Training in progress, epoch 1, checkpoint
Browse files
last-checkpoint/adapter_config.json
CHANGED
|
@@ -8,16 +8,16 @@
|
|
| 8 |
"init_lora_weights": true,
|
| 9 |
"layers_pattern": null,
|
| 10 |
"layers_to_transform": null,
|
| 11 |
-
"lora_alpha":
|
| 12 |
"lora_dropout": 0.1,
|
| 13 |
"modules_to_save": null,
|
| 14 |
"peft_type": "LORA",
|
| 15 |
-
"r":
|
| 16 |
"rank_pattern": {},
|
| 17 |
"revision": null,
|
| 18 |
"target_modules": [
|
| 19 |
-
"
|
| 20 |
-
"
|
| 21 |
],
|
| 22 |
"task_type": "SEQ_2_SEQ_LM"
|
| 23 |
}
|
|
|
|
| 8 |
"init_lora_weights": true,
|
| 9 |
"layers_pattern": null,
|
| 10 |
"layers_to_transform": null,
|
| 11 |
+
"lora_alpha": 16,
|
| 12 |
"lora_dropout": 0.1,
|
| 13 |
"modules_to_save": null,
|
| 14 |
"peft_type": "LORA",
|
| 15 |
+
"r": 16,
|
| 16 |
"rank_pattern": {},
|
| 17 |
"revision": null,
|
| 18 |
"target_modules": [
|
| 19 |
+
"q",
|
| 20 |
+
"v"
|
| 21 |
],
|
| 22 |
"task_type": "SEQ_2_SEQ_LM"
|
| 23 |
}
|
last-checkpoint/adapter_model.safetensors
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 sha256:6889584cb792de55b0a13842442a58ad2f3568b3c86e00c59b4d55790ac37264
|
| 3 |
+
size 37789864
|
last-checkpoint/optimizer.pt
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 sha256:1827bfaa4ae59f92b5bc715e7e5f5aec5bc6fdcb175cdb1df735d2dddf4f95be
|
| 3 |
+
size 75739130
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:91bba6ff03c8e09a85332cb162792e309f2a25041c51c9145b10b5cd75ee7dfb
|
| 3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:917b485f18156b96c93bc0dd12087e5d8b63a3098bce56ebac8eace0334f00f3
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,663 +1,854 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"learning_rate":
|
| 14 |
-
"loss":
|
| 15 |
"step": 500
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
-
"learning_rate":
|
| 20 |
-
"loss":
|
| 21 |
"step": 1000
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
-
"learning_rate":
|
| 26 |
-
"loss":
|
| 27 |
"step": 1500
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 0.
|
| 31 |
-
"learning_rate":
|
| 32 |
-
"loss":
|
| 33 |
"step": 2000
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"learning_rate":
|
| 38 |
-
"loss":
|
| 39 |
"step": 2500
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss":
|
| 45 |
"step": 3000
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 0.
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
| 51 |
"step": 3500
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"learning_rate":
|
| 56 |
-
"loss":
|
| 57 |
"step": 4000
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch": 0.
|
| 61 |
-
"learning_rate":
|
| 62 |
-
"loss":
|
| 63 |
"step": 4500
|
| 64 |
},
|
| 65 |
{
|
| 66 |
-
"epoch": 0.
|
| 67 |
-
"learning_rate":
|
| 68 |
-
"loss":
|
| 69 |
"step": 5000
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch": 0.
|
| 73 |
-
"learning_rate":
|
| 74 |
-
"loss":
|
| 75 |
"step": 5500
|
| 76 |
},
|
| 77 |
{
|
| 78 |
-
"epoch": 0.
|
| 79 |
-
"learning_rate":
|
| 80 |
-
"loss":
|
| 81 |
"step": 6000
|
| 82 |
},
|
| 83 |
{
|
| 84 |
-
"epoch": 0.
|
| 85 |
-
"learning_rate":
|
| 86 |
-
"loss":
|
| 87 |
"step": 6500
|
| 88 |
},
|
| 89 |
{
|
| 90 |
-
"epoch": 0.
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss":
|
| 93 |
"step": 7000
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"learning_rate":
|
| 98 |
-
"loss":
|
| 99 |
"step": 7500
|
| 100 |
},
|
| 101 |
{
|
| 102 |
-
"epoch": 0.
|
| 103 |
-
"learning_rate":
|
| 104 |
-
"loss":
|
| 105 |
"step": 8000
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"epoch": 0.
|
| 109 |
-
"learning_rate":
|
| 110 |
-
"loss":
|
| 111 |
"step": 8500
|
| 112 |
},
|
| 113 |
{
|
| 114 |
-
"epoch": 0.
|
| 115 |
-
"learning_rate":
|
| 116 |
-
"loss":
|
| 117 |
"step": 9000
|
| 118 |
},
|
| 119 |
{
|
| 120 |
-
"epoch": 0.
|
| 121 |
-
"learning_rate":
|
| 122 |
-
"loss":
|
| 123 |
"step": 9500
|
| 124 |
},
|
| 125 |
{
|
| 126 |
-
"epoch": 0.
|
| 127 |
-
"learning_rate":
|
| 128 |
-
"loss":
|
| 129 |
"step": 10000
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch": 0.
|
| 133 |
-
"learning_rate":
|
| 134 |
-
"loss":
|
| 135 |
"step": 10500
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"learning_rate":
|
| 140 |
-
"loss":
|
| 141 |
"step": 11000
|
| 142 |
},
|
| 143 |
{
|
| 144 |
-
"epoch": 0.
|
| 145 |
-
"learning_rate":
|
| 146 |
-
"loss":
|
| 147 |
"step": 11500
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch": 0.
|
| 151 |
-
"learning_rate":
|
| 152 |
-
"loss":
|
| 153 |
"step": 12000
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"epoch": 0.
|
| 157 |
-
"learning_rate":
|
| 158 |
-
"loss":
|
| 159 |
"step": 12500
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"epoch": 0.
|
| 163 |
-
"learning_rate":
|
| 164 |
-
"loss":
|
| 165 |
"step": 13000
|
| 166 |
},
|
| 167 |
{
|
| 168 |
-
"epoch": 0.
|
| 169 |
-
"learning_rate":
|
| 170 |
-
"loss":
|
| 171 |
"step": 13500
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch": 0.
|
| 175 |
-
"learning_rate":
|
| 176 |
-
"loss":
|
| 177 |
"step": 14000
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"learning_rate":
|
| 182 |
-
"loss":
|
| 183 |
"step": 14500
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"epoch": 0.
|
| 187 |
-
"learning_rate":
|
| 188 |
-
"loss":
|
| 189 |
"step": 15000
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch": 0.
|
| 193 |
-
"learning_rate":
|
| 194 |
-
"loss":
|
| 195 |
"step": 15500
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch": 0.
