Upload folder using huggingface_hub
Browse files- adapter_model.safetensors +1 -1
- checkpoint-200/adapter_model.safetensors +1 -1
- checkpoint-200/optimizer.pt +1 -1
- checkpoint-200/rng_state.pth +1 -1
- checkpoint-200/scheduler.pt +1 -1
- checkpoint-200/trainer_state.json +790 -790
- checkpoint-200/training_args.bin +1 -1
- training_args.bin +1 -1
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 34100216
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a115dacf1f2435d487c24b16dcdf2e2ded0bc635269a44e344367ce2938d8594
|
| 3 |
size 34100216
|
checkpoint-200/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 34100216
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a115dacf1f2435d487c24b16dcdf2e2ded0bc635269a44e344367ce2938d8594
|
| 3 |
size 34100216
|
checkpoint-200/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 68292346
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f7f3eb5fe27f4339c2b2499f040822b2be66cf8314166a745797c2d8e5e499f8
|
| 3 |
size 68292346
|
checkpoint-200/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:f50f0b8cf8a8f5247d0e9729d14f2b46d5491d4b47ff4bafcc913f88950008b8
|
| 3 |
size 14244
|
checkpoint-200/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 sha256:112b68374ca1c950140f0870a98c4f3bc5a36bbea3cfec174b29f0bafea1fbee
|
| 3 |
size 1064
|
checkpoint-200/trainer_state.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 500,
|
| 6 |
"global_step": 200,
|
| 7 |
"is_hyper_param_search": false,
|
|
@@ -9,1403 +9,1403 @@
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch":
|
| 13 |
-
"grad_norm":
|
| 14 |
"learning_rate": 2.0000000000000003e-06,
|
| 15 |
-
"loss":
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch":
|
| 20 |
-
"grad_norm":
|
| 21 |
"learning_rate": 4.000000000000001e-06,
|
| 22 |
-
"loss":
|
| 23 |
"step": 2
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch":
|
| 27 |
-
"grad_norm":
|
| 28 |
"learning_rate": 6e-06,
|
| 29 |
-
"loss":
|
| 30 |
"step": 3
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch":
|
| 34 |
-
"grad_norm":
|
| 35 |
"learning_rate": 8.000000000000001e-06,
|
| 36 |
-
"loss": 1.
|
| 37 |
"step": 4
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
"learning_rate": 1e-05,
|
| 43 |
-
"loss":
|
| 44 |
"step": 5
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
"learning_rate": 1.2e-05,
|
| 50 |
-
"loss":
|
| 51 |
"step": 6
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
"learning_rate": 1.4000000000000001e-05,
|
| 57 |
-
"loss":
|
| 58 |
"step": 7
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
"learning_rate": 1.6000000000000003e-05,
|
| 64 |
-
"loss":
|
| 65 |
"step": 8
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
"learning_rate": 1.8e-05,
|
| 71 |
-
"loss":
|
| 72 |
"step": 9
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
"learning_rate": 2e-05,
|
| 78 |
-
"loss": 2.
|
| 79 |
"step": 10
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss": 2.
|
| 86 |
"step": 11
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate": 2.
|
| 92 |
-
"loss": 1.
|
| 93 |
"step": 12
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm":
|
| 98 |
-
"learning_rate": 2.
|
| 99 |
-
"loss": 2.
|
| 100 |
"step": 13
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm":
|
| 105 |
-
"learning_rate": 2.
|
| 106 |
-
"loss":
|
| 107 |
"step": 14
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm":
|
| 112 |
-
"learning_rate":
|
| 113 |
-
"loss":
|
| 114 |
"step": 15
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm":
|
| 119 |
-
"learning_rate":
|
| 120 |
-
"loss": 2.
|
| 121 |
"step": 16
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm":
|
| 126 |
-
"learning_rate":
|
| 127 |
-
"loss": 1.
|
| 128 |
"step": 17
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm":
|
| 133 |
-
"learning_rate": 3.
|
| 134 |
-
"loss": 2.
|
| 135 |
"step": 18
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm":
|
| 140 |
-
"learning_rate": 3.
|
| 141 |
-
"loss":
|
| 142 |
"step": 19
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm":
|
| 147 |
-
"learning_rate":
|
| 148 |
-
"loss":
|
| 149 |
"step": 20
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm":
|
| 154 |
-
"learning_rate":
|
| 155 |
-
"loss":
|
| 156 |
"step": 21
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm":
|
| 161 |
-
"learning_rate":
|
| 162 |
-
"loss":
|
| 163 |
"step": 22
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm":
|
| 168 |
-
"learning_rate": 4.
|
| 169 |
-
"loss": 1.
|
| 170 |
"step": 23
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"grad_norm":
|
| 175 |
-
"learning_rate": 4.
|
| 176 |
-
"loss":
|
| 177 |
"step": 24
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"grad_norm":
|
| 182 |
-
"learning_rate":
|
| 183 |
-
"loss":
|
| 184 |
"step": 25
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch": 0.
|
| 188 |
-
"grad_norm":
|
| 189 |
-
"learning_rate":
|
| 190 |
-
"loss": 2.
|
| 191 |
"step": 26
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch": 0.
|
| 195 |
-
"grad_norm":
|
| 196 |
-
"learning_rate":
|
| 197 |
-
"loss":
|
| 198 |
"step": 27
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch": 0.
|
| 202 |
-
"grad_norm":
|
| 203 |
-
"learning_rate":
|
| 204 |
-
"loss":
|
| 205 |
"step": 28
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch": 0.
|
| 209 |
-
"grad_norm":
|
| 210 |
-
"learning_rate":
|
| 211 |
-
"loss":
|
| 212 |
"step": 29
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"grad_norm":
|
| 217 |
-
"learning_rate":
|
| 218 |
-
"loss": 1.
|
| 219 |
"step": 30
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"grad_norm":
|
| 224 |
-
"learning_rate":
|
| 225 |
-
"loss":
|
| 226 |
"step": 31
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 0.
|
| 230 |
-
"grad_norm":
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss": 2.
|
| 233 |
"step": 32
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 0.
|
| 237 |
-
"grad_norm":
|
| 238 |
-
"learning_rate":
|
| 239 |
-
"loss":
|
| 240 |
"step": 33
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
-
"grad_norm":
|
| 245 |
-
"learning_rate":
|
| 246 |
-
"loss":
|
| 247 |
"step": 34
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch": 0.
|
| 251 |
-
"grad_norm":
|
| 252 |
-
"learning_rate":
|
| 253 |
-
"loss": 1.
|
| 254 |
"step": 35
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch": 0.
|
| 258 |
-
"grad_norm":
|
| 259 |
-
"learning_rate":
|
| 260 |
-
"loss":
|
| 261 |
"step": 36
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 0.
|
| 265 |
-
"grad_norm":
|
| 266 |
-
"learning_rate":
|
| 267 |
-
"loss":
|
| 268 |
"step": 37
|
| 269 |
},
|
| 270 |
{
|
| 271 |
-
"epoch": 0.
|
| 272 |
-
"grad_norm":
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"loss":
|
| 275 |
"step": 38
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 0.
|
| 279 |
-
"grad_norm":
|
| 280 |
-
"learning_rate":
|
| 281 |
-
"loss":
|
| 282 |
"step": 39
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch": 0.
|
| 286 |
-
"grad_norm":
|
| 287 |
-
"learning_rate":
|
| 288 |
-
"loss":
|
| 289 |
"step": 40
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch": 0.
|
| 293 |
-
"grad_norm":
|
| 294 |
-
"learning_rate":
|
| 295 |
-
"loss":
|
| 296 |
"step": 41
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch": 0.
|
| 300 |
-
"grad_norm":
|
| 301 |
-
"learning_rate":
|
| 302 |
-
"loss":
|
| 303 |
"step": 42
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 0.
|
| 307 |
-
"grad_norm":
|
| 308 |
-
"learning_rate":
|
| 309 |
-
"loss":
|
| 310 |
"step": 43
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"epoch": 0.
|
| 314 |
-
"grad_norm": 3.
|
| 315 |
-
"learning_rate":
|
| 316 |
-
"loss": 2.
|
| 317 |
"step": 44
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch": 0.
|
| 321 |
-
"grad_norm":
|
| 322 |
-
"learning_rate":
|
| 323 |
-
"loss": 1.
|
| 324 |
"step": 45
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch": 0.
|
| 328 |
-
"grad_norm":
|
| 329 |
-
"learning_rate":
|
| 330 |
-
"loss":
|
| 331 |
"step": 46
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch": 0.
|
| 335 |
-
"grad_norm":
|
| 336 |
-
"learning_rate":
|
| 337 |
-
"loss": 1.
|
| 338 |
"step": 47
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch": 0.
|
| 342 |
-
"grad_norm":
|
| 343 |
-
"learning_rate":
|
| 344 |
-
"loss":
|
| 345 |
"step": 48
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch": 0.
|
| 349 |
-
"grad_norm":
|
| 350 |
-
"learning_rate":
|
| 351 |
-
"loss":
|
| 352 |
"step": 49
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"epoch": 0.
|
| 356 |
-
"grad_norm":
|
| 357 |
-
"learning_rate":
|
| 358 |
-
"loss":
|
| 359 |
"step": 50
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 0.
|
| 363 |
-
"grad_norm":
|
| 364 |
-
"learning_rate":
|
| 365 |
-
"loss": 2.
|
| 366 |
"step": 51
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch": 0.
|
| 370 |
-
"grad_norm":
|
| 371 |
-
"learning_rate":
|
| 372 |
-
"loss":
|
| 373 |
"step": 52
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch": 0.
|
| 377 |
-
"grad_norm":
|
| 378 |
-
"learning_rate":
|
| 379 |
-
"loss":
|
| 380 |
"step": 53
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"epoch": 0.
|
| 384 |
-
"grad_norm":
|
| 385 |
-
"learning_rate": 0.
|
| 386 |
-
"loss":
|
| 387 |
"step": 54
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 0.
|
| 391 |
-
"grad_norm":
|
| 392 |
-
"learning_rate": 0.
|
| 393 |
-
"loss":
|
| 394 |
"step": 55
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch": 0.
|
| 398 |
-
"grad_norm":
|
| 399 |
-
"learning_rate": 0.
|
| 400 |
-
"loss": 1.
|
| 401 |
"step": 56
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 0.
|
| 405 |
-
"grad_norm":
|
| 406 |
-
"learning_rate": 0.
|
| 407 |
-
"loss":
|
| 408 |
"step": 57
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch": 0.
|
| 412 |
-
"grad_norm":
|
| 413 |
-
"learning_rate": 0.
|
| 414 |
-
"loss": 1.
|
| 415 |
"step": 58
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch": 0.
|
| 419 |
-
"grad_norm":
|
| 420 |
-
"learning_rate": 0.
|
| 421 |
-
"loss":
|
| 422 |
"step": 59
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch": 0.
|
| 426 |
-
"grad_norm":
|
| 427 |
-
"learning_rate": 0.
|
| 428 |
-
"loss": 1.
|
| 429 |
"step": 60
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 0.
|
| 433 |
-
"grad_norm":
|
| 434 |
-
"learning_rate": 0.
|
| 435 |
-
"loss":
|
| 436 |
"step": 61
|
| 437 |
},
|
| 438 |
{
|
| 439 |
-
"epoch": 0.
|
| 440 |
-
"grad_norm":
|
| 441 |
-
"learning_rate": 0.
|
| 442 |
-
"loss":
|
| 443 |
"step": 62
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 0.
|
| 447 |
-
"grad_norm":
|
| 448 |
-
"learning_rate": 0.
|
| 449 |
-
"loss":
|
| 450 |
"step": 63
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"epoch": 0.
|
| 454 |
-
"grad_norm":
|
| 455 |
-
"learning_rate": 0.
|
| 456 |
-
"loss":
|
| 457 |
"step": 64
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch": 0.
|
| 461 |
-
"grad_norm":
|
| 462 |
-
"learning_rate": 0.
|
| 463 |
-
"loss":
|
| 464 |
"step": 65
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch": 0.
|
| 468 |
-
"grad_norm":
|
| 469 |
-
"learning_rate": 0.
|
| 470 |
-
"loss": 1.
|
| 471 |
"step": 66
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 0.
|
| 475 |
-
"grad_norm":
|
| 476 |
-
"learning_rate": 0.
|
| 477 |
-
"loss":
|
| 478 |
"step": 67
|
| 479 |
},
|
| 480 |
{
|
| 481 |
-
"epoch": 0.
|
| 482 |
-
"grad_norm":
|
| 483 |
-
"learning_rate": 0.
