loss": 0.2723,
|
| 455 |
"step": 1825
|
| 456 |
},
|
| 457 |
{
|
| 458 |
+
"epoch": 0.26,
|
| 459 |
+
"learning_rate": 7.927692307692308e-06,
|
| 460 |
+
"loss": 0.316,
|
| 461 |
"step": 1850
|
| 462 |
},
|
| 463 |
{
|
| 464 |
+
"epoch": 0.27,
|
| 465 |
+
"learning_rate": 7.88923076923077e-06,
|
| 466 |
+
"loss": 0.2648,
|
| 467 |
"step": 1875
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 0.27,
|
| 471 |
+
"learning_rate": 7.850769230769232e-06,
|
| 472 |
+
"loss": 0.2965,
|
| 473 |
"step": 1900
|
| 474 |
},
|
| 475 |
{
|
| 476 |
+
"epoch": 0.28,
|
| 477 |
+
"learning_rate": 7.812307692307694e-06,
|
| 478 |
+
"loss": 0.256,
|
| 479 |
"step": 1925
|
| 480 |
},
|
| 481 |
{
|
| 482 |
+
"epoch": 0.28,
|
| 483 |
+
"learning_rate": 7.773846153846155e-06,
|
| 484 |
+
"loss": 0.2156,
|
| 485 |
"step": 1950
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.28,
|
| 489 |
+
"learning_rate": 7.735384615384615e-06,
|
| 490 |
+
"loss": 0.2418,
|
| 491 |
"step": 1975
|
| 492 |
},
|
| 493 |
{
|
| 494 |
+
"epoch": 0.29,
|
| 495 |
+
"learning_rate": 7.696923076923077e-06,
|
| 496 |
+
"loss": 0.2693,
|
| 497 |
"step": 2000
|
| 498 |
},
|
| 499 |
{
|
| 500 |
+
"epoch": 0.29,
|
| 501 |
+
"eval_loss": 0.3239877223968506,
|
| 502 |
+
"eval_runtime": 1817.7748,
|
| 503 |
+
"eval_samples_per_second": 3.626,
|
| 504 |
+
"eval_steps_per_second": 0.453,
|
| 505 |
+
"eval_wer": 22.052330997610273,
|
| 506 |
"step": 2000
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 0.29,
|
| 510 |
+
"learning_rate": 7.65846153846154e-06,
|
| 511 |
+
"loss": 0.2191,
|
| 512 |
"step": 2025
|
| 513 |
},
|
| 514 |
{
|
| 515 |
+
"epoch": 0.29,
|
| 516 |
+
"learning_rate": 7.620000000000001e-06,
|
| 517 |
+
"loss": 0.2437,
|
| 518 |
"step": 2050
|
| 519 |
},
|
| 520 |
{
|
| 521 |
+
"epoch": 0.3,
|
| 522 |
+
"learning_rate": 7.581538461538462e-06,
|
| 523 |
+
"loss": 0.2519,
|
| 524 |
"step": 2075
|
| 525 |
},
|
| 526 |
{
|
| 527 |
+
"epoch": 0.3,
|
| 528 |
+
"learning_rate": 7.544615384615386e-06,
|
| 529 |
+
"loss": 0.3198,
|
| 530 |
"step": 2100
|
| 531 |
},
|
| 532 |
{
|
| 533 |
+
"epoch": 0.3,
|
| 534 |
+
"learning_rate": 7.5061538461538465e-06,
|
| 535 |
+
"loss": 0.2708,
|
| 536 |
"step": 2125
|
| 537 |
},
|
| 538 |
{
|
| 539 |
+
"epoch": 0.31,
|
| 540 |
+
"learning_rate": 7.467692307692308e-06,
|
| 541 |
+
"loss": 0.265,
|
| 542 |
"step": 2150
|
| 543 |
},
|
| 544 |
{
|
| 545 |
+
"epoch": 0.31,
|
| 546 |
+
"learning_rate": 7.42923076923077e-06,
|
| 547 |
+
"loss": 0.2935,
|
| 548 |
"step": 2175
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 0.31,