|
| 199 |
-
"learning_rate":
|
| 200 |
-
"loss":
|
| 201 |
"step": 16000
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch": 0.
|
| 205 |
-
"learning_rate":
|
| 206 |
-
"loss":
|
| 207 |
"step": 16500
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch": 0.
|
| 211 |
-
"learning_rate":
|
| 212 |
-
"loss":
|
| 213 |
"step": 17000
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch": 0.
|
| 217 |
-
"learning_rate":
|
| 218 |
-
"loss":
|
| 219 |
"step": 17500
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"learning_rate":
|
| 224 |
-
"loss":
|
| 225 |
"step": 18000
|
| 226 |
},
|
| 227 |
{
|
| 228 |
-
"epoch": 0.
|
| 229 |
-
"learning_rate":
|
| 230 |
-
"loss":
|
| 231 |
"step": 18500
|
| 232 |
},
|
| 233 |
{
|
| 234 |
-
"epoch": 0.
|
| 235 |
-
"learning_rate":
|
| 236 |
-
"loss":
|
| 237 |
"step": 19000
|
| 238 |
},
|
| 239 |
{
|
| 240 |
-
"epoch": 0.
|
| 241 |
-
"learning_rate":
|
| 242 |
-
"loss":
|
| 243 |
"step": 19500
|
| 244 |
},
|
| 245 |
{
|
| 246 |
-
"epoch": 0.
|
| 247 |
-
"learning_rate":
|
| 248 |
-
"loss":
|
| 249 |
"step": 20000
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch": 0.
|
| 253 |
-
"learning_rate":
|
| 254 |
-
"loss":
|
| 255 |
"step": 20500
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch": 0.
|
| 259 |
-
"learning_rate":
|
| 260 |
-
"loss":
|
| 261 |
"step": 21000
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 0.
|
| 265 |
-
"learning_rate":
|
| 266 |
-
"loss":
|
| 267 |
"step": 21500
|
| 268 |
},
|
| 269 |
{
|
| 270 |
-
"epoch": 0.
|
| 271 |
-
"learning_rate":
|
| 272 |
-
"loss":
|
| 273 |
"step": 22000
|
| 274 |
},
|
| 275 |
{
|
| 276 |
-
"epoch": 0.
|
| 277 |
-
"learning_rate":
|
| 278 |
-
"loss":
|
| 279 |
"step": 22500
|
| 280 |
},
|
| 281 |
{
|
| 282 |
-
"epoch": 0.
|
| 283 |
-
"learning_rate":
|
| 284 |
-
"loss":
|
| 285 |
"step": 23000
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"epoch": 0.
|
| 289 |
-
"learning_rate":
|
| 290 |
-
"loss":
|
| 291 |
"step": 23500
|
| 292 |
},
|
| 293 |
{
|
| 294 |
-
"epoch": 0.
|
| 295 |
-
"learning_rate":
|
| 296 |
-
"loss":
|
| 297 |
"step": 24000
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch": 0.
|
| 301 |
-
"learning_rate":
|
| 302 |
-
"loss":
|
| 303 |
"step": 24500
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 0.
|
| 307 |
-
"learning_rate":
|
| 308 |
-
"loss":
|
| 309 |
"step": 25000
|
| 310 |
},
|
| 311 |
{
|
| 312 |
-
"epoch":
|
| 313 |
-
"
|
| 314 |
-
"
|
| 315 |
-
"eval_length_ratio": 1.0,
|
| 316 |
-
"eval_loss": 2.3814520835876465,
|
| 317 |
-
"eval_precisions": [
|
| 318 |
-
1.0,
|
| 319 |
-
1.0,
|
| 320 |
-
1.0,
|
| 321 |
-
1.0
|
| 322 |
-
],
|
| 323 |
-
"eval_reference_length": 1439232,
|
| 324 |
-
"eval_runtime": 2544.152,
|
| 325 |
-
"eval_samples_per_second": 1.105,
|
| 326 |
-
"eval_steps_per_second": 1.105,
|
| 327 |
-
"eval_translation_length": 1439232,
|
| 328 |
-
"step": 25291
|
| 329 |
-
},
|
| 330 |
-
{
|
| 331 |
-
"epoch": 1.01,
|
| 332 |
-
"learning_rate": 9.973200326855681e-05,
|
| 333 |
-
"loss": 2.3745,
|
| 334 |
"step": 25500
|
| 335 |
},
|
| 336 |
{
|
| 337 |
-
"epoch":
|
| 338 |
-
"learning_rate":
|
| 339 |
-
"loss":
|
| 340 |
"step": 26000
|
| 341 |
},
|
| 342 |
{
|
| 343 |
-
"epoch":
|
| 344 |
-
"learning_rate":
|
| 345 |
-
"loss":
|
| 346 |
"step": 26500
|
| 347 |
},
|
| 348 |
{
|
| 349 |
-
"epoch":
|
| 350 |
-
"learning_rate":
|
| 351 |
-
"loss":
|
| 352 |
"step": 27000
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"epoch":
|
| 356 |
-
"learning_rate":
|
| 357 |
-
"loss":
|
| 358 |
"step": 27500
|
| 359 |
},
|
| 360 |
{
|
| 361 |
-
"epoch":
|
| 362 |
-
"learning_rate":
|
| 363 |
-
"loss":
|
| 364 |
"step": 28000
|
| 365 |
},
|
| 366 |
{
|
| 367 |
-
"epoch":
|
| 368 |
-
"learning_rate":
|
| 369 |
-
"loss":
|
| 370 |
"step": 28500
|
| 371 |
},
|
| 372 |
{
|
| 373 |
-
"epoch":
|
| 374 |
-
"learning_rate":
|
| 375 |
-
"loss":
|
| 376 |
"step": 29000
|
| 377 |
},
|
| 378 |
{
|
| 379 |
-
"epoch":
|
| 380 |
-
"learning_rate":
|
| 381 |
-
"loss":
|
| 382 |
"step": 29500
|
| 383 |
},
|
| 384 |
{
|
| 385 |
-
"epoch":
|
| 386 |
-
"learning_rate":
|
| 387 |
-
"loss":
|
| 388 |
"step": 30000
|
| 389 |
},
|
| 390 |
{
|
| 391 |
-
"epoch":
|
| 392 |
-
"learning_rate":
|
| 393 |
-
"loss":
|
| 394 |
"step": 30500
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch":
|
| 398 |
-
"learning_rate":
|
| 399 |
-
"loss":
|
| 400 |
"step": 31000
|
| 401 |
},
|
| 402 |
{
|
| 403 |
-
"epoch":
|
| 404 |
-
"learning_rate":
|
| 405 |
-
"loss":
|
| 406 |
"step": 31500
|
| 407 |
},
|
| 408 |
{
|
| 409 |
-
"epoch":
|
| 410 |
-
"learning_rate":
|
| 411 |
-
"loss":
|
| 412 |
"step": 32000
|
| 413 |
},
|
| 414 |
{
|
| 415 |
-