|
| 484 |
-
"loss": 1.
|
| 485 |
"step": 68
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 0.
|
| 489 |
-
"grad_norm":
|
| 490 |
-
"learning_rate": 0.
|
| 491 |
-
"loss": 1.
|
| 492 |
"step": 69
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch": 0.
|
| 496 |
-
"grad_norm":
|
| 497 |
-
"learning_rate": 0.
|
| 498 |
-
"loss": 1.
|
| 499 |
"step": 70
|
| 500 |
},
|
| 501 |
{
|
| 502 |
-
"epoch": 0.
|
| 503 |
-
"grad_norm":
|
| 504 |
-
"learning_rate": 0.
|
| 505 |
-
"loss":
|
| 506 |
"step": 71
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch": 0.
|
| 510 |
-
"grad_norm":
|
| 511 |
-
"learning_rate": 0.
|
| 512 |
-
"loss":
|
| 513 |
"step": 72
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch": 0.
|
| 517 |
-
"grad_norm":
|
| 518 |
-
"learning_rate": 0.
|
| 519 |
-
"loss":
|
| 520 |
"step": 73
|
| 521 |
},
|
| 522 |
{
|
| 523 |
-
"epoch": 0.
|
| 524 |
-
"grad_norm":
|
| 525 |
-
"learning_rate": 0.
|
| 526 |
-
"loss": 1.
|
| 527 |
"step": 74
|
| 528 |
},
|
| 529 |
{
|
| 530 |
-
"epoch": 0.
|
| 531 |
-
"grad_norm":
|
| 532 |
-
"learning_rate": 0.
|
| 533 |
-
"loss": 1.
|
| 534 |
"step": 75
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch": 0.
|
| 538 |
-
"grad_norm":
|
| 539 |
-
"learning_rate": 0.
|
| 540 |
-
"loss":
|
| 541 |
"step": 76
|
| 542 |
},
|
| 543 |
{
|
| 544 |
-
"epoch": 0.
|
| 545 |
-
"grad_norm":
|
| 546 |
-
"learning_rate": 0.
|
| 547 |
-
"loss":
|
| 548 |
"step": 77
|
| 549 |
},
|
| 550 |
{
|
| 551 |
-
"epoch": 0.
|
| 552 |
-
"grad_norm":
|
| 553 |
-
"learning_rate": 0.
|
| 554 |
-
"loss": 1.
|
| 555 |
"step": 78
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"epoch": 0.
|
| 559 |
-
"grad_norm":
|
| 560 |
-
"learning_rate": 0.
|
| 561 |
-
"loss": 1.
|
| 562 |
"step": 79
|
| 563 |
},
|
| 564 |
{
|
| 565 |
-
"epoch": 0.
|
| 566 |
-
"grad_norm":
|
| 567 |
-
"learning_rate": 0.
|
| 568 |
-
"loss": 1.
|
| 569 |
"step": 80
|
| 570 |
},
|
| 571 |
{
|
| 572 |
-
"epoch": 0.
|
| 573 |
-
"grad_norm":
|
| 574 |
-
"learning_rate": 0.
|
| 575 |
-
"loss":
|
| 576 |
"step": 81
|
| 577 |
},
|
| 578 |
{
|
| 579 |
-
"epoch": 0.
|
| 580 |
-
"grad_norm": 4.
|
| 581 |
-
"learning_rate": 0.
|
| 582 |
-
"loss": 1.
|
| 583 |
"step": 82
|
| 584 |
},
|
| 585 |
{
|
| 586 |
-
"epoch": 0.
|
| 587 |
-
"grad_norm":
|
| 588 |
-
"learning_rate": 0.
|
| 589 |
-
"loss": 1.
|
| 590 |
"step": 83
|
| 591 |
},
|
| 592 |
{
|
| 593 |
-
"epoch": 0.
|
| 594 |
-
"grad_norm":
|
| 595 |
-
"learning_rate": 0.
|
| 596 |
-
"loss":
|
| 597 |
"step": 84
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch": 0.
|
| 601 |
-
"grad_norm": 7.
|
| 602 |
-
"learning_rate": 0.
|
| 603 |
-
"loss": 1.
|
| 604 |
"step": 85
|
| 605 |
},
|
| 606 |
{
|
| 607 |
-
"epoch": 0.
|
| 608 |
-
"grad_norm":
|
| 609 |
-
"learning_rate": 0.
|
| 610 |
-
"loss":
|
| 611 |
"step": 86
|
| 612 |
},
|
| 613 |
{
|
| 614 |
-
"epoch": 0.
|
| 615 |
-
"grad_norm":
|
| 616 |
-
"learning_rate": 0.
|
| 617 |
-
"loss": 1.
|
| 618 |
"step": 87
|
| 619 |
},
|
| 620 |
{
|
| 621 |
-
"epoch": 0.
|
| 622 |
-
"grad_norm":
|
| 623 |
-
"learning_rate": 0.
|
| 624 |
-
"loss": 1.
|
| 625 |
"step": 88
|
| 626 |
},
|
| 627 |
{
|
| 628 |
-
"epoch": 0.
|
| 629 |
-
"grad_norm": 1.
|
| 630 |
-
"learning_rate": 0.
|
| 631 |
-
"loss": 1.
|
| 632 |
"step": 89
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch": 0.
|
| 636 |
-
"grad_norm":
|
| 637 |
-
"learning_rate": 0.
|
| 638 |
-
"loss": 1.
|
| 639 |
"step": 90
|
| 640 |
},
|
| 641 |
{
|
| 642 |
-
"epoch": 0.
|
| 643 |
-
"grad_norm":
|
| 644 |
-
"learning_rate": 0.
|
| 645 |
-
"loss": 1.
|
| 646 |
"step": 91
|
| 647 |
},
|
| 648 |
{
|
| 649 |
-
"epoch": 0.
|
| 650 |
-
"grad_norm":
|
| 651 |
-
"learning_rate": 0.
|
| 652 |
-
"loss":
|
| 653 |
"step": 92
|
| 654 |
},
|
| 655 |
{
|
| 656 |
-
"epoch": 0.
|
| 657 |
-
"grad_norm":
|
| 658 |
-
"learning_rate": 0.
|
| 659 |
-
"loss": 1.
|
| 660 |
"step": 93
|
| 661 |
},
|
| 662 |
{
|
| 663 |
-
"epoch": 0.
|
| 664 |
-
"grad_norm":
|
| 665 |
-
"learning_rate": 0.
|
| 666 |
-
"loss":
|
| 667 |
"step": 94
|
| 668 |
},
|
| 669 |
{
|
| 670 |
-
"epoch": 0.
|
| 671 |
-
"grad_norm":
|
| 672 |
-
"learning_rate": 0.
|
| 673 |
-
"loss":
|
| 674 |
"step": 95
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch": 0.
|
| 678 |
-
"grad_norm":
|
| 679 |
-
"learning_rate": 0.
|
| 680 |
-
"loss": 1.
|
| 681 |
"step": 96
|
| 682 |
},
|
| 683 |
{
|
| 684 |
-
"epoch": 0.
|
| 685 |
-
"grad_norm":
|
| 686 |
-
"learning_rate": 0.
|
| 687 |
-
"loss":
|
| 688 |
"step": 97
|
| 689 |
},
|
| 690 |
{
|
| 691 |
-
"epoch": 0.
|
| 692 |
-
"grad_norm":
|
| 693 |
-
"learning_rate": 0.
|
| 694 |
-
"loss": 1.
|
| 695 |
"step": 98
|
| 696 |
},
|
| 697 |
{
|
| 698 |
-
"epoch": 0.
|
| 699 |
-
"grad_norm":
|
| 700 |
-
"learning_rate": 0.
|
| 701 |
-
"loss": 1.
|
| 702 |
"step": 99
|
| 703 |
},
|
| 704 |
{
|
| 705 |
-
"epoch": 0.
|
| 706 |
-
"grad_norm":
|
| 707 |
-
"learning_rate": 0.
|
| 708 |
-
"loss": 1.
|
| 709 |
"step": 100
|
| 710 |
},
|
| 711 |
{
|
| 712 |
-
"epoch": 0.
|
| 713 |
-
"grad_norm":
|
| 714 |
-
"learning_rate": 0.
|
| 715 |
-
"loss":
|
| 716 |
"step": 101
|
| 717 |
},
|
| 718 |
{
|
| 719 |
-
"epoch": 0.
|
| 720 |
-
"grad_norm":
|
| 721 |
-
"learning_rate": 0.
|
| 722 |
-
"loss":
|
| 723 |
"step": 102
|
| 724 |
},
|
| 725 |
{
|
| 726 |
-
"epoch": 0.
|
| 727 |
-
"grad_norm":
|
| 728 |
"learning_rate": 0.00019800000000000002,
|
| 729 |
-
"loss":
|
| 730 |
"step": 103
|
| 731 |
},
|
| 732 |
{
|
| 733 |
-
"epoch": 0.
|
| 734 |
-
"grad_norm":
|
| 735 |
-
"learning_rate": 0.
|
| 736 |
-
"loss": 1.
|
| 737 |
"step": 104
|
| 738 |
},
|
| 739 |
{
|
| 740 |
-
"epoch": 0.
|
| 741 |
-
"grad_norm":
|
| 742 |
-
"learning_rate": 0.
|
| 743 |
-
"loss": 1.
|
| 744 |
"step": 105
|
| 745 |
},
|
| 746 |
{
|
| 747 |
-
"epoch": 0.
|
| 748 |
-
"grad_norm":
|
| 749 |
-
"learning_rate": 0.
|
| 750 |
-
"loss": 1.
|
| 751 |
"step": 106
|
| 752 |
},
|
| 753 |
{
|
| 754 |
-
"epoch": 0.
|
| 755 |
-
"grad_norm": 5.
|
| 756 |
-
"learning_rate": 0.
|
| 757 |
-
"loss": 1.
|
| 758 |
"step": 107
|
| 759 |
},
|
| 760 |
{
|
| 761 |
-
"epoch": 0.
|
| 762 |
-
"grad_norm":
|
| 763 |
-
"learning_rate": 0.
|
| 764 |
-
"loss":
|
| 765 |
"step": 108
|
| 766 |
},
|
| 767 |
{
|
| 768 |
-
"epoch": 0.
|
| 769 |
-
"grad_norm":
|
| 770 |
-
"learning_rate": 0.
|
| 771 |
"loss": 1.4454,
|
| 772 |
"step": 109
|
| 773 |
},
|
| 774 |
{
|
| 775 |
-
"epoch": 0.
|
| 776 |
-
"grad_norm":
|
| 777 |
-
"learning_rate": 0.
|
| 778 |
-
"loss": 1.
|
| 779 |
"step": 110
|
| 780 |
},
|
| 781 |
{
|
| 782 |
-
"epoch": 0.
|
| 783 |
-
"grad_norm":
|
| 784 |
-
"learning_rate": 0.
|
| 785 |
-
"loss": 1.
|
| 786 |
"step": 111
|
| 787 |
},
|
| 788 |
{
|
| 789 |
-
"epoch": 0.
|
| 790 |
-
"grad_norm":
|
| 791 |
-
"learning_rate": 0.
|
| 792 |
-
"loss": 1.
|
| 793 |
"step": 112
|
| 794 |
},
|
| 795 |
{
|
| 796 |
-
"epoch": 0.
|
| 797 |
-
"grad_norm":
|
| 798 |
-
"learning_rate": 0.
|
| 799 |
-
"loss": 1.
|
| 800 |
"step": 113
|
| 801 |
},
|
| 802 |
{
|
| 803 |
-
"epoch": 0.
|
| 804 |
-
"grad_norm":
|
| 805 |
-
"learning_rate": 0.
|
| 806 |
-
"loss":
|
| 807 |
"step": 114
|
| 808 |
},
|
| 809 |
{
|
| 810 |
-
"epoch": 0.
|
| 811 |
-
"grad_norm":
|
| 812 |
-
"learning_rate": 0.
|
| 813 |
-
"loss": 1.
|
| 814 |
"step": 115
|
| 815 |
},
|
| 816 |
{
|
| 817 |
-
"epoch": 0.
|
| 818 |
-
"grad_norm":
|
| 819 |
-
"learning_rate": 0.
|
| 820 |
-
"loss": 1.
|
| 821 |
"step": 116
|
| 822 |
},
|
| 823 |
{
|
| 824 |
-
"epoch": 0.
|
| 825 |
-
"grad_norm":
|
| 826 |
-
"learning_rate": 0.
|
| 827 |
-
"loss":
|
| 828 |
"step": 117
|
| 829 |
},
|
| 830 |
{
|
| 831 |
-
"epoch": 0.