|
| 552 |
+
"learning_rate": 7.390769230769231e-06,
|
| 553 |
+
"loss": 0.2595,
|
| 554 |
"step": 2200
|
| 555 |
},
|
| 556 |
{
|
| 557 |
+
"epoch": 0.32,
|
| 558 |
+
"learning_rate": 7.3523076923076935e-06,
|
| 559 |
+
"loss": 0.2613,
|
| 560 |
"step": 2225
|
| 561 |
},
|
| 562 |
{
|
| 563 |
+
"epoch": 0.32,
|
| 564 |
+
"learning_rate": 7.313846153846155e-06,
|
| 565 |
+
"loss": 0.2229,
|
| 566 |
"step": 2250
|
| 567 |
},
|
| 568 |
{
|
| 569 |
+
"epoch": 0.33,
|
| 570 |
+
"learning_rate": 7.275384615384616e-06,
|
| 571 |
+
"loss": 0.2814,
|
| 572 |
"step": 2275
|
| 573 |
},
|
| 574 |
{
|
| 575 |
+
"epoch": 0.33,
|
| 576 |
+
"learning_rate": 7.236923076923078e-06,
|
| 577 |
+
"loss": 0.2681,
|
| 578 |
"step": 2300
|
| 579 |
},
|
| 580 |
{
|
| 581 |
+
"epoch": 0.33,
|
| 582 |
+
"learning_rate": 7.198461538461539e-06,
|
| 583 |
+
"loss": 0.2531,
|
| 584 |
"step": 2325
|
| 585 |
},
|
| 586 |
{
|
| 587 |
+
"epoch": 0.34,
|
| 588 |
+
"learning_rate": 7.16e-06,
|
| 589 |
+
"loss": 0.2104,
|
| 590 |
"step": 2350
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 0.34,
|
| 594 |
+
"learning_rate": 7.121538461538462e-06,
|
| 595 |
+
"loss": 0.2294,
|
| 596 |
"step": 2375
|
| 597 |
},
|
| 598 |
{
|
| 599 |
+
"epoch": 0.34,
|
| 600 |
+
"learning_rate": 7.0830769230769235e-06,
|
| 601 |
+
"loss": 0.2556,
|
| 602 |
"step": 2400
|
| 603 |
},
|
| 604 |
{
|
| 605 |
+
"epoch": 0.35,
|
| 606 |
+
"learning_rate": 7.044615384615386e-06,
|
| 607 |
+
"loss": 0.2185,
|
| 608 |
"step": 2425
|
| 609 |
},
|
| 610 |
{
|
| 611 |
+
"epoch": 0.35,
|
| 612 |
+
"learning_rate": 7.006153846153847e-06,
|
| 613 |
+
"loss": 0.2585,
|
| 614 |
"step": 2450
|
| 615 |
},
|
| 616 |
{
|
| 617 |
+
"epoch": 0.35,
|
| 618 |
+
"learning_rate": 6.967692307692308e-06,
|
| 619 |
+
"loss": 0.2325,
|
| 620 |
"step": 2475
|
| 621 |
},
|
| 622 |
{
|
| 623 |
+
"epoch": 0.36,
|
| 624 |
+
"learning_rate": 6.9292307692307705e-06,
|
| 625 |
+
"loss": 0.2068,
|
| 626 |
"step": 2500
|
| 627 |
},
|
| 628 |
{
|
| 629 |
+
"epoch": 0.36,
|
| 630 |
+
"learning_rate": 6.890769230769231e-06,
|
| 631 |
+
"loss": 0.2233,
|
| 632 |
"step": 2525
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 0.36,
|
| 636 |
+
"learning_rate": 6.852307692307692e-06,
|
| 637 |
+
"loss": 0.2649,
|
| 638 |
"step": 2550
|
| 639 |
},
|
| 640 |
{
|
| 641 |
+
"epoch": 0.37,
|
| 642 |
+
"learning_rate": 6.8138461538461545e-06,
|
| 643 |
+
"loss": 0.2524,
|
| 644 |
"step": 2575
|
| 645 |
},
|
| 646 |
{
|
| 647 |
+
"epoch": 0.37,
|
| 648 |
+
"learning_rate": 6.775384615384616e-06,