"epoch":
|
| 416 |
-
"learning_rate":
|
| 417 |
-
"loss":
|
| 418 |
"step": 32500
|
| 419 |
},
|
| 420 |
{
|
| 421 |
-
"epoch":
|
| 422 |
-
"learning_rate":
|
| 423 |
-
"loss":
|
| 424 |
"step": 33000
|
| 425 |
},
|
| 426 |
{
|
| 427 |
-
"epoch":
|
| 428 |
-
"learning_rate":
|
| 429 |
-
"loss":
|
| 430 |
"step": 33500
|
| 431 |
},
|
| 432 |
{
|
| 433 |
-
"epoch":
|
| 434 |
-
"learning_rate":
|
| 435 |
-
"loss":
|
| 436 |
"step": 34000
|
| 437 |
},
|
| 438 |
{
|
| 439 |
-
"epoch":
|
| 440 |
-
"learning_rate":
|
| 441 |
-
"loss":
|
| 442 |
"step": 34500
|
| 443 |
},
|
| 444 |
{
|
| 445 |
-
"epoch":
|
| 446 |
-
"learning_rate":
|
| 447 |
-
"loss":
|
| 448 |
"step": 35000
|
| 449 |
},
|
| 450 |
{
|
| 451 |
-
"epoch":
|
| 452 |
-
"learning_rate":
|
| 453 |
-
"loss":
|
| 454 |
"step": 35500
|
| 455 |
},
|
| 456 |
{
|
| 457 |
-
"epoch":
|
| 458 |
-
"learning_rate":
|
| 459 |
-
"loss":
|
| 460 |
"step": 36000
|
| 461 |
},
|
| 462 |
{
|
| 463 |
-
"epoch":
|
| 464 |
-
"learning_rate":
|
| 465 |
-
"loss":
|
| 466 |
"step": 36500
|
| 467 |
},
|
| 468 |
{
|
| 469 |
-
"epoch":
|
| 470 |
-
"learning_rate":
|
| 471 |
-
"loss":
|
| 472 |
"step": 37000
|
| 473 |
},
|
| 474 |
{
|
| 475 |
-
"epoch":
|
| 476 |
-
"learning_rate":
|
| 477 |
-
"loss":
|
| 478 |
"step": 37500
|
| 479 |
},
|
| 480 |
{
|
| 481 |
-
"epoch":
|
| 482 |
-
"learning_rate":
|
| 483 |
-
"loss":
|
| 484 |
"step": 38000
|
| 485 |
},
|
| 486 |
{
|
| 487 |
-
"epoch":
|
| 488 |
-
"learning_rate":
|
| 489 |
-
"loss":
|
| 490 |
"step": 38500
|
| 491 |
},
|
| 492 |
{
|
| 493 |
-
"epoch":
|
| 494 |
-
"learning_rate":
|
| 495 |
-
"loss":
|
| 496 |
"step": 39000
|
| 497 |
},
|
| 498 |
{
|
| 499 |
-
"epoch":
|
| 500 |
-
"learning_rate":
|
| 501 |
-
"loss":
|
| 502 |
"step": 39500
|
| 503 |
},
|
| 504 |
{
|
| 505 |
-
"epoch":
|
| 506 |
-
"learning_rate":
|
| 507 |
-
"loss":
|
| 508 |
"step": 40000
|
| 509 |
},
|
| 510 |
{
|
| 511 |
-
"epoch":
|
| 512 |
-
"learning_rate":
|
| 513 |
-
"loss":
|
| 514 |
"step": 40500
|
| 515 |
},
|
| 516 |
{
|
| 517 |
-
"epoch":
|
| 518 |
-
"learning_rate":
|
| 519 |
-
"loss":
|
| 520 |
"step": 41000
|
| 521 |
},
|
| 522 |
{
|
| 523 |
-
"epoch":
|
| 524 |
-
"learning_rate":
|
| 525 |
-
"loss":
|
| 526 |
"step": 41500
|
| 527 |
},
|
| 528 |
{
|
| 529 |
-
"epoch":
|
| 530 |
-
"learning_rate":
|
| 531 |
-
"loss":
|
| 532 |
"step": 42000
|
| 533 |
},
|
| 534 |
{
|
| 535 |
-
"epoch":
|
| 536 |
-
"learning_rate":
|
| 537 |
-
"loss":
|
| 538 |
"step": 42500
|
| 539 |
},
|
| 540 |
{
|
| 541 |
-
"epoch":
|
| 542 |
-
"learning_rate":
|
| 543 |
-
"loss":
|
| 544 |
"step": 43000
|
| 545 |
},
|
| 546 |
{
|
| 547 |
-
"epoch":
|
| 548 |
-
"learning_rate":
|
| 549 |
-
"loss":
|
| 550 |
"step": 43500
|
| 551 |
},
|
| 552 |
{
|
| 553 |
-
"epoch":
|
| 554 |
-
"learning_rate":
|
| 555 |
-
"loss":
|
| 556 |
"step": 44000
|
| 557 |
},
|
| 558 |
{
|
| 559 |
-
"epoch":
|
| 560 |
-
"learning_rate":
|
| 561 |
-
"loss":
|
| 562 |
"step": 44500
|
| 563 |
},
|
| 564 |
{
|
| 565 |
-
"epoch":
|
| 566 |
-
"learning_rate":
|
| 567 |
-
"loss":
|
| 568 |
"step": 45000
|
| 569 |
},
|
| 570 |
{
|
| 571 |
-
"epoch":
|
| 572 |
-
"learning_rate":
|
| 573 |
-
"loss":
|
| 574 |
"step": 45500
|
| 575 |
},
|
| 576 |
{
|
| 577 |
-
"epoch":
|
| 578 |
-
"learning_rate":
|
| 579 |
-
"loss":
|
| 580 |
"step": 46000
|
| 581 |
},
|
| 582 |
{
|
| 583 |
-
"epoch":
|
| 584 |
-
"learning_rate":
|
| 585 |
-
"loss":
|
| 586 |
"step": 46500
|
| 587 |
},
|
| 588 |
{
|
| 589 |
-
"epoch":
|
| 590 |
-
"learning_rate":
|
| 591 |
-
"loss":
|
| 592 |
"step": 47000
|
| 593 |
},
|
| 594 |
{
|
| 595 |
-
"epoch":
|
| 596 |
-
"learning_rate":
|
| 597 |
-
"loss":
|
| 598 |
"step": 47500
|
| 599 |
},
|
| 600 |
{
|
| 601 |
-
"epoch":
|
| 602 |
-
"learning_rate":
|
| 603 |
-
"loss":
|
| 604 |
"step": 48000
|
| 605 |
},
|
| 606 |
{
|
| 607 |
-
"epoch":
|
| 608 |
-
"learning_rate":
|
| 609 |
-
"loss":
|
| 610 |
"step": 48500
|
| 611 |
},
|
| 612 |
{
|
| 613 |
-
"epoch":
|
| 614 |
-
"learning_rate":
|
| 615 |
-
"loss":
|
| 616 |
"step": 49000
|
| 617 |
},
|
| 618 |
{
|
| 619 |
-
"epoch":
|
| 620 |
-
"learning_rate":
|
| 621 |
-
"loss":
|
| 622 |
"step": 49500
|
| 623 |
},
|
| 624 |
{
|
| 625 |
-
"epoch":
|
| 626 |
-
"learning_rate":
|
| 627 |
-
"loss":
|
| 628 |
"step": 50000
|
| 629 |
},
|
| 630 |
{
|
| 631 |
-
"epoch":
|
| 632 |
-
"learning_rate":
|
| 633 |
-
"loss":
|
| 634 |
"step": 50500
|
| 635 |
},
|
| 636 |
{
|
| 637 |
-
"epoch":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 638 |
"eval_bleu": 1.0,
|
| 639 |
"eval_brevity_penalty": 1.0,
|
| 640 |
"eval_length_ratio": 1.0,
|
| 641 |
-
"eval_loss":
|
| 642 |
"eval_precisions": [
|
| 643 |
1.0,
|
| 644 |
1.0,
|
| 645 |