|
| 832 |
-
"grad_norm":
|
| 833 |
-
"learning_rate": 0.
|
| 834 |
-
"loss": 1.
|
| 835 |
"step": 118
|
| 836 |
},
|
| 837 |
{
|
| 838 |
-
"epoch": 0.
|
| 839 |
-
"grad_norm":
|
| 840 |
-
"learning_rate": 0.
|
| 841 |
-
"loss": 1.
|
| 842 |
"step": 119
|
| 843 |
},
|
| 844 |
{
|
| 845 |
-
"epoch": 0.
|
| 846 |
-
"grad_norm":
|
| 847 |
-
"learning_rate": 0.
|
| 848 |
-
"loss":
|
| 849 |
"step": 120
|
| 850 |
},
|
| 851 |
{
|
| 852 |
-
"epoch": 0.
|
| 853 |
-
"grad_norm":
|
| 854 |
-
"learning_rate": 0.
|
| 855 |
-
"loss":
|
| 856 |
"step": 121
|
| 857 |
},
|
| 858 |
{
|
| 859 |
-
"epoch": 0.
|
| 860 |
-
"grad_norm":
|
| 861 |
-
"learning_rate": 0.
|
| 862 |
-
"loss":
|
| 863 |
"step": 122
|
| 864 |
},
|
| 865 |
{
|
| 866 |
-
"epoch": 0.
|
| 867 |
-
"grad_norm":
|
| 868 |
-
"learning_rate": 0.
|
| 869 |
-
"loss":
|
| 870 |
"step": 123
|
| 871 |
},
|
| 872 |
{
|
| 873 |
-
"epoch": 0.
|
| 874 |
-
"grad_norm":
|
| 875 |
-
"learning_rate": 0.
|
| 876 |
-
"loss": 1.
|
| 877 |
"step": 124
|
| 878 |
},
|
| 879 |
{
|
| 880 |
-
"epoch": 0.
|
| 881 |
-
"grad_norm":
|
| 882 |
-
"learning_rate": 0.
|
| 883 |
-
"loss": 1.
|
| 884 |
"step": 125
|
| 885 |
},
|
| 886 |
{
|
| 887 |
-
"epoch": 0.
|
| 888 |
-
"grad_norm":
|
| 889 |
-
"learning_rate": 0.
|
| 890 |
-
"loss": 1.
|
| 891 |
"step": 126
|
| 892 |
},
|
| 893 |
{
|
| 894 |
-
"epoch": 0.
|
| 895 |
-
"grad_norm":
|
| 896 |
-
"learning_rate": 0.
|
| 897 |
-
"loss": 1.
|
| 898 |
"step": 127
|
| 899 |
},
|
| 900 |
{
|
| 901 |
-
"epoch": 0.
|
| 902 |
-
"grad_norm":
|
| 903 |
-
"learning_rate": 0.
|
| 904 |
-
"loss": 1.
|
| 905 |
"step": 128
|
| 906 |
},
|
| 907 |
{
|
| 908 |
-
"epoch": 0.
|
| 909 |
-
"grad_norm":
|
| 910 |
-
"learning_rate": 0.
|
| 911 |
-
"loss": 1.
|
| 912 |
"step": 129
|
| 913 |
},
|
| 914 |
{
|
| 915 |
-
"epoch": 0.
|
| 916 |
-
"grad_norm":
|
| 917 |
-
"learning_rate": 0.
|
| 918 |
-
"loss": 1.
|
| 919 |
"step": 130
|
| 920 |
},
|
| 921 |
{
|
| 922 |
-
"epoch": 0.
|
| 923 |
-
"grad_norm":
|
| 924 |
-
"learning_rate": 0.
|
| 925 |
-
"loss": 1.
|
| 926 |
"step": 131
|
| 927 |
},
|
| 928 |
{
|
| 929 |
-
"epoch": 0.
|
| 930 |
-
"grad_norm":
|
| 931 |
-
"learning_rate": 0.
|
| 932 |
-
"loss": 1.
|
| 933 |
"step": 132
|
| 934 |
},
|
| 935 |
{
|
| 936 |
-
"epoch": 0.
|
| 937 |
-
"grad_norm":
|
| 938 |
-
"learning_rate": 0.
|
| 939 |
-
"loss":
|
| 940 |
"step": 133
|
| 941 |
},
|
| 942 |
{
|
| 943 |
-
"epoch": 0.
|
| 944 |
-
"grad_norm":
|
| 945 |
-
"learning_rate": 0.
|
| 946 |
-
"loss":
|
| 947 |
"step": 134
|
| 948 |
},
|
| 949 |
{
|
| 950 |
-
"epoch": 0.
|
| 951 |
-
"grad_norm":
|
| 952 |
-
"learning_rate": 0.
|
| 953 |
-
"loss": 1.
|
| 954 |
"step": 135
|
| 955 |
},
|
| 956 |
{
|
| 957 |
-
"epoch": 0.
|
| 958 |
-
"grad_norm":
|
| 959 |
-
"learning_rate": 0.
|
| 960 |
-
"loss":
|
| 961 |
"step": 136
|
| 962 |
},
|
| 963 |
{
|
| 964 |
-
"epoch": 0.
|
| 965 |
-
"grad_norm":
|
| 966 |
-
"learning_rate": 0.
|
| 967 |
-
"loss": 1.
|
| 968 |
"step": 137
|
| 969 |
},
|
| 970 |
{
|
| 971 |
-
"epoch": 0.
|
| 972 |
-
"grad_norm":
|
| 973 |
-
"learning_rate": 0.
|
| 974 |
-
"loss": 1.
|
| 975 |
"step": 138
|
| 976 |
},
|
| 977 |
{
|
| 978 |
-
"epoch": 0.
|
| 979 |
-
"grad_norm":
|
| 980 |
-
"learning_rate": 0.
|
| 981 |
-
"loss": 1.
|
| 982 |
"step": 139
|
| 983 |
},
|
| 984 |
{
|
| 985 |
-
"epoch": 0.
|
| 986 |
-
"grad_norm":
|
| 987 |
-
"learning_rate": 0.
|
| 988 |
-
"loss": 1.
|
| 989 |
"step": 140
|
| 990 |
},
|
| 991 |
{
|
| 992 |
-
"epoch": 0.
|
| 993 |
-
"grad_norm":
|
| 994 |
-
"learning_rate": 0.
|
| 995 |
-
"loss": 1.
|
| 996 |
"step": 141
|
| 997 |
},
|
| 998 |
{
|
| 999 |
-
"epoch": 0.
|
| 1000 |
-
"grad_norm":
|
| 1001 |
-
"learning_rate": 0.
|
| 1002 |
-
"loss": 1.
|
| 1003 |
"step": 142
|
| 1004 |
},
|
| 1005 |
{
|
| 1006 |
-
"epoch": 0.
|
| 1007 |
-
"grad_norm":
|
| 1008 |
-
"learning_rate": 0.
|
| 1009 |
-
"loss": 1.
|
| 1010 |
"step": 143
|
| 1011 |
},
|
| 1012 |
{
|
| 1013 |
-
"epoch": 0.
|
| 1014 |
-
"grad_norm":
|
| 1015 |
-
"learning_rate": 0.
|
| 1016 |
-
"loss": 1.
|
| 1017 |
"step": 144
|
| 1018 |
},
|
| 1019 |
{
|
| 1020 |
-
"epoch": 0.
|
| 1021 |
-
"grad_norm":
|
| 1022 |
-
"learning_rate": 0.
|
| 1023 |
-
"loss": 0.
|
| 1024 |
"step": 145
|
| 1025 |
},
|
| 1026 |
{
|
| 1027 |
-
"epoch": 0.
|
| 1028 |
-
"grad_norm":
|
| 1029 |
-
"learning_rate": 0.
|
| 1030 |
-
"loss": 1.
|
| 1031 |
"step": 146
|
| 1032 |
},
|
| 1033 |
{
|
| 1034 |
-
"epoch": 0.
|
| 1035 |
-
"grad_norm":
|
| 1036 |
-
"learning_rate": 0.
|
| 1037 |
-
"loss": 1.
|
| 1038 |
"step": 147
|
| 1039 |
},
|
| 1040 |
{
|
| 1041 |
-
"epoch": 0.
|
| 1042 |
-
"grad_norm":
|
| 1043 |
-
"learning_rate": 0.
|
| 1044 |
-
"loss":
|
| 1045 |
"step": 148
|
| 1046 |
},
|
| 1047 |
{
|
| 1048 |
-
"epoch": 0.
|
| 1049 |
-
"grad_norm":
|
| 1050 |
-
"learning_rate": 0.
|
| 1051 |
-
"loss":
|
| 1052 |
"step": 149
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
-
"epoch": 0.
|
| 1056 |
-
"grad_norm":
|
| 1057 |
-
"learning_rate": 0.
|
| 1058 |
-
"loss":
|
| 1059 |
"step": 150
|
| 1060 |
},
|
| 1061 |
{
|
| 1062 |
-
"epoch": 0.
|
| 1063 |
-
"grad_norm":
|
| 1064 |
-
"learning_rate": 0.
|
| 1065 |
-
"loss":
|
| 1066 |
"step": 151
|
| 1067 |
},
|
| 1068 |
{
|
| 1069 |
-
"epoch": 0.
|
| 1070 |
-
"grad_norm":
|
| 1071 |
-
"learning_rate": 0.
|
| 1072 |
-
"loss": 1.
|
| 1073 |
"step": 152
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
-
"epoch": 0.
|
| 1077 |
-
"grad_norm":
|
| 1078 |
-
"learning_rate":
|
| 1079 |
-
"loss": 1.
|
| 1080 |
"step": 153
|
| 1081 |
},
|
| 1082 |
{
|
| 1083 |
-
"epoch": 0.
|
| 1084 |
-
"grad_norm":
|
| 1085 |
-
"learning_rate":
|
| 1086 |
-
"loss": 1.
|
| 1087 |
"step": 154
|
| 1088 |
},
|
| 1089 |
{
|
| 1090 |
-
"epoch": 0.
|
| 1091 |
-
"grad_norm":
|
| 1092 |
-
"learning_rate": 9.
|
| 1093 |
-
"loss":
|
| 1094 |
"step": 155
|
| 1095 |
},
|
| 1096 |
{
|
| 1097 |
-
"epoch": 0.
|
| 1098 |
-
"grad_norm":
|
| 1099 |
-
"learning_rate": 9.
|
| 1100 |
-
"loss": 1.
|
| 1101 |
"step": 156
|
| 1102 |
},
|
| 1103 |
{
|
| 1104 |
-
"epoch": 0.
|
| 1105 |
-
"grad_norm":
|
| 1106 |
-
"learning_rate":
|
| 1107 |
-
"loss": 1.
|
| 1108 |
"step": 157
|
| 1109 |
},
|
| 1110 |
{
|
| 1111 |
-
"epoch": 0.
|
| 1112 |
-
"grad_norm":
|
| 1113 |
-
"learning_rate":
|
| 1114 |
-
"loss": 1.
|
| 1115 |
"step": 158
|
| 1116 |
},
|
| 1117 |
{
|
| 1118 |
-
"epoch": 0.
|
| 1119 |
-
"grad_norm":
|
| 1120 |
-
"learning_rate":
|
| 1121 |
-
"loss": 1.
|
| 1122 |
"step": 159
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
-
"epoch": 0.
|
| 1126 |
-
"grad_norm":
|
| 1127 |
-
"learning_rate": 8.
|
| 1128 |
-
"loss": 1.
|
| 1129 |
"step": 160
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
-
"epoch": 0.
|
| 1133 |
-
"grad_norm":
|
| 1134 |
-
"learning_rate": 8.
|
| 1135 |
-
"loss": 1.
|
| 1136 |
"step": 161
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
-
"epoch": 0.
|
| 1140 |
-
"grad_norm":
|
| 1141 |
-
"learning_rate":
|
| 1142 |
-
"loss": 1.
|
| 1143 |
"step": 162
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
-
"epoch": 0.
|
| 1147 |
-
"grad_norm":
|
| 1148 |
-
"learning_rate":
|
| 1149 |
-
"loss": 1.
|
| 1150 |
"step": 163
|
| 1151 |
},
|
| 1152 |
{
|
| 1153 |
-
"epoch": 0.
|
| 1154 |
-
"grad_norm":
|
| 1155 |
-
"learning_rate":
|
| 1156 |
-
"loss": 1.
|
| 1157 |
"step": 164
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
-
"epoch": 0.
|
| 1161 |
-
"grad_norm":
|
| 1162 |
-
"learning_rate": 7.
|
| 1163 |
-
"loss": 1.