|
| 649 |
+
"loss": 0.2847,
|
| 650 |
"step": 2600
|
| 651 |
},
|
| 652 |
{
|
| 653 |
+
"epoch": 0.38,
|
| 654 |
+
"learning_rate": 6.736923076923078e-06,
|
| 655 |
+
"loss": 0.1954,
|
| 656 |
"step": 2625
|
| 657 |
},
|
| 658 |
{
|
| 659 |
+
"epoch": 0.38,
|
| 660 |
+
"learning_rate": 6.698461538461539e-06,
|
| 661 |
+
"loss": 0.2553,
|
| 662 |
"step": 2650
|
| 663 |
},
|
| 664 |
{
|
| 665 |
+
"epoch": 0.38,
|
| 666 |
+
"learning_rate": 6.660000000000001e-06,
|
| 667 |
+
"loss": 0.2138,
|
| 668 |
"step": 2675
|
| 669 |
},
|
| 670 |
{
|
| 671 |
+
"epoch": 0.39,
|
| 672 |
+
"learning_rate": 6.621538461538463e-06,
|
| 673 |
+
"loss": 0.2245,
|
| 674 |
"step": 2700
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 0.39,
|
| 678 |
+
"learning_rate": 6.583076923076923e-06,
|
| 679 |
+
"loss": 0.2137,
|
| 680 |
"step": 2725
|
| 681 |
},
|
| 682 |
{
|
| 683 |
+
"epoch": 0.39,
|
| 684 |
+
"learning_rate": 6.5446153846153846e-06,
|
| 685 |
+
"loss": 0.2558,
|
| 686 |
"step": 2750
|
| 687 |
},
|
| 688 |
{
|
| 689 |
+
"epoch": 0.4,
|
| 690 |
+
"learning_rate": 6.506153846153847e-06,
|
| 691 |
+
"loss": 0.1981,
|
| 692 |
"step": 2775
|
| 693 |
},
|
| 694 |
{
|
| 695 |
+
"epoch": 0.4,
|
| 696 |
+
"learning_rate": 6.467692307692308e-06,
|
| 697 |
+
"loss": 0.2327,
|
| 698 |
"step": 2800
|
| 699 |
},
|
| 700 |
{
|
| 701 |
+
"epoch": 0.4,
|
| 702 |
+
"learning_rate": 6.42923076923077e-06,
|
| 703 |
+
"loss": 0.219,
|
| 704 |
"step": 2825
|
| 705 |
},
|
| 706 |
{
|
| 707 |
+
"epoch": 0.41,
|
| 708 |
+
"learning_rate": 6.3907692307692315e-06,
|
| 709 |
+
"loss": 0.226,
|
| 710 |
"step": 2850
|
| 711 |
},
|
| 712 |
{
|
| 713 |
+
"epoch": 0.41,
|
| 714 |
+
"learning_rate": 6.352307692307693e-06,
|
| 715 |
+
"loss": 0.2192,
|
| 716 |
"step": 2875
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 0.41,
|
| 720 |
+
"learning_rate": 6.313846153846155e-06,
|
| 721 |
+
"loss": 0.2146,
|
| 722 |
"step": 2900
|
| 723 |
},
|
| 724 |
{
|
| 725 |
+
"epoch": 0.42,
|
| 726 |
+
"learning_rate": 6.2753846153846155e-06,
|
| 727 |
+
"loss": 0.2058,
|
| 728 |
"step": 2925
|
| 729 |
},
|
| 730 |
{
|
| 731 |
+
"epoch": 0.42,
|
| 732 |
+
"learning_rate": 6.236923076923077e-06,
|
| 733 |
+
"loss": 0.2369,
|
| 734 |
"step": 2950
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 0.42,
|
| 738 |
+
"learning_rate": 6.198461538461539e-06,
|
| 739 |
+
"loss": 0.2202,
|
| 740 |
"step": 2975
|
| 741 |
},
|
| 742 |
{
|
| 743 |
+
"epoch": 0.43,
|
| 744 |
+
"learning_rate": 6.16e-06,
|
| 745 |
+
"loss": 0.2228,
|
| 746 |