1.0,
|
| 646 |
1.0
|
| 647 |
],
|
| 648 |
-
"eval_reference_length":
|
| 649 |
-
"eval_runtime":
|
| 650 |
-
"eval_samples_per_second":
|
| 651 |
-
"eval_steps_per_second":
|
| 652 |
-
"eval_translation_length":
|
| 653 |
-
"step":
|
| 654 |
}
|
| 655 |
],
|
| 656 |
"logging_steps": 500,
|
| 657 |
-
"max_steps":
|
| 658 |
-
"num_train_epochs":
|
| 659 |
-
"save_steps":
|
| 660 |
-
"total_flos": 1.
|
| 661 |
"trial_name": null,
|
| 662 |
"trial_params": null
|
| 663 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 68219,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.01,
|
| 13 |
+
"learning_rate": 8e-05,
|
| 14 |
+
"loss": 4.3977,
|
| 15 |
"step": 500
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.01,
|
| 19 |
+
"learning_rate": 7.999957460493864e-05,
|
| 20 |
+
"loss": 3.6945,
|
| 21 |
"step": 1000
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.02,
|
| 25 |
+
"learning_rate": 7.999829842880257e-05,
|
| 26 |
+
"loss": 3.5381,
|
| 27 |
"step": 1500
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.03,
|
| 31 |
+
"learning_rate": 7.999617149873574e-05,
|
| 32 |
+
"loss": 3.5732,
|
| 33 |
"step": 2000
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.04,
|
| 37 |
+
"learning_rate": 7.999319385997746e-05,
|
| 38 |
+
"loss": 3.5564,
|
| 39 |
"step": 2500
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 0.04,
|
| 43 |
+
"learning_rate": 7.998936557586135e-05,
|
| 44 |
+
"loss": 3.6373,
|
| 45 |
"step": 3000
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 0.05,
|
| 49 |
+
"learning_rate": 7.998468672781407e-05,
|
| 50 |
+
"loss": 3.45,
|
| 51 |
"step": 3500
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.06,
|
| 55 |
+
"learning_rate": 7.997915741535355e-05,
|
| 56 |
+
"loss": 3.4784,
|
| 57 |
"step": 4000
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 0.07,
|
| 61 |
+
"learning_rate": 7.997277775608694e-05,
|
| 62 |
+
"loss": 3.5638,
|
| 63 |
"step": 4500
|
| 64 |
},
|
| 65 |
{
|
| 66 |
+
"epoch": 0.07,
|
| 67 |
+
"learning_rate": 7.996554788570796e-05,
|
| 68 |
+
"loss": 3.4186,
|
| 69 |
"step": 5000
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 0.08,
|
| 73 |
+
"learning_rate": 7.995746795799422e-05,
|
| 74 |
+
"loss": 3.4193,
|
| 75 |
"step": 5500
|
| 76 |
},
|
| 77 |
{
|
| 78 |
+
"epoch": 0.09,
|
| 79 |
+
"learning_rate": 7.994853814480376e-05,
|
| 80 |
+
"loss": 3.4769,
|
| 81 |
"step": 6000
|
| 82 |
},
|
| 83 |
{
|
| 84 |
+
"epoch": 0.1,
|
| 85 |
+
"learning_rate": 7.99387586360715e-05,
|
| 86 |
+
"loss": 3.3855,
|
| 87 |
"step": 6500
|
| 88 |
},
|
| 89 |
{
|
| 90 |
+
"epoch": 0.1,
|
| 91 |
+
"learning_rate": 7.992812963980518e-05,
|
| 92 |
+
"loss": 3.4352,
|
| 93 |
"step": 7000
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.11,
|
| 97 |
+
"learning_rate": 7.991665138208094e-05,
|
| 98 |
+
"loss": 3.3517,
|
| 99 |
"step": 7500
|
| 100 |
},
|
| 101 |
{
|
| 102 |
+
"epoch": 0.12,
|
| 103 |
+
"learning_rate": 7.990432410703848e-05,
|
| 104 |
+
"loss": 3.507,
|
| 105 |
"step": 8000
|
| 106 |
},
|
| 107 |
{
|
| 108 |
+
"epoch": 0.12,
|
| 109 |
+
"learning_rate": 7.989114807687589e-05,
|
| 110 |
+
"loss": 3.3754,
|
| 111 |
"step": 8500
|
| 112 |
},
|
| 113 |
{
|
| 114 |
+
"epoch": 0.13,
|
| 115 |
+
"learning_rate": 7.987712357184408e-05,
|
| 116 |
+
"loss": 3.2636,
|
| 117 |
"step": 9000
|
| 118 |
},
|
| 119 |
{
|
| 120 |
+
"epoch": 0.14,
|
| 121 |
+
"learning_rate": 7.98622508902408e-05,
|
| 122 |
+
"loss": 3.5107,
|
| 123 |
"step": 9500
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 0.15,
|
| 127 |
+
"learning_rate": 7.984653034840432e-05,
|
| 128 |
+
"loss": 3.3726,
|
| 129 |
"step": 10000
|
| 130 |
},
|
| 131 |
{
|
| 132 |
+
"epoch": 0.15,
|
| 133 |
+
"learning_rate": 7.982996228070671e-05,
|
| 134 |
+
"loss": 3.3568,
|
| 135 |
"step": 10500
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.16,
|
| 139 |
+
"learning_rate": 7.981254703954664e-05,
|
| 140 |
+
"loss": 3.3137,
|
| 141 |
"step": 11000
|
| 142 |
},
|
| 143 |
{
|
| 144 |
+
"epoch": 0.17,
|
| 145 |
+
"learning_rate": 7.979428499534201e-05,
|
| 146 |
+
"loss": 3.4073,
|
| 147 |
"step": 11500
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"epoch": 0.18,
|
| 151 |
+
"learning_rate": 7.977517653652199e-05,
|
| 152 |
+
"loss": 3.4058,
|
| 153 |
"step": 12000
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"epoch": 0.18,
|
| 157 |
+
"learning_rate": 7.975522206951876e-05,
|
| 158 |
+
"loss": 3.3605,
|
| 159 |
"step": 12500
|
| 160 |
},
|
| 161 |
{
|
| 162 |
+
"epoch": 0.19,
|
| 163 |
+
"learning_rate": 7.973442201875895e-05,
|
| 164 |
+
"loss": 3.2736,
|
| 165 |
"step": 13000
|
| 166 |
},
|
| 167 |
{
|
| 168 |
+
"epoch": 0.2,