|
| 1164 |
"step": 165
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
-
"epoch": 0.
|
| 1168 |
-
"grad_norm":
|
| 1169 |
-
"learning_rate": 7.
|
| 1170 |
-
"loss": 1.
|
| 1171 |
"step": 166
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
-
"epoch": 0.
|
| 1175 |
-
"grad_norm":
|
| 1176 |
-
"learning_rate":
|
| 1177 |
-
"loss": 1.
|
| 1178 |
"step": 167
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
-
"epoch": 0.
|
| 1182 |
-
"grad_norm":
|
| 1183 |
-
"learning_rate":
|
| 1184 |
-
"loss": 1.
|
| 1185 |
"step": 168
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
-
"epoch": 0.
|
| 1189 |
-
"grad_norm":
|
| 1190 |
-
"learning_rate":
|
| 1191 |
-
"loss":
|
| 1192 |
"step": 169
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
-
"epoch": 0.
|
| 1196 |
-
"grad_norm":
|
| 1197 |
-
"learning_rate": 6.
|
| 1198 |
-
"loss": 1.
|
| 1199 |
"step": 170
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
-
"epoch": 0.
|
| 1203 |
-
"grad_norm":
|
| 1204 |
-
"learning_rate": 6.
|
| 1205 |
-
"loss":
|
| 1206 |
"step": 171
|
| 1207 |
},
|
| 1208 |
{
|
| 1209 |
-
"epoch": 0.
|
| 1210 |
-
"grad_norm":
|
| 1211 |
-
"learning_rate":
|
| 1212 |
-
"loss": 1.
|
| 1213 |
"step": 172
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
-
"epoch": 0.
|
| 1217 |
-
"grad_norm":
|
| 1218 |
-
"learning_rate":
|
| 1219 |
-
"loss": 1.
|
| 1220 |
"step": 173
|
| 1221 |
},
|
| 1222 |
{
|
| 1223 |
-
"epoch": 0.
|
| 1224 |
-
"grad_norm":
|
| 1225 |
-
"learning_rate":
|
| 1226 |
-
"loss": 1.
|
| 1227 |
"step": 174
|
| 1228 |
},
|
| 1229 |
{
|
| 1230 |
-
"epoch": 0.
|
| 1231 |
-
"grad_norm":
|
| 1232 |
-
"learning_rate": 5.
|
| 1233 |
-
"loss": 1.
|
| 1234 |
"step": 175
|
| 1235 |
},
|
| 1236 |
{
|
| 1237 |
-
"epoch": 0.
|
| 1238 |
-
"grad_norm":
|
| 1239 |
-
"learning_rate": 5.
|
| 1240 |
-
"loss": 1.
|
| 1241 |
"step": 176
|
| 1242 |
},
|
| 1243 |
{
|
| 1244 |
-
"epoch": 0.
|
| 1245 |
-
"grad_norm":
|
| 1246 |
-
"learning_rate":
|
| 1247 |
-
"loss":
|
| 1248 |
"step": 177
|
| 1249 |
},
|
| 1250 |
{
|
| 1251 |
-
"epoch": 0.
|
| 1252 |
-
"grad_norm":
|
| 1253 |
-
"learning_rate":
|
| 1254 |
-
"loss":
|
| 1255 |
"step": 178
|
| 1256 |
},
|
| 1257 |
{
|
| 1258 |
-
"epoch": 0.
|
| 1259 |
-
"grad_norm":
|
| 1260 |
-
"learning_rate":
|
| 1261 |
-
"loss":
|
| 1262 |
"step": 179
|
| 1263 |
},
|
| 1264 |
{
|
| 1265 |
-
"epoch": 0.
|
| 1266 |
-
"grad_norm":
|
| 1267 |
-
"learning_rate": 4.
|
| 1268 |
-
"loss": 1.
|
| 1269 |
"step": 180
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
-
"epoch": 0.
|
| 1273 |
-
"grad_norm":
|
| 1274 |
-
"learning_rate": 4.
|
| 1275 |
-
"loss": 1.
|
| 1276 |
"step": 181
|
| 1277 |
},
|
| 1278 |
{
|
| 1279 |
-
"epoch": 0.
|
| 1280 |
-
"grad_norm":
|
| 1281 |
-
"learning_rate":
|
| 1282 |
-
"loss":
|
| 1283 |
"step": 182
|
| 1284 |
},
|
| 1285 |
{
|
| 1286 |
-
"epoch": 0.
|
| 1287 |
-
"grad_norm":
|
| 1288 |
-
"learning_rate":
|
| 1289 |
-
"loss": 1.
|
| 1290 |
"step": 183
|
| 1291 |
},
|
| 1292 |
{
|
| 1293 |
-
"epoch": 0.
|
| 1294 |
-
"grad_norm":
|
| 1295 |
-
"learning_rate":
|
| 1296 |
-
"loss": 1.
|
| 1297 |
"step": 184
|
| 1298 |
},
|
| 1299 |
{
|
| 1300 |
-
"epoch": 0.
|
| 1301 |
-
"grad_norm":
|
| 1302 |
-
"learning_rate": 3.
|
| 1303 |
-
"loss": 1.
|
| 1304 |
"step": 185
|
| 1305 |
},
|
| 1306 |
{
|
| 1307 |
-
"epoch": 0.
|
| 1308 |
-
"grad_norm":
|
| 1309 |
-
"learning_rate": 3.
|
| 1310 |
-
"loss": 1.
|
| 1311 |
"step": 186
|
| 1312 |
},
|
| 1313 |
{
|
| 1314 |
-
"epoch": 0.
|
| 1315 |
-
"grad_norm":
|
| 1316 |
-
"learning_rate":
|
| 1317 |
-
"loss": 1.
|
| 1318 |
"step": 187
|
| 1319 |
},
|
| 1320 |
{
|
| 1321 |
-
"epoch": 0.
|
| 1322 |
-
"grad_norm":
|
| 1323 |
-
"learning_rate":
|
| 1324 |
-
"loss": 1.
|
| 1325 |
"step": 188
|
| 1326 |
},
|
| 1327 |
{
|
| 1328 |
-
"epoch": 0.
|
| 1329 |
-
"grad_norm":
|
| 1330 |
-
"learning_rate":
|
| 1331 |
-
"loss": 1.
|
| 1332 |
"step": 189
|
| 1333 |
},
|
| 1334 |
{
|
| 1335 |
-
"epoch": 0.
|
| 1336 |
-
"grad_norm":
|
| 1337 |
-
"learning_rate": 2.
|
| 1338 |
-
"loss": 1.
|
| 1339 |
"step": 190
|
| 1340 |
},
|
| 1341 |
{
|
| 1342 |
-
"epoch": 0.
|
| 1343 |
-
"grad_norm":
|
| 1344 |
-
"learning_rate": 2.
|
| 1345 |
-
"loss": 1.
|
| 1346 |
"step": 191
|
| 1347 |
},
|
| 1348 |
{
|
| 1349 |
-
"epoch": 0.
|
| 1350 |
-
"grad_norm":
|
| 1351 |
-
"learning_rate":
|
| 1352 |
-
"loss": 1.
|
| 1353 |
"step": 192
|
| 1354 |
},
|
| 1355 |
{
|
| 1356 |
-
"epoch": 0.
|
| 1357 |
-
"grad_norm": 1.
|
| 1358 |
-
"learning_rate":
|
| 1359 |
-
"loss":
|
| 1360 |
"step": 193
|
| 1361 |
},
|
| 1362 |
{
|
| 1363 |
-
"epoch": 0.
|
| 1364 |
-
"grad_norm":
|
| 1365 |
-
"learning_rate":
|
| 1366 |
-
"loss": 1.
|
| 1367 |
"step": 194
|
| 1368 |
},
|
| 1369 |
{
|
| 1370 |
-
"epoch": 0.
|
| 1371 |
-
"grad_norm":
|
| 1372 |
-
"learning_rate": 1.
|
| 1373 |
-
"loss": 1.
|
| 1374 |
"step": 195
|
| 1375 |
},
|
| 1376 |
{
|
| 1377 |
-
"epoch": 0.
|
| 1378 |
-
"grad_norm":
|
| 1379 |
-
"learning_rate": 1.
|
| 1380 |
-
"loss":
|
| 1381 |
"step": 196
|
| 1382 |
},
|
| 1383 |
{
|
| 1384 |
-
"epoch": 0.
|
| 1385 |
-
"grad_norm":
|
| 1386 |
-
"learning_rate":
|
| 1387 |
-
"loss": 1.
|
| 1388 |
"step": 197
|
| 1389 |
},
|
| 1390 |
{
|
| 1391 |
-
"epoch": 0.
|
| 1392 |
-
"grad_norm":
|
| 1393 |
-
"learning_rate":
|
| 1394 |
-
"loss": 1.
|
| 1395 |
"step": 198
|
| 1396 |
},
|
| 1397 |
{
|
| 1398 |
-
"epoch": 0.
|
| 1399 |
-
"grad_norm":
|
| 1400 |
-
"learning_rate":
|
| 1401 |
-
"loss": 1.
|
| 1402 |
"step": 199
|
| 1403 |
},
|
| 1404 |
{
|
| 1405 |
-
"epoch": 0.
|
| 1406 |
-
"grad_norm":
|
| 1407 |
-
"learning_rate":
|
| 1408 |
-
"loss": 1.