"step": 3000
|
| 747 |
},
|
| 748 |
{
|
| 749 |
+
"epoch": 0.43,
|
| 750 |
+
"eval_loss": 0.27374517917633057,
|
| 751 |
+
"eval_runtime": 1805.7724,
|
| 752 |
+
"eval_samples_per_second": 3.65,
|
| 753 |
+
"eval_steps_per_second": 0.456,
|
| 754 |
+
"eval_wer": 18.14370772408765,
|
| 755 |
"step": 3000
|
| 756 |
},
|
| 757 |
{
|
| 758 |
+
"epoch": 0.43,
|
| 759 |
+
"learning_rate": 6.1215384615384625e-06,
|
| 760 |
+
"loss": 0.2029,
|
| 761 |
"step": 3025
|
| 762 |
},
|
| 763 |
{
|
| 764 |
+
"epoch": 0.44,
|
| 765 |
+
"learning_rate": 6.083076923076924e-06,
|
| 766 |
+
"loss": 0.2085,
|
| 767 |
"step": 3050
|
| 768 |
},
|
| 769 |
{
|
| 770 |
+
"epoch": 0.44,
|
| 771 |
+
"learning_rate": 6.044615384615385e-06,
|
| 772 |
+
"loss": 0.196,
|
| 773 |
"step": 3075
|
| 774 |
},
|
| 775 |
{
|
| 776 |
+
"epoch": 0.44,
|
| 777 |
+
"learning_rate": 6.006153846153847e-06,
|
| 778 |
+
"loss": 0.1999,
|
| 779 |
"step": 3100
|
| 780 |
},
|
| 781 |
{
|
| 782 |
+
"epoch": 0.45,
|
| 783 |
+
"learning_rate": 5.967692307692308e-06,
|
| 784 |
+
"loss": 0.2076,
|
| 785 |
"step": 3125
|
| 786 |
},
|
| 787 |
{
|
| 788 |
+
"epoch": 0.45,
|
| 789 |
+
"learning_rate": 5.929230769230769e-06,
|
| 790 |
+
"loss": 0.1999,
|
| 791 |
"step": 3150
|
| 792 |
},
|
| 793 |
{
|
| 794 |
+
"epoch": 0.45,
|
| 795 |
+
"learning_rate": 5.890769230769231e-06,
|
| 796 |
+
"loss": 0.2361,
|
| 797 |
"step": 3175
|
| 798 |
},
|
| 799 |
{
|
| 800 |
+
"epoch": 0.46,
|
| 801 |
+
"learning_rate": 5.8523076923076926e-06,
|
| 802 |
+
"loss": 0.2003,
|
| 803 |
"step": 3200
|
| 804 |
},
|
| 805 |
{
|
| 806 |
+
"epoch": 0.46,
|
| 807 |
+
"learning_rate": 5.813846153846155e-06,
|
| 808 |
+
"loss": 0.249,
|
| 809 |
"step": 3225
|
| 810 |
},
|
| 811 |
{
|
| 812 |
+
"epoch": 0.46,
|
| 813 |
+
"learning_rate": 5.775384615384616e-06,
|
| 814 |
+
"loss": 0.2724,
|
| 815 |
"step": 3250
|
| 816 |
},
|
| 817 |
{
|
| 818 |
+
"epoch": 0.47,
|
| 819 |
+
"learning_rate": 5.736923076923077e-06,
|
| 820 |
+
"loss": 0.2001,
|
| 821 |
"step": 3275
|
| 822 |
},
|
| 823 |
{
|
| 824 |
"epoch": 1.0,
|
| 825 |
+
"learning_rate": 5.6984615384615395e-06,
|
| 826 |
+
"loss": 0.2355,
|
| 827 |
"step": 3300
|
| 828 |
},
|
| 829 |
{
|
| 830 |
+
"epoch": 1.0,
|
| 831 |
+
"learning_rate": 5.66e-06,
|
| 832 |
+
"loss": 0.2297,
|
| 833 |
"step": 3325
|
| 834 |
},
|
| 835 |
{
|
| 836 |
"epoch": 1.01,
|
| 837 |
+
"learning_rate": 5.621538461538461e-06,
|
| 838 |
+
"loss": 0.2265,
|
| 839 |
"step": 3350
|
| 840 |
},
|
| 841 |
{
|
| 842 |
+
"epoch": 1.01,
|
| 843 |
+
"learning_rate": 5.5830769230769235e-06,