|
| 169 |
+
"learning_rate": 7.971277682665446e-05,
|
| 170 |
+
"loss": 3.3546,
|
| 171 |
"step": 13500
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 0.21,
|
| 175 |
+
"learning_rate": 7.969028695359319e-05,
|
| 176 |
+
"loss": 3.3789,
|
| 177 |
"step": 14000
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.21,
|
| 181 |
+
"learning_rate": 7.966695287792921e-05,
|
| 182 |
+
"loss": 3.361,
|
| 183 |
"step": 14500
|
| 184 |
},
|
| 185 |
{
|
| 186 |
+
"epoch": 0.22,
|
| 187 |
+
"learning_rate": 7.96427750959725e-05,
|
| 188 |
+
"loss": 3.3593,
|
| 189 |
"step": 15000
|
| 190 |
},
|
| 191 |
{
|
| 192 |
+
"epoch": 0.23,
|
| 193 |
+
"learning_rate": 7.961775412197857e-05,
|
| 194 |
+
"loss": 3.3542,
|
| 195 |
"step": 15500
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 0.23,
|
| 199 |
+
"learning_rate": 7.959189048813735e-05,
|
| 200 |
+
"loss": 3.3135,
|
| 201 |
"step": 16000
|
| 202 |
},
|
| 203 |
{
|
| 204 |
+
"epoch": 0.24,
|
| 205 |
+
"learning_rate": 7.95651847445619e-05,
|
| 206 |
+
"loss": 3.3155,
|
| 207 |
"step": 16500
|
| 208 |
},
|
| 209 |
{
|
| 210 |
+
"epoch": 0.25,
|
| 211 |
+
"learning_rate": 7.953763745927682e-05,
|
| 212 |
+
"loss": 3.3343,
|
| 213 |
"step": 17000
|
| 214 |
},
|
| 215 |
{
|
| 216 |
+
"epoch": 0.26,
|
| 217 |
+
"learning_rate": 7.950924921820606e-05,
|
| 218 |
+
"loss": 3.3081,
|
| 219 |
"step": 17500
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.26,
|
| 223 |
+
"learning_rate": 7.948002062516052e-05,
|
| 224 |
+
"loss": 3.2967,
|
| 225 |
"step": 18000
|
| 226 |
},
|
| 227 |
{
|
| 228 |
+
"epoch": 0.27,
|
| 229 |
+
"learning_rate": 7.944995230182513e-05,
|
| 230 |
+
"loss": 3.2729,
|
| 231 |
"step": 18500
|
| 232 |
},
|
| 233 |
{
|
| 234 |
+
"epoch": 0.28,
|
| 235 |
+
"learning_rate": 7.941904488774571e-05,
|
| 236 |
+
"loss": 3.2456,
|
| 237 |
"step": 19000
|
| 238 |
},
|
| 239 |
{
|
| 240 |
+
"epoch": 0.29,
|
| 241 |
+
"learning_rate": 7.938729904031533e-05,
|
| 242 |
+
"loss": 3.2926,
|
| 243 |
"step": 19500
|
| 244 |
},
|
| 245 |
{
|
| 246 |
+
"epoch": 0.29,
|
| 247 |
+
"learning_rate": 7.93547154347603e-05,
|
| 248 |
+
"loss": 3.3109,
|
| 249 |
"step": 20000
|
| 250 |
},
|
| 251 |
{
|
| 252 |
+
"epoch": 0.3,
|
| 253 |
+
"learning_rate": 7.932129476412592e-05,
|
| 254 |
+
"loss": 3.3393,
|
| 255 |
"step": 20500
|
| 256 |
},
|
| 257 |
{
|
| 258 |
+
"epoch": 0.31,
|
| 259 |
+
"learning_rate": 7.928703773926155e-05,
|
| 260 |
+
"loss": 3.2807,
|
| 261 |
"step": 21000
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.32,
|
| 265 |
+
"learning_rate": 7.925194508880567e-05,
|
| 266 |
+
"loss": 3.2521,
|
| 267 |
"step": 21500
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 0.32,
|
| 271 |
+
"learning_rate": 7.921601755917029e-05,
|
| 272 |
+
"loss": 3.3654,
|
| 273 |
"step": 22000
|
| 274 |
},
|
| 275 |
{
|
| 276 |
+
"epoch": 0.33,
|
| 277 |
+
"learning_rate": 7.917925591452508e-05,
|
| 278 |
+
"loss": 3.1937,
|
| 279 |
"step": 22500
|
| 280 |
},
|
| 281 |
{
|
| 282 |
+
"epoch": 0.34,
|
| 283 |
+
"learning_rate": 7.914166093678117e-05,
|
| 284 |
+
"loss": 3.3133,
|
| 285 |
"step": 23000
|
| 286 |
},
|
| 287 |
{
|
| 288 |
+
"epoch": 0.34,
|
| 289 |
+
"learning_rate": 7.910323342557442e-05,
|
| 290 |
+
"loss": 3.2901,
|
| 291 |
"step": 23500
|
| 292 |
},
|
| 293 |
{
|
| 294 |
+
"epoch": 0.35,
|
| 295 |
+
"learning_rate": 7.906397419824855e-05,
|
| 296 |
+
"loss": 3.2792,
|
| 297 |
"step": 24000
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 0.36,
|
| 301 |
+
"learning_rate": 7.902388408983759e-05,
|
| 302 |
+
"loss": 3.2885,
|
| 303 |
"step": 24500
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 0.37,
|
| 307 |
+
"learning_rate": 7.898296395304824e-05,
|
| 308 |
+
"loss": 3.21,
|
| 309 |
"step": 25000
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 0.37,
|
| 313 |
+
"learning_rate": 7.894121465824175e-05,
|
| 314 |
+
"loss": 3.2116,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 315 |
"step": 25500
|
| 316 |
},
|
| 317 |
{
|
| 318 |
+
"epoch": 0.38,
|
| 319 |
+
"learning_rate": 7.889863709341528e-05,
|
| 320 |
+
"loss": 3.3064,
|
| 321 |
"step": 26000
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"epoch": 0.39,
|
| 325 |
+
"learning_rate": 7.885523216418312e-05,
|
| 326 |
+
"loss": 3.2047,
|
| 327 |
"step": 26500
|
| 328 |
},
|
| 329 |
{
|
| 330 |
+
"epoch": 0.4,
|
| 331 |
+
"learning_rate": 7.881100079375742e-05,
|
| 332 |
+
"loss": 3.3359,
|
| 333 |
"step": 27000
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 0.4,
|
| 337 |
+
"learning_rate": 7.876594392292848e-05,
|
| 338 |
+
"loss": 3.3722,
|
| 339 |
"step": 27500
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 0.41,