|
| 1409 |
"step": 200
|
| 1410 |
}
|
| 1411 |
],
|
|
@@ -1426,7 +1426,7 @@
|
|
| 1426 |
"attributes": {}
|
| 1427 |
}
|
| 1428 |
},
|
| 1429 |
-
"total_flos":
|
| 1430 |
"train_batch_size": 1,
|
| 1431 |
"trial_name": null,
|
| 1432 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.3189792663476874,
|
| 5 |
"eval_steps": 500,
|
| 6 |
"global_step": 200,
|
| 7 |
"is_hyper_param_search": false,
|
|
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.001594896331738437,
|
| 13 |
+
"grad_norm": 9.529223442077637,
|
| 14 |
"learning_rate": 2.0000000000000003e-06,
|
| 15 |
+
"loss": 3.8964,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.003189792663476874,
|
| 20 |
+
"grad_norm": 3.4800214767456055,
|
| 21 |
"learning_rate": 4.000000000000001e-06,
|
| 22 |
+
"loss": 2.2247,
|
| 23 |
"step": 2
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.004784688995215311,
|
| 27 |
+
"grad_norm": 3.026850461959839,
|
| 28 |
"learning_rate": 6e-06,
|
| 29 |
+
"loss": 2.2337,
|
| 30 |
"step": 3
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.006379585326953748,
|
| 34 |
+
"grad_norm": 7.065993785858154,
|
| 35 |
"learning_rate": 8.000000000000001e-06,
|
| 36 |
+
"loss": 1.9733,
|
| 37 |
"step": 4
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.007974481658692184,
|
| 41 |
+
"grad_norm": 5.338789463043213,
|
| 42 |
"learning_rate": 1e-05,
|
| 43 |
+
"loss": 2.2808,
|
| 44 |
"step": 5
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.009569377990430622,
|
| 48 |
+
"grad_norm": 12.933262825012207,
|
| 49 |
"learning_rate": 1.2e-05,
|
| 50 |
+
"loss": 2.2984,
|
| 51 |
"step": 6
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.011164274322169059,
|
| 55 |
+
"grad_norm": 3.535792350769043,
|
| 56 |
"learning_rate": 1.4000000000000001e-05,
|
| 57 |
+
"loss": 2.6236,
|
| 58 |
"step": 7
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.012759170653907496,
|
| 62 |
+
"grad_norm": 7.564616680145264,
|
| 63 |
"learning_rate": 1.6000000000000003e-05,
|
| 64 |
+
"loss": 2.4849,
|
| 65 |
"step": 8
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.014354066985645933,
|
| 69 |
+
"grad_norm": 7.379055976867676,
|
| 70 |
"learning_rate": 1.8e-05,
|
| 71 |
+
"loss": 2.9017,
|
| 72 |
"step": 9
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.01594896331738437,
|
| 76 |
+
"grad_norm": 6.781640529632568,
|
| 77 |
"learning_rate": 2e-05,
|
| 78 |
+
"loss": 2.5836,
|
| 79 |
"step": 10
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.017543859649122806,
|
| 83 |
+
"grad_norm": Infinity,
|
| 84 |
+
"learning_rate": 2e-05,
|
| 85 |
+
"loss": 2.6482,
|
| 86 |
"step": 11
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.019138755980861243,
|
| 90 |
+
"grad_norm": 9.666836738586426,
|
| 91 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 92 |
+
"loss": 1.5775,
|
| 93 |
"step": 12
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.02073365231259968,
|
| 97 |
+
"grad_norm": 7.910143852233887,
|
| 98 |
+
"learning_rate": 2.4e-05,
|
| 99 |
+
"loss": 2.259,
|
| 100 |
"step": 13
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.022328548644338118,
|
| 104 |
+
"grad_norm": 5.05098295211792,
|
| 105 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 106 |
+
"loss": 2.1146,
|
| 107 |
"step": 14
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.023923444976076555,
|
| 111 |
+
"grad_norm": 4.457101345062256,
|
| 112 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 113 |
+
"loss": 2.1058,
|
| 114 |
"step": 15
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.025518341307814992,
|
| 118 |
+
"grad_norm": NaN,
|
| 119 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 120 |
+
"loss": 2.4278,
|
| 121 |
"step": 16
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.02711323763955343,
|
| 125 |
+
"grad_norm": 4.420550346374512,
|
| 126 |
+
"learning_rate": 3e-05,
|
| 127 |
+
"loss": 1.9174,
|
| 128 |
"step": 17
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.028708133971291867,
|
| 132 |
+
"grad_norm": 9.368809700012207,
|
| 133 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 134 |
+
"loss": 2.0925,
|
| 135 |
"step": 18
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.030303030303030304,
|
| 139 |
+
"grad_norm": 4.382827281951904,
|
| 140 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 141 |
+
"loss": 1.3716,
|
| 142 |
"step": 19
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.03189792663476874,
|
| 146 |
+
"grad_norm": 24.012392044067383,
|
| 147 |
+
"learning_rate": 3.6e-05,
|
| 148 |
+
"loss": 2.348,
|
| 149 |
"step": 20
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.03349282296650718,
|
| 153 |
+
"grad_norm": 2.227252721786499,
|
| 154 |
+
"learning_rate": 3.8e-05,
|
| 155 |
+
"loss": 2.0678,
|
| 156 |
"step": 21
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.03508771929824561,
|
| 160 |
+
"grad_norm": 6.851357460021973,
|
| 161 |
+
"learning_rate": 4e-05,
|
| 162 |
+
"loss": 3.2008,
|
| 163 |
"step": 22
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.03668261562998405,
|
| 167 |
+
"grad_norm": 5.75844144821167,
|
| 168 |
+
"learning_rate": 4.2e-05,
|
| 169 |
+
"loss": 1.8704,
|
| 170 |
"step": 23
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.03827751196172249,
|
| 174 |
+
"grad_norm": 14.692899703979492,
|
| 175 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 176 |
+
"loss": 2.9055,
|
| 177 |
"step": 24
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.03987240829346093,
|
| 181 |
+
"grad_norm": NaN,
|
| 182 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 183 |
+
"loss": 2.4497,
|
| 184 |
"step": 25
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.04146730462519936,
|
| 188 |
+
"grad_norm": NaN,
|
| 189 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 190 |
+
"loss": 2.4382,
|
| 191 |
"step": 26
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.0430622009569378,
|
| 195 |
+
"grad_norm": 6.065037727355957,
|
| 196 |
+
"learning_rate": 4.600000000000001e-05,
|
| 197 |
+
"loss": 2.105,
|
| 198 |
"step": 27
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.044657097288676235,
|
| 202 |
+
"grad_norm": 4.765100479125977,
|
| 203 |
+
"learning_rate": 4.8e-05,
|
| 204 |
+
"loss": 2.5292,
|
| 205 |
"step": 28
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.046251993620414676,
|
| 209 |
+
"grad_norm": 10.693521499633789,
|
| 210 |
+
"learning_rate": 5e-05,
|
| 211 |
+
"loss": 1.4471,
|
| 212 |
"step": 29
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.04784688995215311,
|
| 216 |
+
"grad_norm": 5.703505992889404,
|
| 217 |
+
"learning_rate": 5.2000000000000004e-05,
|
| 218 |
+
"loss": 1.8759,
|
| 219 |
"step": 30
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.049441786283891544,
|
| 223 |
+
"grad_norm": 6.594554424285889,
|
| 224 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 225 |
+
"loss": 2.3555,
|
| 226 |
"step": 31
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.051036682615629984,
|
| 230 |
+
"grad_norm": 10.944120407104492,
|
| 231 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 232 |
+
"loss": 2.6607,
|
| 233 |
"step": 32
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.05263157894736842,
|
| 237 |
+
"grad_norm": 4.437515735626221,
|
| 238 |
+
"learning_rate": 5.8e-05,
|
| 239 |
+
"loss": 2.1366,
|
| 240 |
"step": 33
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.05422647527910686,
|
| 244 |
+
"grad_norm": 18.751192092895508,
|
| 245 |
+
"learning_rate": 6e-05,
|
| 246 |
+
"loss": 2.8504,
|
| 247 |
"step": 34
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.05582137161084529,
|
| 251 |
+
"grad_norm": 4.625219821929932,
|
| 252 |
+
"learning_rate": 6.2e-05,
|
| 253 |
+
"loss": 1.7409,
|
| 254 |
"step": 35
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.05741626794258373,
|
| 258 |
+
"grad_norm": 5.256422996520996,
|
| 259 |
+
"learning_rate": 6.400000000000001e-05,
|
| 260 |
+
"loss": 2.965,
|
| 261 |
"step": 36
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.05901116427432217,
|
| 265 |
+
"grad_norm": 5.383383274078369,
|
| 266 |
+
"learning_rate": 6.6e-05,
|
| 267 |
+
"loss": 2.5555,
|
| 268 |
"step": 37
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 0.06060606060606061,
|
| 272 |
+
"grad_norm": 20.616491317749023,
|
| 273 |
+
"learning_rate": 6.800000000000001e-05,
|
| 274 |
+
"loss": 2.1693,
|
| 275 |
"step": 38
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.06220095693779904,
|
| 279 |
+
"grad_norm": 14.188122749328613,
|
| 280 |
+
"learning_rate": 7e-05,
|
| 281 |
+
"loss": 3.2098,
|
| 282 |
"step": 39
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.06379585326953748,
|
| 286 |
+
"grad_norm": 7.294703483581543,
|
| 287 |
+
"learning_rate": 7.2e-05,
|
| 288 |
+
"loss": 2.8137,
|
| 289 |
"step": 40
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 0.06539074960127592,
|
| 293 |
+
"grad_norm": 16.109020233154297,
|
| 294 |
+
"learning_rate": 7.4e-05,
|
| 295 |
+
"loss": 3.4821,
|
| 296 |
"step": 41
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 0.06698564593301436,
|
| 300 |
+
"grad_norm": 12.77609920501709,
|
| 301 |
+
"learning_rate": 7.6e-05,
|
| 302 |
+
"loss": 3.2586,
|
| 303 |
"step": 42
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 0.0685805422647528,
|
| 307 |
+
"grad_norm": 7.247976303100586,
|
| 308 |
+
"learning_rate": 7.800000000000001e-05,
|
| 309 |
+
"loss": 2.1131,
|
| 310 |
"step": 43
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 0.07017543859649122,
|
| 314 |
+
"grad_norm": 3.0302693843841553,
|
| 315 |
+
"learning_rate": 8e-05,
|
| 316 |
+
"loss": 2.3047,
|
| 317 |
"step": 44
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.07177033492822966,
|
| 321 |
+
"grad_norm": 6.998560905456543,
|
| 322 |
+
"learning_rate": 8.2e-05,
|
| 323 |
+
"loss": 1.9685,
|
| 324 |
"step": 45
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 0.0733652312599681,
|
| 328 |
+
"grad_norm": 11.82168197631836,
|
| 329 |
+
"learning_rate": 8.4e-05,
|
| 330 |
+
"loss": 2.412,
|
| 331 |