|
| 844 |
+
"loss": 0.21,
|
| 845 |
"step": 3375
|
| 846 |
},
|
| 847 |
{
|
| 848 |
+
"epoch": 1.01,
|
| 849 |
+
"learning_rate": 5.544615384615385e-06,
|
| 850 |
+
"loss": 0.1944,
|
| 851 |
"step": 3400
|
| 852 |
},
|
| 853 |
{
|
| 854 |
+
"epoch": 1.02,
|
| 855 |
+
"learning_rate": 5.506153846153847e-06,
|
| 856 |
+
"loss": 0.1844,
|
| 857 |
"step": 3425
|
| 858 |
},
|
| 859 |
{
|
| 860 |
+
"epoch": 1.02,
|
| 861 |
+
"learning_rate": 5.467692307692308e-06,
|
| 862 |
+
"loss": 0.1774,
|
| 863 |
"step": 3450
|
| 864 |
},
|
| 865 |
{
|
| 866 |
+
"epoch": 1.03,
|
| 867 |
+
"learning_rate": 5.42923076923077e-06,
|
| 868 |
+
"loss": 0.1688,
|
| 869 |
"step": 3475
|
| 870 |
},
|
| 871 |
{
|
| 872 |
+
"epoch": 1.03,
|
| 873 |
+
"learning_rate": 5.390769230769232e-06,
|
| 874 |
+
"loss": 0.1473,
|
| 875 |
"step": 3500
|
| 876 |
},
|
| 877 |
{
|
| 878 |
+
"epoch": 1.03,
|
| 879 |
+
"learning_rate": 5.352307692307692e-06,
|
| 880 |
+
"loss": 0.1743,
|
| 881 |
"step": 3525
|
| 882 |
},
|
| 883 |
{
|
| 884 |
+
"epoch": 1.04,
|
| 885 |
+
"learning_rate": 5.313846153846154e-06,
|
| 886 |
+
"loss": 0.1759,
|
| 887 |
"step": 3550
|
| 888 |
},
|
| 889 |
{
|
| 890 |
+
"epoch": 1.04,
|
| 891 |
+
"learning_rate": 5.275384615384616e-06,
|
| 892 |
+
"loss": 0.1507,
|
| 893 |
"step": 3575
|
| 894 |
},
|
| 895 |
{
|
| 896 |
+
"epoch": 1.04,
|
| 897 |
+
"learning_rate": 5.236923076923077e-06,
|
| 898 |
+
"loss": 0.1556,
|
| 899 |
"step": 3600
|
| 900 |
},
|
| 901 |
{
|
| 902 |
+
"epoch": 1.05,
|
| 903 |
+
"learning_rate": 5.198461538461539e-06,
|
| 904 |
+
"loss": 0.1545,
|
| 905 |
"step": 3625
|
| 906 |
},
|
| 907 |
{
|
| 908 |
+
"epoch": 1.05,
|
| 909 |
+
"learning_rate": 5.1600000000000006e-06,
|
| 910 |
+
"loss": 0.1143,
|
| 911 |
"step": 3650
|
| 912 |
},
|
| 913 |
{
|
| 914 |
+
"epoch": 1.05,
|
| 915 |
+
"learning_rate": 5.121538461538462e-06,
|
| 916 |
+
"loss": 0.1168,
|
| 917 |
"step": 3675
|
| 918 |
},
|
| 919 |
{
|
| 920 |
+
"epoch": 1.06,
|
| 921 |
+
"learning_rate": 5.083076923076924e-06,
|
| 922 |
+
"loss": 0.1095,
|
| 923 |
"step": 3700
|
| 924 |
},
|
| 925 |
{
|
| 926 |
+
"epoch": 1.06,
|
| 927 |
+
"learning_rate": 5.044615384615385e-06,
|
| 928 |
+
"loss": 0.1098,
|
| 929 |
"step": 3725
|
| 930 |
},
|
| 931 |
{
|
| 932 |
+
"epoch": 1.06,
|
| 933 |
+
"learning_rate": 5.006153846153846e-06,
|
| 934 |
+
"loss": 0.102,
|
| 935 |
"step": 3750
|
| 936 |
},
|
| 937 |
{
|
| 938 |
+
"epoch": 1.07,
|
| 939 |
+
"learning_rate": 4.967692307692308e-06,
|
| 940 |
+
"loss": 0.1263,
|
| 941 |
"step": 3775
|
| 942 |