|
| 343 |
+
"learning_rate": 7.872006251004482e-05,
|
| 344 |
+
"loss": 3.3781,
|
| 345 |
"step": 28000
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 0.42,
|
| 349 |
+
"learning_rate": 7.867335753099278e-05,
|
| 350 |
+
"loss": 3.2035,
|
| 351 |
"step": 28500
|
| 352 |
},
|
| 353 |
{
|
| 354 |
+
"epoch": 0.43,
|
| 355 |
+
"learning_rate": 7.86258299791757e-05,
|
| 356 |
+
"loss": 3.1994,
|
| 357 |
"step": 29000
|
| 358 |
},
|
| 359 |
{
|
| 360 |
+
"epoch": 0.43,
|
| 361 |
+
"learning_rate": 7.857748086549292e-05,
|
| 362 |
+
"loss": 3.3272,
|
| 363 |
"step": 29500
|
| 364 |
},
|
| 365 |
{
|
| 366 |
+
"epoch": 0.44,
|
| 367 |
+
"learning_rate": 7.852831121831812e-05,
|
| 368 |
+
"loss": 3.2537,
|
| 369 |
"step": 30000
|
| 370 |
},
|
| 371 |
{
|
| 372 |
+
"epoch": 0.45,
|
| 373 |
+
"learning_rate": 7.847832208347754e-05,
|
| 374 |
+
"loss": 3.2472,
|
| 375 |
"step": 30500
|
| 376 |
},
|
| 377 |
{
|
| 378 |
+
"epoch": 0.45,
|
| 379 |
+
"learning_rate": 7.842751452422775e-05,
|
| 380 |
+
"loss": 3.2387,
|
| 381 |
"step": 31000
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 0.46,
|
| 385 |
+
"learning_rate": 7.8375889621233e-05,
|
| 386 |
+
"loss": 3.2847,
|
| 387 |
"step": 31500
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 0.47,
|
| 391 |
+
"learning_rate": 7.83234484725422e-05,
|
| 392 |
+
"loss": 3.1932,
|
| 393 |
"step": 32000
|
| 394 |
},
|
| 395 |
{
|
| 396 |
+
"epoch": 0.48,
|
| 397 |
+
"learning_rate": 7.827019219356568e-05,
|
| 398 |
+
"loss": 3.287,
|
| 399 |
"step": 32500
|
| 400 |
},
|
| 401 |
{
|
| 402 |
+
"epoch": 0.48,
|
| 403 |
+
"learning_rate": 7.821612191705128e-05,
|
| 404 |
+
"loss": 3.3191,
|
| 405 |
"step": 33000
|
| 406 |
},
|
| 407 |
{
|
| 408 |
+
"epoch": 0.49,
|
| 409 |
+
"learning_rate": 7.816123879306048e-05,
|
| 410 |
+
"loss": 3.2097,
|
| 411 |
"step": 33500
|
| 412 |
},
|
| 413 |
{
|
| 414 |
+
"epoch": 0.5,
|
| 415 |
+
"learning_rate": 7.810554398894376e-05,
|
| 416 |
+
"loss": 3.1723,
|
| 417 |
"step": 34000
|
| 418 |
},
|
| 419 |
{
|
| 420 |
+
"epoch": 0.51,
|
| 421 |
+
"learning_rate": 7.804903868931584e-05,
|
| 422 |
+
"loss": 3.2233,
|
| 423 |
"step": 34500
|
| 424 |
},
|
| 425 |
{
|
| 426 |
+
"epoch": 0.51,
|
| 427 |
+
"learning_rate": 7.79917240960305e-05,
|
| 428 |
+
"loss": 3.1182,
|
| 429 |
"step": 35000
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 0.52,
|
| 433 |
+
"learning_rate": 7.7933601428155e-05,
|
| 434 |
+
"loss": 3.2478,
|
| 435 |
"step": 35500
|
| 436 |
},
|
| 437 |
{
|
| 438 |
+
"epoch": 0.53,
|
| 439 |
+
"learning_rate": 7.78746719219441e-05,
|
| 440 |
+
"loss": 3.229,
|
| 441 |
"step": 36000
|
| 442 |
},
|
| 443 |
{
|
| 444 |
+
"epoch": 0.54,
|
| 445 |
+
"learning_rate": 7.781493683081388e-05,
|
| 446 |
+
"loss": 3.2293,
|
| 447 |
"step": 36500
|
| 448 |
},
|
| 449 |
{
|
| 450 |
+
"epoch": 0.54,
|
| 451 |
+
"learning_rate": 7.775439742531495e-05,
|
| 452 |
+
"loss": 3.1808,
|
| 453 |
"step": 37000
|
| 454 |
},
|
| 455 |
{
|
| 456 |
+
"epoch": 0.55,
|
| 457 |
+
"learning_rate": 7.769305499310553e-05,
|
| 458 |
+
"loss": 3.2062,
|
| 459 |
"step": 37500
|
| 460 |
},
|
| 461 |
{
|
| 462 |
+
"epoch": 0.56,
|
| 463 |
+
"learning_rate": 7.763091083892402e-05,
|
| 464 |
+
"loss": 3.1863,
|
| 465 |
"step": 38000
|
| 466 |
},
|
| 467 |
{
|
| 468 |
+
"epoch": 0.56,
|
| 469 |
+
"learning_rate": 7.756796628456121e-05,
|
| 470 |
+
"loss": 3.193,
|
| 471 |
"step": 38500
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 0.57,
|
| 475 |
+
"learning_rate": 7.750422266883222e-05,
|
| 476 |
+
"loss": 3.1244,
|
| 477 |
"step": 39000
|
| 478 |
},
|
| 479 |
{
|
| 480 |
+
"epoch": 0.58,
|
| 481 |
+
"learning_rate": 7.743968134754806e-05,
|
| 482 |
+
"loss": 3.273,
|
| 483 |
"step": 39500
|
| 484 |
},
|
| 485 |
{
|
| 486 |
+
"epoch": 0.59,
|
| 487 |
+
"learning_rate": 7.737434369348664e-05,
|
| 488 |
+
"loss": 3.2606,
|
| 489 |
"step": 40000
|
| 490 |
},
|
| 491 |
{
|
| 492 |
+
"epoch": 0.59,
|
| 493 |
+
"learning_rate": 7.730821109636379e-05,
|
| 494 |
+
"loss": 3.1016,
|
| 495 |
"step": 40500
|
| 496 |
},
|
| 497 |
{
|
| 498 |
+
"epoch": 0.6,
|
| 499 |
+
"learning_rate": 7.724128496280346e-05,
|
| 500 |
+
"loss": 3.2559,
|
| 501 |
"step": 41000
|
| 502 |
},
|
| 503 |
{
|
| 504 |
+
"epoch": 0.61,
|
| 505 |
+
"learning_rate": 7.717356671630802e-05,
|
| 506 |
+
"loss": 3.2853,
|
| 507 |
"step": 41500
|
| 508 |
},
|
| 509 |
{
|
| 510 |
+
"epoch": 0.62,
|
| 511 |
+
"learning_rate": 7.710505779722786e-05,
|
| 512 |
+
"loss": 3.1973,
|
| 513 |
"step": 42000
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 0.62,
|
| 517 |
+
"learning_rate": 7.703575966273073e-05,
|
| 518 |
+
"loss": 3.3002,
|
| 519 |
"step": 42500