"step": 46
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 0.07496012759170653,
|
| 335 |
+
"grad_norm": 2.4560766220092773,
|
| 336 |
+
"learning_rate": 8.6e-05,
|
| 337 |
+
"loss": 1.5546,
|
| 338 |
"step": 47
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.07655502392344497,
|
| 342 |
+
"grad_norm": 7.5531110763549805,
|
| 343 |
+
"learning_rate": 8.800000000000001e-05,
|
| 344 |
+
"loss": 2.0706,
|
| 345 |
"step": 48
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 0.07814992025518341,
|
| 349 |
+
"grad_norm": 9.41441535949707,
|
| 350 |
+
"learning_rate": 9e-05,
|
| 351 |
+
"loss": 2.0769,
|
| 352 |
"step": 49
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 0.07974481658692185,
|
| 356 |
+
"grad_norm": 5.027219295501709,
|
| 357 |
+
"learning_rate": 9.200000000000001e-05,
|
| 358 |
+
"loss": 1.0112,
|
| 359 |
"step": 50
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.08133971291866028,
|
| 363 |
+
"grad_norm": 4.590622901916504,
|
| 364 |
+
"learning_rate": 9.4e-05,
|
| 365 |
+
"loss": 2.4014,
|
| 366 |
"step": 51
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 0.08293460925039872,
|
| 370 |
+
"grad_norm": 110.29186248779297,
|
| 371 |
+
"learning_rate": 9.6e-05,
|
| 372 |
+
"loss": 2.2312,
|
| 373 |
"step": 52
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 0.08452950558213716,
|
| 377 |
+
"grad_norm": 13.907599449157715,
|
| 378 |
+
"learning_rate": 9.8e-05,
|
| 379 |
+
"loss": 3.74,
|
| 380 |
"step": 53
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 0.0861244019138756,
|
| 384 |
+
"grad_norm": 21.45476722717285,
|
| 385 |
+
"learning_rate": 0.0001,
|
| 386 |
+
"loss": 2.2449,
|
| 387 |
"step": 54
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 0.08771929824561403,
|
| 391 |
+
"grad_norm": 26.367050170898438,
|
| 392 |
+
"learning_rate": 0.00010200000000000001,
|
| 393 |
+
"loss": 2.4117,
|
| 394 |
"step": 55
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 0.08931419457735247,
|
| 398 |
+
"grad_norm": 11.473040580749512,
|
| 399 |
+
"learning_rate": 0.00010400000000000001,
|
| 400 |
+
"loss": 1.9575,
|
| 401 |
"step": 56
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.09090909090909091,
|
| 405 |
+
"grad_norm": 7.996298789978027,
|
| 406 |
+
"learning_rate": 0.00010600000000000002,
|
| 407 |
+
"loss": 2.2506,
|
| 408 |
"step": 57
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 0.09250398724082935,
|
| 412 |
+
"grad_norm": 5.780291557312012,
|
| 413 |
+
"learning_rate": 0.00010800000000000001,
|
| 414 |
+
"loss": 1.685,
|
| 415 |
"step": 58
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 0.09409888357256778,
|
| 419 |
+
"grad_norm": 14.359979629516602,
|
| 420 |
+
"learning_rate": 0.00011000000000000002,
|
| 421 |
+
"loss": 2.1997,
|
| 422 |
"step": 59
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 0.09569377990430622,
|
| 426 |
+
"grad_norm": 2.4853787422180176,
|
| 427 |
+
"learning_rate": 0.00011200000000000001,
|
| 428 |
+
"loss": 1.2731,
|
| 429 |
"step": 60
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 0.09728867623604466,
|
| 433 |
+
"grad_norm": 7.136777877807617,
|
| 434 |
+
"learning_rate": 0.00011399999999999999,
|
| 435 |
+
"loss": 2.1987,
|
| 436 |
"step": 61
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 0.09888357256778309,
|
| 440 |
+
"grad_norm": 7.219399929046631,
|
| 441 |
+
"learning_rate": 0.000116,
|
| 442 |
+
"loss": 2.3118,
|
| 443 |
"step": 62
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.10047846889952153,
|
| 447 |
+
"grad_norm": 15.516590118408203,
|
| 448 |
+
"learning_rate": 0.000118,
|
| 449 |
+
"loss": 2.2534,
|
| 450 |
"step": 63
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 0.10207336523125997,
|
| 454 |
+
"grad_norm": 4.05905818939209,
|
| 455 |
+
"learning_rate": 0.00012,
|
| 456 |
+
"loss": 2.0509,
|
| 457 |
"step": 64
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 0.10366826156299841,
|
| 461 |
+
"grad_norm": 51.53736877441406,
|
| 462 |
+
"learning_rate": 0.000122,
|
| 463 |
+
"loss": 3.7686,
|
| 464 |
"step": 65
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 0.10526315789473684,
|
| 468 |
+
"grad_norm": 4.316274642944336,
|
| 469 |
+
"learning_rate": 0.000124,
|
| 470 |
+
"loss": 1.332,
|
| 471 |
"step": 66
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 0.10685805422647528,
|
| 475 |
+
"grad_norm": 47.97142028808594,
|
| 476 |
+
"learning_rate": 0.000126,
|
| 477 |
+
"loss": 1.4256,
|
| 478 |
"step": 67
|
| 479 |
},
|
| 480 |
{
|
| 481 |
+
"epoch": 0.10845295055821372,
|
| 482 |
+
"grad_norm": 10.22885513305664,
|
| 483 |
+
"learning_rate": 0.00012800000000000002,
|
| 484 |
+
"loss": 1.4572,
|
| 485 |
"step": 68
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.11004784688995216,
|
| 489 |
+
"grad_norm": 5.966179370880127,
|
| 490 |
+
"learning_rate": 0.00013000000000000002,
|
| 491 |
+
"loss": 1.3998,
|
| 492 |
"step": 69
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 0.11164274322169059,
|
| 496 |
+
"grad_norm": 3.413966178894043,
|
| 497 |
+
"learning_rate": 0.000132,
|
| 498 |
+
"loss": 1.6306,
|
| 499 |
"step": 70
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 0.11323763955342903,
|
| 503 |
+
"grad_norm": 197.0709991455078,
|
| 504 |
+
"learning_rate": 0.000134,
|
| 505 |
+
"loss": 1.6587,
|
| 506 |
"step": 71
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 0.11483253588516747,
|
| 510 |
+
"grad_norm": 12.735696792602539,
|
| 511 |
+
"learning_rate": 0.00013600000000000003,
|
| 512 |
+
"loss": 2.0253,
|
| 513 |
"step": 72
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 0.11642743221690591,
|
| 517 |
+
"grad_norm": 14.38369083404541,
|
| 518 |
+
"learning_rate": 0.000138,
|
| 519 |
+
"loss": 2.0833,
|
| 520 |
"step": 73
|
| 521 |
},
|
| 522 |
{
|
| 523 |
+
"epoch": 0.11802232854864433,
|
| 524 |
+
"grad_norm": 5.746401786804199,
|
| 525 |
+
"learning_rate": 0.00014,
|
| 526 |
+
"loss": 1.959,
|
| 527 |
"step": 74
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 0.11961722488038277,
|
| 531 |
+
"grad_norm": 4.266767501831055,
|
| 532 |
+
"learning_rate": 0.000142,
|
| 533 |
+
"loss": 1.5752,
|
| 534 |
"step": 75
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 0.12121212121212122,
|
| 538 |
+
"grad_norm": 16.618648529052734,
|
| 539 |
+
"learning_rate": 0.000144,
|
| 540 |
+
"loss": 3.0947,
|
| 541 |
"step": 76
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 0.12280701754385964,
|
| 545 |
+
"grad_norm": 2.3666510581970215,
|
| 546 |
+
"learning_rate": 0.000146,
|
| 547 |
+
"loss": 2.0715,
|
| 548 |
"step": 77
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 0.12440191387559808,
|
| 552 |
+
"grad_norm": 6.622753143310547,
|
| 553 |
+
"learning_rate": 0.000148,
|
| 554 |
+
"loss": 1.8844,
|
| 555 |
"step": 78
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 0.12599681020733652,
|
| 559 |
+
"grad_norm": 43.816375732421875,
|
| 560 |
+
"learning_rate": 0.00015000000000000001,
|
| 561 |
+
"loss": 1.9108,
|
| 562 |
"step": 79
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 0.12759170653907495,
|
| 566 |
+
"grad_norm": 6.682168006896973,
|
| 567 |
+
"learning_rate": 0.000152,
|
| 568 |
+
"loss": 1.5588,
|
| 569 |
"step": 80
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 0.1291866028708134,
|
| 573 |
+
"grad_norm": 3.0997276306152344,
|
| 574 |
+
"learning_rate": 0.000154,
|
| 575 |
+
"loss": 2.1988,
|
| 576 |
"step": 81
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 0.13078149920255183,
|
| 580 |
+
"grad_norm": 4.309176445007324,
|
| 581 |
+
"learning_rate": 0.00015600000000000002,
|
| 582 |
+
"loss": 1.406,
|
| 583 |
"step": 82
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"epoch": 0.13237639553429026,
|
| 587 |
+
"grad_norm": 5.416724681854248,
|
| 588 |
+
"learning_rate": 0.00015800000000000002,
|
| 589 |
+
"loss": 1.576,
|
| 590 |
"step": 83
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 0.1339712918660287,
|
| 594 |
+
"grad_norm": 4.361196517944336,
|
| 595 |
+
"learning_rate": 0.00016,
|
| 596 |
+
"loss": 2.2985,
|
| 597 |
"step": 84
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 0.13556618819776714,
|
| 601 |
+
"grad_norm": 7.655956745147705,
|
| 602 |
+
"learning_rate": 0.000162,
|
| 603 |
+
"loss": 1.3693,
|
| 604 |
"step": 85
|
| 605 |
},
|
| 606 |
{
|
| 607 |
+
"epoch": 0.1371610845295056,
|
| 608 |
+
"grad_norm": 5.973109245300293,
|
| 609 |
+
"learning_rate": 0.000164,
|
| 610 |
+
"loss": 2.4914,
|
| 611 |
"step": 86
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 0.13875598086124402,
|
| 615 |
+
"grad_norm": 5.1589436531066895,
|
| 616 |
+
"learning_rate": 0.000166,
|
| 617 |
+
"loss": 1.6638,
|
| 618 |
"step": 87
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 0.14035087719298245,
|
| 622 |
+
"grad_norm": 6.356327056884766,
|
| 623 |
+
"learning_rate": 0.000168,
|
| 624 |
+
"loss": 1.381,
|
| 625 |
"step": 88
|
| 626 |
},
|
| 627 |
{
|
| 628 |
+
"epoch": 0.1419457735247209,
|
| 629 |
+
"grad_norm": 1.5582388639450073,
|
| 630 |
+
"learning_rate": 0.00017,
|
| 631 |
+
"loss": 1.4864,
|
| 632 |
"step": 89
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 0.14354066985645933,
|
| 636 |
+
"grad_norm": 2.830798387527466,
|
| 637 |
+
"learning_rate": 0.000172,
|
| 638 |
+
"loss": 1.897,
|
| 639 |
"step": 90
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 0.14513556618819776,
|
| 643 |
+
"grad_norm": 20.24082374572754,
|
| 644 |
+
"learning_rate": 0.000174,
|
| 645 |
+
"loss": 1.7698,
|
| 646 |
"step": 91
|
| 647 |
},
|
| 648 |
{
|
| 649 |
+
"epoch": 0.1467304625199362,
|
| 650 |
+
"grad_norm": 35.429466247558594,
|
| 651 |
+
"learning_rate": 0.00017600000000000002,
|
| 652 |
+
"loss": 3.6051,
|
| 653 |
"step": 92
|
| 654 |
},
|
| 655 |
{
|
| 656 |
+
"epoch": 0.14832535885167464,
|
| 657 |
+
"grad_norm": 9.8428373336792,
|
| 658 |
+
"learning_rate": 0.00017800000000000002,
|
| 659 |
+
"loss": 1.9402,
|
| 660 |
"step": 93
|
| 661 |
},
|
| 662 |
{
|
| 663 |
+
"epoch": 0.14992025518341306,
|
| 664 |
+