},
|
| 943 |
{
|
| 944 |
+
"epoch": 1.07,
|
| 945 |
+
"learning_rate": 4.929230769230769e-06,
|
| 946 |
+
"loss": 0.093,
|
| 947 |
"step": 3800
|
| 948 |
},
|
| 949 |
{
|
| 950 |
+
"epoch": 1.08,
|
| 951 |
+
"learning_rate": 4.8907692307692315e-06,
|
| 952 |
+
"loss": 0.1163,
|
| 953 |
"step": 3825
|
| 954 |
},
|
| 955 |
{
|
| 956 |
+
"epoch": 1.08,
|
| 957 |
+
"learning_rate": 4.852307692307693e-06,
|
| 958 |
+
"loss": 0.1005,
|
| 959 |
"step": 3850
|
| 960 |
},
|
| 961 |
{
|
| 962 |
+
"epoch": 1.08,
|
| 963 |
+
"learning_rate": 4.813846153846154e-06,
|
| 964 |
+
"loss": 0.1036,
|
| 965 |
"step": 3875
|
| 966 |
},
|
| 967 |
{
|
| 968 |
+
"epoch": 1.09,
|
| 969 |
+
"learning_rate": 4.7753846153846155e-06,
|
| 970 |
+
"loss": 0.1289,
|
| 971 |
"step": 3900
|
| 972 |
},
|
| 973 |
{
|
| 974 |
+
"epoch": 1.09,
|
| 975 |
+
"learning_rate": 4.736923076923078e-06,
|
| 976 |
+
"loss": 0.1238,
|
| 977 |
"step": 3925
|
| 978 |
},
|
| 979 |
{
|
| 980 |
+
"epoch": 1.09,
|
| 981 |
+
"learning_rate": 4.698461538461539e-06,
|
| 982 |
+
"loss": 0.1218,
|
| 983 |
"step": 3950
|
| 984 |
},
|
| 985 |
{
|
| 986 |
+
"epoch": 1.1,
|
| 987 |
+
"learning_rate": 4.66e-06,
|
| 988 |
+
"loss": 0.1223,
|
| 989 |
"step": 3975
|
| 990 |
},
|
| 991 |
{
|
| 992 |
+
"epoch": 1.1,
|
| 993 |
+
"learning_rate": 4.621538461538462e-06,
|
| 994 |
+
"loss": 0.1002,
|
| 995 |
"step": 4000
|
| 996 |
},
|
| 997 |
{
|
| 998 |
+
"epoch": 1.1,
|
| 999 |
+
"eval_loss": 0.2553846836090088,
|
| 1000 |
+
"eval_runtime": 1819.1705,
|
| 1001 |
+
"eval_samples_per_second": 3.623,
|
| 1002 |
+
"eval_steps_per_second": 0.453,
|
| 1003 |
+
"eval_wer": 16.353436753209934,
|
| 1004 |
"step": 4000
|
| 1005 |
}
|
| 1006 |
],
|
| 1007 |
+
"max_steps": 7000,
|
| 1008 |
"num_train_epochs": 9223372036854775807,
|
| 1009 |
"total_flos": 1.632763731050496e+19,
|
| 1010 |
"trial_name": null,
|
checkpoint-4000/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3643
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1ca27d9d1bef820c87cfd52d5adb15af5250e1ac4555cd1703a088ad0dad1a4
|
| 3 |
size 3643
|
nohup.out
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3055754841
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:def26fdaf433678a4a7ce858c8c045bcb0081f1a9d18211fd996f65174b4429e
|
| 3 |
size 3055754841
|
runs/Jul23_08-44-43_tknadmin-System-Product-Name/events.out.tfevents.1690094707.tknadmin-System-Product-Name.1702135.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fd2b84874237941afbfb748358971b34ae5f5412082b7a191a6634ffde33352
|
| 3 |
+
size 30658
|