|
| 520 |
},
|
| 521 |
{
|
| 522 |
+
"epoch": 0.63,
|
| 523 |
+
"learning_rate": 7.696567378677089e-05,
|
| 524 |
+
"loss": 3.1585,
|
| 525 |
"step": 43000
|
| 526 |
},
|
| 527 |
{
|
| 528 |
+
"epoch": 0.64,
|
| 529 |
+
"learning_rate": 7.689480166005756e-05,
|
| 530 |
+
"loss": 3.2222,
|
| 531 |
"step": 43500
|
| 532 |
},
|
| 533 |
{
|
| 534 |
+
"epoch": 0.64,
|
| 535 |
+
"learning_rate": 7.682314479002344e-05,
|
| 536 |
+
"loss": 3.1983,
|
| 537 |
"step": 44000
|
| 538 |
},
|
| 539 |
{
|
| 540 |
+
"epoch": 0.65,
|
| 541 |
+
"learning_rate": 7.67507047007924e-05,
|
| 542 |
+
"loss": 3.1874,
|
| 543 |
"step": 44500
|
| 544 |
},
|
| 545 |
{
|
| 546 |
+
"epoch": 0.66,
|
| 547 |
+
"learning_rate": 7.667748293314729e-05,
|
| 548 |
+
"loss": 3.2168,
|
| 549 |
"step": 45000
|
| 550 |
},
|
| 551 |
{
|
| 552 |
+
"epoch": 0.67,
|
| 553 |
+
"learning_rate": 7.6603481044497e-05,
|
| 554 |
+
"loss": 3.2015,
|
| 555 |
"step": 45500
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 0.67,
|
| 559 |
+
"learning_rate": 7.652870060884345e-05,
|
| 560 |
+
"loss": 3.2585,
|
| 561 |
"step": 46000
|
| 562 |
},
|
| 563 |
{
|
| 564 |
+
"epoch": 0.68,
|
| 565 |
+
"learning_rate": 7.645314321674803e-05,
|
| 566 |
+
"loss": 3.2109,
|
| 567 |
"step": 46500
|
| 568 |
},
|
| 569 |
{
|
| 570 |
+
"epoch": 0.69,
|
| 571 |
+
"learning_rate": 7.637681047529781e-05,
|
| 572 |
+
"loss": 3.1592,
|
| 573 |
"step": 47000
|
| 574 |
},
|
| 575 |
{
|
| 576 |
+
"epoch": 0.7,
|
| 577 |
+
"learning_rate": 7.629970400807136e-05,
|
| 578 |
+
"loss": 3.1815,
|
| 579 |
"step": 47500
|
| 580 |
},
|
| 581 |
{
|
| 582 |
+
"epoch": 0.7,
|
| 583 |
+
"learning_rate": 7.622182545510419e-05,
|
| 584 |
+
"loss": 3.1461,
|
| 585 |
"step": 48000
|
| 586 |
},
|
| 587 |
{
|
| 588 |
+
"epoch": 0.71,
|
| 589 |
+
"learning_rate": 7.61431764728539e-05,
|
| 590 |
+
"loss": 3.2464,
|
| 591 |
"step": 48500
|
| 592 |
},
|
| 593 |
{
|
| 594 |
+
"epoch": 0.72,
|
| 595 |
+
"learning_rate": 7.606375873416491e-05,
|
| 596 |
+
"loss": 3.1823,
|
| 597 |
"step": 49000
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 0.73,
|
| 601 |
+
"learning_rate": 7.598357392823292e-05,
|
| 602 |
+
"loss": 3.1304,
|
| 603 |
"step": 49500
|
| 604 |
},
|
| 605 |
{
|
| 606 |
+
"epoch": 0.73,
|
| 607 |
+
"learning_rate": 7.590262376056896e-05,
|
| 608 |
+
"loss": 3.1312,
|
| 609 |
"step": 50000
|
| 610 |
},
|
| 611 |
{
|
| 612 |
+
"epoch": 0.74,
|
| 613 |
+
"learning_rate": 7.58209099529631e-05,
|
| 614 |
+
"loss": 3.0755,
|
| 615 |
"step": 50500
|
| 616 |
},
|
| 617 |
{
|
| 618 |
+
"epoch": 0.75,
|
| 619 |
+
"learning_rate": 7.573843424344783e-05,
|
| 620 |
+
"loss": 3.1699,
|
| 621 |
+
"step": 51000
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 0.75,
|
| 625 |
+
"learning_rate": 7.565519838626113e-05,
|
| 626 |
+
"loss": 3.1478,
|
| 627 |
+
"step": 51500
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 0.76,
|
| 631 |
+
"learning_rate": 7.557120415180916e-05,
|
| 632 |
+
"loss": 3.1874,
|
| 633 |
+
"step": 52000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.77,
|
| 637 |
+
"learning_rate": 7.548645332662853e-05,
|
| 638 |
+
"loss": 3.2038,
|
| 639 |
+
"step": 52500
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.78,
|
| 643 |
+
"learning_rate": 7.540094771334835e-05,
|
| 644 |
+
"loss": 3.3039,
|
| 645 |
+
"step": 53000
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 0.78,
|
| 649 |
+
"learning_rate": 7.531468913065192e-05,
|
| 650 |
+
"loss": 3.3008,
|
| 651 |
+
"step": 53500
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 0.79,
|
| 655 |
+
"learning_rate": 7.522767941323798e-05,
|
| 656 |
+
"loss": 3.1248,
|
| 657 |
+
"step": 54000
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.8,
|
| 661 |
+
"learning_rate": 7.513992041178174e-05,
|
| 662 |
+
"loss": 3.1972,
|
| 663 |
+
"step": 54500
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 0.81,
|
| 667 |
+
"learning_rate": 7.505141399289549e-05,
|
| 668 |
+
"loss": 3.1768,
|
| 669 |
+
"step": 55000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.81,
|
| 673 |
+
"learning_rate": 7.496216203908891e-05,
|
| 674 |
+
"loss": 3.0954,
|
| 675 |
+
"step": 55500
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.82,
|
| 679 |
+
"learning_rate": 7.487216644872901e-05,
|
| 680 |
+
"loss": 3.1659,
|
| 681 |
+
"step": 56000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.83,
|
| 685 |
+
"learning_rate": 7.478142913599978e-05,
|
| 686 |
+
"loss": 3.2247,
|
| 687 |
+
"step": 56500
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 0.84,
|
| 691 |
+
"learning_rate": 7.468995203086146e-05,
|
| 692 |
+
"loss": 3.1733,
|
| 693 |
+
"step": 57000