"grad_norm": 3.7131340503692627,
|
| 665 |
+
"learning_rate": 0.00018,
|
| 666 |
+
"loss": 2.2192,
|
| 667 |
"step": 94
|
| 668 |
},
|
| 669 |
{
|
| 670 |
+
"epoch": 0.15151515151515152,
|
| 671 |
+
"grad_norm": 8.838521003723145,
|
| 672 |
+
"learning_rate": 0.000182,
|
| 673 |
+
"loss": 2.0167,
|
| 674 |
"step": 95
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 0.15311004784688995,
|
| 678 |
+
"grad_norm": 9.63149356842041,
|
| 679 |
+
"learning_rate": 0.00018400000000000003,
|
| 680 |
+
"loss": 1.6169,
|
| 681 |
"step": 96
|
| 682 |
},
|
| 683 |
{
|
| 684 |
+
"epoch": 0.1547049441786284,
|
| 685 |
+
"grad_norm": 6.012756824493408,
|
| 686 |
+
"learning_rate": 0.00018600000000000002,
|
| 687 |
+
"loss": 1.4477,
|
| 688 |
"step": 97
|
| 689 |
},
|
| 690 |
{
|
| 691 |
+
"epoch": 0.15629984051036683,
|
| 692 |
+
"grad_norm": 9.02092456817627,
|
| 693 |
+
"learning_rate": 0.000188,
|
| 694 |
+
"loss": 1.5999,
|
| 695 |
"step": 98
|
| 696 |
},
|
| 697 |
{
|
| 698 |
+
"epoch": 0.15789473684210525,
|
| 699 |
+
"grad_norm": 5.8646416664123535,
|
| 700 |
+
"learning_rate": 0.00019,
|
| 701 |
+
"loss": 1.2515,
|
| 702 |
"step": 99
|
| 703 |
},
|
| 704 |
{
|
| 705 |
+
"epoch": 0.1594896331738437,
|
| 706 |
+
"grad_norm": 3.3447864055633545,
|
| 707 |
+
"learning_rate": 0.000192,
|
| 708 |
+
"loss": 1.257,
|
| 709 |
"step": 100
|
| 710 |
},
|
| 711 |
{
|
| 712 |
+
"epoch": 0.16108452950558214,
|
| 713 |
+
"grad_norm": 8.706202507019043,
|
| 714 |
+
"learning_rate": 0.000194,
|
| 715 |
+
"loss": 2.0047,
|
| 716 |
"step": 101
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 0.16267942583732056,
|
| 720 |
+
"grad_norm": 17.863378524780273,
|
| 721 |
+
"learning_rate": 0.000196,
|
| 722 |
+
"loss": 2.421,
|
| 723 |
"step": 102
|
| 724 |
},
|
| 725 |
{
|
| 726 |
+
"epoch": 0.16427432216905902,
|
| 727 |
+
"grad_norm": 5.376917362213135,
|
| 728 |
"learning_rate": 0.00019800000000000002,
|
| 729 |
+
"loss": 2.1448,
|
| 730 |
"step": 103
|
| 731 |
},
|
| 732 |
{
|
| 733 |
+
"epoch": 0.16586921850079744,
|
| 734 |
+
"grad_norm": 3.9824860095977783,
|
| 735 |
+
"learning_rate": 0.0002,
|
| 736 |
+
"loss": 1.7987,
|
| 737 |
"step": 104
|
| 738 |
},
|
| 739 |
{
|
| 740 |
+
"epoch": 0.1674641148325359,
|
| 741 |
+
"grad_norm": 6.7771382331848145,
|
| 742 |
+
"learning_rate": 0.00019800000000000002,
|
| 743 |
+
"loss": 1.8173,
|
| 744 |
"step": 105
|
| 745 |
},
|
| 746 |
{
|
| 747 |
+
"epoch": 0.16905901116427433,
|
| 748 |
+
"grad_norm": 6.204737186431885,
|
| 749 |
+
"learning_rate": 0.000196,
|
| 750 |
+
"loss": 1.2615,
|
| 751 |
"step": 106
|
| 752 |
},
|
| 753 |
{
|
| 754 |
+
"epoch": 0.17065390749601275,
|
| 755 |
+
"grad_norm": 5.108335018157959,
|
| 756 |
+
"learning_rate": 0.000194,
|
| 757 |
+
"loss": 1.6966,
|
| 758 |
"step": 107
|
| 759 |
},
|
| 760 |
{
|
| 761 |
+
"epoch": 0.1722488038277512,
|
| 762 |
+
"grad_norm": 51.948585510253906,
|
| 763 |
+
"learning_rate": 0.000192,
|
| 764 |
+
"loss": 2.8226,
|
| 765 |
"step": 108
|
| 766 |
},
|
| 767 |
{
|
| 768 |
+
"epoch": 0.17384370015948963,
|
| 769 |
+
"grad_norm": 3.510647773742676,
|
| 770 |
+
"learning_rate": 0.00019,
|
| 771 |
"loss": 1.4454,
|
| 772 |
"step": 109
|
| 773 |
},
|
| 774 |
{
|
| 775 |
+
"epoch": 0.17543859649122806,
|
| 776 |
+
"grad_norm": 5.786037921905518,
|
| 777 |
+
"learning_rate": 0.000188,
|
| 778 |
+
"loss": 1.3979,
|
| 779 |
"step": 110
|
| 780 |
},
|
| 781 |
{
|
| 782 |
+
"epoch": 0.17703349282296652,
|
| 783 |
+
"grad_norm": 5.4267964363098145,
|
| 784 |
+
"learning_rate": 0.00018600000000000002,
|
| 785 |
+
"loss": 1.8775,
|
| 786 |
"step": 111
|
| 787 |
},
|
| 788 |
{
|
| 789 |
+
"epoch": 0.17862838915470494,
|
| 790 |
+
"grad_norm": 7.448083400726318,
|
| 791 |
+
"learning_rate": 0.00018400000000000003,
|
| 792 |
+
"loss": 1.8377,
|
| 793 |
"step": 112
|
| 794 |
},
|
| 795 |
{
|
| 796 |
+
"epoch": 0.18022328548644337,
|
| 797 |
+
"grad_norm": 6.855521202087402,
|
| 798 |
+
"learning_rate": 0.000182,
|
| 799 |
+
"loss": 1.7079,
|
| 800 |
"step": 113
|
| 801 |
},
|
| 802 |
{
|
| 803 |
+
"epoch": 0.18181818181818182,
|
| 804 |
+
"grad_norm": 6.257575988769531,
|
| 805 |
+
"learning_rate": 0.00018,
|
| 806 |
+
"loss": 2.2364,
|
| 807 |
"step": 114
|
| 808 |
},
|
| 809 |
{
|
| 810 |
+
"epoch": 0.18341307814992025,
|
| 811 |
+
"grad_norm": 15.47193431854248,
|
| 812 |
+
"learning_rate": 0.00017800000000000002,
|
| 813 |
+
"loss": 1.7272,
|
| 814 |
"step": 115
|
| 815 |
},
|
| 816 |
{
|
| 817 |
+
"epoch": 0.1850079744816587,
|
| 818 |
+
"grad_norm": 12.51389217376709,
|
| 819 |
+
"learning_rate": 0.00017600000000000002,
|
| 820 |
+
"loss": 1.8385,
|
| 821 |
"step": 116
|
| 822 |
},
|
| 823 |
{
|
| 824 |
+
"epoch": 0.18660287081339713,
|
| 825 |
+
"grad_norm": 8.150806427001953,
|
| 826 |
+
"learning_rate": 0.000174,
|
| 827 |
+
"loss": 2.429,
|
| 828 |
"step": 117
|
| 829 |
},
|
| 830 |
{
|
| 831 |
+
"epoch": 0.18819776714513556,
|
| 832 |
+
"grad_norm": 29.065210342407227,
|
| 833 |
+
"learning_rate": 0.000172,
|
| 834 |
+
"loss": 1.3318,
|
| 835 |
"step": 118
|
| 836 |
},
|
| 837 |
{
|
| 838 |
+
"epoch": 0.189792663476874,
|
| 839 |
+
"grad_norm": 5.457766532897949,
|
| 840 |
+
"learning_rate": 0.00017,
|
| 841 |
+
"loss": 1.2329,
|
| 842 |
"step": 119
|
| 843 |
},
|
| 844 |
{
|
| 845 |
+
"epoch": 0.19138755980861244,
|
| 846 |
+
"grad_norm": 3.7080814838409424,
|
| 847 |
+
"learning_rate": 0.000168,
|
| 848 |
+
"loss": 1.1145,
|
| 849 |
"step": 120
|
| 850 |
},
|
| 851 |
{
|
| 852 |
+
"epoch": 0.19298245614035087,
|
| 853 |
+
"grad_norm": 5.2060065269470215,
|
| 854 |
+
"learning_rate": 0.000166,
|
| 855 |
+
"loss": 1.6808,
|
| 856 |
"step": 121
|
| 857 |
},
|
| 858 |
{
|
| 859 |
+
"epoch": 0.19457735247208932,
|
| 860 |
+
"grad_norm": 15.03097152709961,
|
| 861 |
+
"learning_rate": 0.000164,
|
| 862 |
+
"loss": 2.207,
|
| 863 |
"step": 122
|
| 864 |
},
|
| 865 |
{
|
| 866 |
+
"epoch": 0.19617224880382775,
|
| 867 |
+
"grad_norm": 3.1523945331573486,
|
| 868 |
+
"learning_rate": 0.000162,
|
| 869 |
+
"loss": 0.8121,
|
| 870 |
"step": 123
|
| 871 |
},
|
| 872 |
{
|
| 873 |
+
"epoch": 0.19776714513556617,
|
| 874 |
+
"grad_norm": 12.531198501586914,
|
| 875 |
+
"learning_rate": 0.00016,
|
| 876 |
+
"loss": 1.6493,
|
| 877 |
"step": 124
|
| 878 |
},
|
| 879 |
{
|
| 880 |
+
"epoch": 0.19936204146730463,
|
| 881 |
+
"grad_norm": 18.13838768005371,
|
| 882 |
+
"learning_rate": 0.00015800000000000002,
|
| 883 |
+
"loss": 1.9895,
|
| 884 |
"step": 125
|
| 885 |
},
|
| 886 |
{
|
| 887 |
+
"epoch": 0.20095693779904306,
|
| 888 |
+
"grad_norm": 3.7068324089050293,
|
| 889 |
+
"learning_rate": 0.00015600000000000002,
|
| 890 |
+
"loss": 1.2397,
|
| 891 |
"step": 126
|
| 892 |
},
|
| 893 |
{
|
| 894 |
+
"epoch": 0.2025518341307815,
|
| 895 |
+
"grad_norm": 3.565275192260742,
|
| 896 |
+
"learning_rate": 0.000154,
|
| 897 |
+
"loss": 1.1711,
|
| 898 |
"step": 127
|
| 899 |
},
|
| 900 |
{
|
| 901 |
+
"epoch": 0.20414673046251994,
|
| 902 |
+
"grad_norm": 2.20239520072937,
|
| 903 |
+
"learning_rate": 0.000152,
|
| 904 |
+
"loss": 1.8436,
|
| 905 |
"step": 128
|
| 906 |
},
|
| 907 |
{
|
| 908 |
+
"epoch": 0.20574162679425836,
|
| 909 |
+
"grad_norm": 3.103001117706299,
|
| 910 |
+
"learning_rate": 0.00015000000000000001,
|
| 911 |
+
"loss": 1.0085,
|
| 912 |
"step": 129
|
| 913 |
},
|
| 914 |
{
|
| 915 |
+
"epoch": 0.20733652312599682,
|
| 916 |
+
"grad_norm": 3.8708302974700928,
|
| 917 |
+
"learning_rate": 0.000148,
|
| 918 |
+
"loss": 1.0336,
|
| 919 |
"step": 130
|
| 920 |
},
|
| 921 |
{
|
| 922 |
+
"epoch": 0.20893141945773525,
|
| 923 |
+
"grad_norm": 4.008622646331787,
|
| 924 |
+
"learning_rate": 0.000146,
|
| 925 |
+
"loss": 1.7669,
|
| 926 |
"step": 131
|
| 927 |
},
|
| 928 |
{
|
| 929 |
+
"epoch": 0.21052631578947367,
|
| 930 |
+
"grad_norm": 3.847217559814453,
|
| 931 |
+
"learning_rate": 0.000144,
|
| 932 |
+
"loss": 1.5163,
|
| 933 |
"step": 132
|
| 934 |
},
|
| 935 |
{
|
| 936 |
+
"epoch": 0.21212121212121213,
|
| 937 |
+
"grad_norm": 9.75152587890625,
|
| 938 |
+
"learning_rate": 0.000142,
|
| 939 |
+
"loss": 2.3286,
|
| 940 |
"step": 133
|
| 941 |
},
|
| 942 |
{
|
| 943 |
+
"epoch": 0.21371610845295055,
|
| 944 |
+
"grad_norm": 9.416215896606445,
|
| 945 |
+
"learning_rate": 0.00014,
|
| 946 |
+
"loss": 2.0397,
|
| 947 |
"step": 134
|
| 948 |
},
|
| 949 |
{
|
| 950 |
+
"epoch": 0.215311004784689,
|
| 951 |
+
"grad_norm": 8.208052635192871,
|
| 952 |
+
"learning_rate": 0.000138,
|
| 953 |
+
"loss": 1.7927,
|
| 954 |
"step": 135
|
| 955 |
},
|
| 956 |
{
|
| 957 |
+
"epoch": 0.21690590111642744,
|
| 958 |
+
"grad_norm": 11.827594757080078,
|
| 959 |
+
"learning_rate": 0.00013600000000000003,
|
| 960 |
+
"loss": 2.1958,
|
| 961 |
"step": 136
|
| 962 |
},
|
| 963 |
{
|
| 964 |
+
"epoch": 0.21850079744816586,
|
| 965 |
+
"grad_norm": 5.788820266723633,
|
| 966 |
+
"learning_rate": 0.000134,
|
| 967 |
+
"loss": 1.8336,
|
| 968 |
"step": 137
|
| 969 |
},
|
| 970 |
{
|
| 971 |
+
"epoch": 0.22009569377990432,
|
| 972 |
+
"grad_norm": 3.6653785705566406,
|
| 973 |
+
"learning_rate": 0.000132,
|
| 974 |
+
"loss": 1.0477,
|
| 975 |
"step": 138
|
| 976 |
},
|
| 977 |
{
|
| 978 |
+
"epoch": 0.22169059011164274,
|
| 979 |
+
"grad_norm": 63.57593536376953,
|
| 980 |
+
"learning_rate": 0.00013000000000000002,
|
| 981 |
+
"loss": 1.5667,
|
| 982 |
"step": 139
|
| 983 |
},
|
| 984 |
{
|
| 985 |
+
"epoch": 0.22328548644338117,
|
| 986 |
+
"grad_norm": 12.209519386291504,
|
| 987 |
+
"learning_rate": 0.00012800000000000002,
|
| 988 |
+
"loss": 1.6414,
|
| 989 |
"step": 140
|
| 990 |
},
|
| 991 |
{
|
| 992 |
+
"epoch": 0.22488038277511962,
|
| 993 |
+
"grad_norm": 8.834230422973633,
|
| 994 |
+
"learning_rate": 0.000126,
|
| 995 |
+
"loss": 1.1292,
|
| 996 |
"step": 141
|
| 997 |
},
|
| 998 |
{
|
| 999 |
+
"epoch": 0.22647527910685805,
|
| 1000 |
+
"grad_norm": 3.4640305042266846,
|
| 1001 |
+
"learning_rate": 0.000124,
|
| 1002 |
+
"loss": 1.3858,
|
| 1003 |
"step": 142
|
| 1004 |
},
|
| 1005 |
{
|
| 1006 |
+
"epoch": 0.22807017543859648,
|
| 1007 |
+
"grad_norm": 5.367834091186523,