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 0.84,
|
| 697 |
+
"learning_rate": 7.459773707900946e-05,
|
| 698 |
+
"loss": 3.2226,
|
| 699 |
+
"step": 57500
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 0.85,
|
| 703 |
+
"learning_rate": 7.450478624183306e-05,
|
| 704 |
+
"loss": 3.1898,
|
| 705 |
+
"step": 58000
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.86,
|
| 709 |
+
"learning_rate": 7.441110149637363e-05,
|
| 710 |
+
"loss": 3.2628,
|
| 711 |
+
"step": 58500
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.86,
|
| 715 |
+
"learning_rate": 7.431668483528254e-05,
|
| 716 |
+
"loss": 3.1605,
|
| 717 |
+
"step": 59000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.87,
|
| 721 |
+
"learning_rate": 7.422153826677887e-05,
|
| 722 |
+
"loss": 3.1439,
|
| 723 |
+
"step": 59500
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.88,
|
| 727 |
+
"learning_rate": 7.412566381460662e-05,
|
| 728 |
+
"loss": 3.177,
|
| 729 |
+
"step": 60000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.89,
|
| 733 |
+
"learning_rate": 7.402906351799175e-05,
|
| 734 |
+
"loss": 3.1037,
|
| 735 |
+
"step": 60500
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 0.89,
|
| 739 |
+
"learning_rate": 7.39317394315987e-05,
|
| 740 |
+
"loss": 3.2293,
|
| 741 |
+
"step": 61000
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 0.9,
|
| 745 |
+
"learning_rate": 7.383369362548674e-05,
|
| 746 |
+
"loss": 3.1304,
|
| 747 |
+
"step": 61500
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 0.91,
|
| 751 |
+
"learning_rate": 7.373492818506597e-05,
|
| 752 |
+
"loss": 3.1243,
|
| 753 |
+
"step": 62000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.92,
|
| 757 |
+
"learning_rate": 7.363544521105292e-05,
|
| 758 |
+
"loss": 3.1219,
|
| 759 |
+
"step": 62500
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.92,
|
| 763 |
+
"learning_rate": 7.353524681942585e-05,
|
| 764 |
+
"loss": 3.1183,
|
| 765 |
+
"step": 63000
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.93,
|
| 769 |
+
"learning_rate": 7.343433514137987e-05,
|
| 770 |
+
"loss": 3.2476,
|
| 771 |
+
"step": 63500
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 0.94,
|
| 775 |
+
"learning_rate": 7.333271232328141e-05,
|
| 776 |
+
"loss": 3.207,
|
| 777 |
+
"step": 64000
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 0.95,
|
| 781 |
+
"learning_rate": 7.32303805266227e-05,
|
| 782 |
+
"loss": 3.0767,
|
| 783 |
+
"step": 64500
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 0.95,
|
| 787 |
+
"learning_rate": 7.312734192797583e-05,
|
| 788 |
+
"loss": 3.1447,
|
| 789 |
+
"step": 65000
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.96,
|
| 793 |
+
"learning_rate": 7.302359871894635e-05,
|
| 794 |
+
"loss": 3.2073,
|
| 795 |
+
"step": 65500
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 0.97,
|
| 799 |
+
"learning_rate": 7.291915310612666e-05,
|
| 800 |
+
"loss": 3.1725,
|
| 801 |
+
"step": 66000
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.97,
|
| 805 |
+
"learning_rate": 7.281400731104918e-05,
|
| 806 |
+
"loss": 3.1812,
|
| 807 |
+
"step": 66500
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.98,
|
| 811 |
+
"learning_rate": 7.2708163570139e-05,
|
| 812 |
+
"loss": 3.1421,
|
| 813 |
+
"step": 67000
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 0.99,
|
| 817 |
+
"learning_rate": 7.260162413466636e-05,
|
| 818 |
+
"loss": 3.1253,
|
| 819 |
+
"step": 67500
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 1.0,
|
| 823 |
+
"learning_rate": 7.249439127069873e-05,
|
| 824 |
+
"loss": 3.1778,
|
| 825 |
+
"step": 68000
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 1.0,
|
| 829 |
"eval_bleu": 1.0,
|
| 830 |
"eval_brevity_penalty": 1.0,
|
| 831 |
"eval_length_ratio": 1.0,
|
| 832 |
+
"eval_loss": 3.0338802337646484,
|
| 833 |
"eval_precisions": [
|
| 834 |
1.0,
|
| 835 |
1.0,
|
| 836 |
1.0,
|
| 837 |
1.0
|
| 838 |
],
|
| 839 |
+
"eval_reference_length": 7761920,
|
| 840 |
+
"eval_runtime": 20729.6074,
|
| 841 |
+
"eval_samples_per_second": 0.731,
|
| 842 |
+
"eval_steps_per_second": 0.366,
|
| 843 |
+
"eval_translation_length": 7761920,
|
| 844 |
+
"step": 68219
|
| 845 |
}
|
| 846 |
],
|
| 847 |
"logging_steps": 500,
|
| 848 |
+
"max_steps": 341095,
|
| 849 |
+
"num_train_epochs": 5,
|
| 850 |
+
"save_steps": 5000,
|
| 851 |
+
"total_flos": 1.1708163520053903e+18,
|
| 852 |
"trial_name": null,
|
| 853 |
"trial_params": null
|
| 854 |
}
|
last-checkpoint/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4728
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6ca830d6d5d553a29fbd015e13f5b91f1ebe5cec81e6874f2564cd16cd029b0
|
| 3 |
size 4728
|