|
| 1008 |
+
"learning_rate": 0.000122,
|
| 1009 |
+
"loss": 1.4735,
|
| 1010 |
"step": 143
|
| 1011 |
},
|
| 1012 |
{
|
| 1013 |
+
"epoch": 0.22966507177033493,
|
| 1014 |
+
"grad_norm": 3.5475995540618896,
|
| 1015 |
+
"learning_rate": 0.00012,
|
| 1016 |
+
"loss": 1.8889,
|
| 1017 |
"step": 144
|
| 1018 |
},
|
| 1019 |
{
|
| 1020 |
+
"epoch": 0.23125996810207336,
|
| 1021 |
+
"grad_norm": 4.0117621421813965,
|
| 1022 |
+
"learning_rate": 0.000118,
|
| 1023 |
+
"loss": 0.6609,
|
| 1024 |
"step": 145
|
| 1025 |
},
|
| 1026 |
{
|
| 1027 |
+
"epoch": 0.23285486443381181,
|
| 1028 |
+
"grad_norm": 2.641918420791626,
|
| 1029 |
+
"learning_rate": 0.000116,
|
| 1030 |
+
"loss": 1.439,
|
| 1031 |
"step": 146
|
| 1032 |
},
|
| 1033 |
{
|
| 1034 |
+
"epoch": 0.23444976076555024,
|
| 1035 |
+
"grad_norm": 24.166297912597656,
|
| 1036 |
+
"learning_rate": 0.00011399999999999999,
|
| 1037 |
+
"loss": 1.5829,
|
| 1038 |
"step": 147
|
| 1039 |
},
|
| 1040 |
{
|
| 1041 |
+
"epoch": 0.23604465709728867,
|
| 1042 |
+
"grad_norm": 12.261051177978516,
|
| 1043 |
+
"learning_rate": 0.00011200000000000001,
|
| 1044 |
+
"loss": 2.6259,
|
| 1045 |
"step": 148
|
| 1046 |
},
|
| 1047 |
{
|
| 1048 |
+
"epoch": 0.23763955342902712,
|
| 1049 |
+
"grad_norm": 4.844594955444336,
|
| 1050 |
+
"learning_rate": 0.00011000000000000002,
|
| 1051 |
+
"loss": 2.2162,
|
| 1052 |
"step": 149
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
+
"epoch": 0.23923444976076555,
|
| 1056 |
+
"grad_norm": 6.697972297668457,
|
| 1057 |
+
"learning_rate": 0.00010800000000000001,
|
| 1058 |
+
"loss": 2.7383,
|
| 1059 |
"step": 150
|
| 1060 |
},
|
| 1061 |
{
|
| 1062 |
+
"epoch": 0.24082934609250398,
|
| 1063 |
+
"grad_norm": 4.018860340118408,
|
| 1064 |
+
"learning_rate": 0.00010600000000000002,
|
| 1065 |
+
"loss": 2.4861,
|
| 1066 |
"step": 151
|
| 1067 |
},
|
| 1068 |
{
|
| 1069 |
+
"epoch": 0.24242424242424243,
|
| 1070 |
+
"grad_norm": 7.205791473388672,
|
| 1071 |
+
"learning_rate": 0.00010400000000000001,
|
| 1072 |
+
"loss": 1.4837,
|
| 1073 |
"step": 152
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
+
"epoch": 0.24401913875598086,
|
| 1077 |
+
"grad_norm": 8.773712158203125,
|
| 1078 |
+
"learning_rate": 0.00010200000000000001,
|
| 1079 |
+
"loss": 1.6268,
|
| 1080 |
"step": 153
|
| 1081 |
},
|
| 1082 |
{
|
| 1083 |
+
"epoch": 0.24561403508771928,
|
| 1084 |
+
"grad_norm": 4.254454135894775,
|
| 1085 |
+
"learning_rate": 0.0001,
|
| 1086 |
+
"loss": 1.6405,
|
| 1087 |
"step": 154
|
| 1088 |
},
|
| 1089 |
{
|
| 1090 |
+
"epoch": 0.24720893141945774,
|
| 1091 |
+
"grad_norm": 4.130655288696289,
|
| 1092 |
+
"learning_rate": 9.8e-05,
|
| 1093 |
+
"loss": 0.9554,
|
| 1094 |
"step": 155
|
| 1095 |
},
|
| 1096 |
{
|
| 1097 |
+
"epoch": 0.24880382775119617,
|
| 1098 |
+
"grad_norm": 7.5216169357299805,
|
| 1099 |
+
"learning_rate": 9.6e-05,
|
| 1100 |
+
"loss": 1.3912,
|
| 1101 |
"step": 156
|
| 1102 |
},
|
| 1103 |
{
|
| 1104 |
+
"epoch": 0.2503987240829346,
|
| 1105 |
+
"grad_norm": 70.0249252319336,
|
| 1106 |
+
"learning_rate": 9.4e-05,
|
| 1107 |
+
"loss": 1.8041,
|
| 1108 |
"step": 157
|
| 1109 |
},
|
| 1110 |
{
|
| 1111 |
+
"epoch": 0.25199362041467305,
|
| 1112 |
+
"grad_norm": 3.9001526832580566,
|
| 1113 |
+
"learning_rate": 9.200000000000001e-05,
|
| 1114 |
+
"loss": 1.7109,
|
| 1115 |
"step": 158
|
| 1116 |
},
|
| 1117 |
{
|
| 1118 |
+
"epoch": 0.2535885167464115,
|
| 1119 |
+
"grad_norm": 5.0965094566345215,
|
| 1120 |
+
"learning_rate": 9e-05,
|
| 1121 |
+
"loss": 1.0314,
|
| 1122 |
"step": 159
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
+
"epoch": 0.2551834130781499,
|
| 1126 |
+
"grad_norm": 3.0462706089019775,
|
| 1127 |
+
"learning_rate": 8.800000000000001e-05,
|
| 1128 |
+
"loss": 1.3541,
|
| 1129 |
"step": 160
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
+
"epoch": 0.2567783094098884,
|
| 1133 |
+
"grad_norm": 2.8334619998931885,
|
| 1134 |
+
"learning_rate": 8.6e-05,
|
| 1135 |
+
"loss": 1.9827,
|
| 1136 |
"step": 161
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
+
"epoch": 0.2583732057416268,
|
| 1140 |
+
"grad_norm": 5.991235256195068,
|
| 1141 |
+
"learning_rate": 8.4e-05,
|
| 1142 |
+
"loss": 1.4319,
|
| 1143 |
"step": 162
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
+
"epoch": 0.25996810207336524,
|
| 1147 |
+
"grad_norm": 6.401395320892334,
|
| 1148 |
+
"learning_rate": 8.2e-05,
|
| 1149 |
+
"loss": 1.912,
|
| 1150 |
"step": 163
|
| 1151 |
},
|
| 1152 |
{
|
| 1153 |
+
"epoch": 0.26156299840510366,
|
| 1154 |
+
"grad_norm": 3.0269479751586914,
|
| 1155 |
+
"learning_rate": 8e-05,
|
| 1156 |
+
"loss": 1.5637,
|
| 1157 |
"step": 164
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
+
"epoch": 0.2631578947368421,
|
| 1161 |
+
"grad_norm": 10.380670547485352,
|
| 1162 |
+
"learning_rate": 7.800000000000001e-05,
|
| 1163 |
+
"loss": 1.8585,
|
| 1164 |
"step": 165
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
+
"epoch": 0.2647527910685805,
|
| 1168 |
+
"grad_norm": 7.375792503356934,
|
| 1169 |
+
"learning_rate": 7.6e-05,
|
| 1170 |
+
"loss": 1.8773,
|
| 1171 |
"step": 166
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
+
"epoch": 0.266347687400319,
|
| 1175 |
+
"grad_norm": 3.75229811668396,
|
| 1176 |
+
"learning_rate": 7.4e-05,
|
| 1177 |
+
"loss": 1.9936,
|
| 1178 |
"step": 167
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
+
"epoch": 0.2679425837320574,
|
| 1182 |
+
"grad_norm": 4.587918281555176,
|
| 1183 |
+
"learning_rate": 7.2e-05,
|
| 1184 |
+
"loss": 1.0402,
|
| 1185 |
"step": 168
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
+
"epoch": 0.26953748006379585,
|
| 1189 |
+
"grad_norm": 4.558103561401367,
|
| 1190 |
+
"learning_rate": 7e-05,
|
| 1191 |
+
"loss": 2.0408,
|
| 1192 |
"step": 169
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
+
"epoch": 0.2711323763955343,
|
| 1196 |
+
"grad_norm": 6.410488605499268,
|
| 1197 |
+
"learning_rate": 6.800000000000001e-05,
|
| 1198 |
+
"loss": 1.448,
|
| 1199 |
"step": 170
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
+
"epoch": 0.2727272727272727,
|
| 1203 |
+
"grad_norm": 9.21990966796875,
|
| 1204 |
+
"learning_rate": 6.6e-05,
|
| 1205 |
+
"loss": 0.963,
|
| 1206 |
"step": 171
|
| 1207 |
},
|
| 1208 |
{
|
| 1209 |
+
"epoch": 0.2743221690590112,
|
| 1210 |
+
"grad_norm": 10.446578025817871,
|
| 1211 |
+
"learning_rate": 6.400000000000001e-05,
|
| 1212 |
+
"loss": 1.4381,
|
| 1213 |
"step": 172
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
+
"epoch": 0.2759170653907496,
|
| 1217 |
+
"grad_norm": 13.717737197875977,
|
| 1218 |
+
"learning_rate": 6.2e-05,
|
| 1219 |
+
"loss": 1.2965,
|
| 1220 |
"step": 173
|
| 1221 |
},
|
| 1222 |
{
|
| 1223 |
+
"epoch": 0.27751196172248804,
|
| 1224 |
+
"grad_norm": 36.117401123046875,
|
| 1225 |
+
"learning_rate": 6e-05,
|
| 1226 |
+
"loss": 1.2144,
|
| 1227 |
"step": 174
|
| 1228 |
},
|
| 1229 |
{
|
| 1230 |
+
"epoch": 0.27910685805422647,
|
| 1231 |
+
"grad_norm": 8.741774559020996,
|
| 1232 |
+
"learning_rate": 5.8e-05,
|
| 1233 |
+
"loss": 1.5059,
|
| 1234 |
"step": 175
|
| 1235 |
},
|
| 1236 |
{
|
| 1237 |
+
"epoch": 0.2807017543859649,
|
| 1238 |
+
"grad_norm": 3.040428400039673,
|
| 1239 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 1240 |
+
"loss": 1.4881,
|
| 1241 |
"step": 176
|
| 1242 |
},
|
| 1243 |
{
|
| 1244 |
+
"epoch": 0.2822966507177033,
|
| 1245 |
+
"grad_norm": 7.401648044586182,
|
| 1246 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 1247 |
+
"loss": 0.7462,
|
| 1248 |
"step": 177
|
| 1249 |
},
|
| 1250 |
{
|
| 1251 |
+
"epoch": 0.2838915470494418,
|
| 1252 |
+
"grad_norm": 3.3486616611480713,
|
| 1253 |
+
"learning_rate": 5.2000000000000004e-05,
|
| 1254 |
+
"loss": 0.8773,
|
| 1255 |
"step": 178
|
| 1256 |
},
|
| 1257 |
{
|
| 1258 |
+
"epoch": 0.28548644338118023,
|
| 1259 |
+
"grad_norm": 9.647512435913086,
|
| 1260 |
+
"learning_rate": 5e-05,
|
| 1261 |
+
"loss": 1.3147,
|
| 1262 |
"step": 179
|
| 1263 |
},
|
| 1264 |
{
|
| 1265 |
+
"epoch": 0.28708133971291866,
|
| 1266 |
+
"grad_norm": 4.068251132965088,
|
| 1267 |
+
"learning_rate": 4.8e-05,
|
| 1268 |
+
"loss": 1.4796,
|
| 1269 |
"step": 180
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
+
"epoch": 0.2886762360446571,
|
| 1273 |
+
"grad_norm": 5.119231700897217,
|
| 1274 |
+
"learning_rate": 4.600000000000001e-05,
|
| 1275 |
+
"loss": 1.2465,
|
| 1276 |
"step": 181
|
| 1277 |
},
|
| 1278 |
{
|
| 1279 |
+
"epoch": 0.2902711323763955,
|
| 1280 |
+
"grad_norm": 11.434249877929688,
|
| 1281 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 1282 |
+
"loss": 1.0839,
|
| 1283 |
"step": 182
|
| 1284 |
},
|
| 1285 |
{
|
| 1286 |
+
"epoch": 0.291866028708134,
|
| 1287 |
+
"grad_norm": 2.6633126735687256,
|
| 1288 |
+
"learning_rate": 4.2e-05,
|
| 1289 |
+
"loss": 1.1653,
|
| 1290 |
"step": 183
|
| 1291 |
},
|
| 1292 |
{
|
| 1293 |
+
"epoch": 0.2934609250398724,
|
| 1294 |
+
"grad_norm": 8.366178512573242,
|
| 1295 |
+
"learning_rate": 4e-05,
|
| 1296 |
+
"loss": 1.8623,
|
| 1297 |
"step": 184
|
| 1298 |
},
|
| 1299 |
{
|
| 1300 |
+
"epoch": 0.29505582137161085,
|
| 1301 |
+
"grad_norm": 13.434002876281738,
|
| 1302 |
+
"learning_rate": 3.8e-05,
|
| 1303 |
+
"loss": 1.4409,
|
| 1304 |
"step": 185
|
| 1305 |
},
|
| 1306 |
{
|
| 1307 |
+
"epoch": 0.2966507177033493,
|
| 1308 |
+
"grad_norm": 3.535421371459961,
|
| 1309 |
+
"learning_rate": 3.6e-05,
|
| 1310 |
+
"loss": 1.2367,
|
| 1311 |
"step": 186
|
| 1312 |
},
|
| 1313 |
{
|
| 1314 |
+
"epoch": 0.2982456140350877,
|
| 1315 |
+
"grad_norm": 5.332886219024658,
|
| 1316 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 1317 |
+
"loss": 1.4744,
|
| 1318 |
"step": 187
|
| 1319 |
},
|
| 1320 |
{
|
| 1321 |
+
"epoch": 0.29984051036682613,
|
| 1322 |
+
"grad_norm": 3.7560222148895264,
|
| 1323 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 1324 |
+
"loss": 1.3489,
|
| 1325 |
"step": 188
|
| 1326 |
},
|
| 1327 |
{
|
| 1328 |
+
"epoch": 0.3014354066985646,
|
| 1329 |
+
"grad_norm": 4.513449668884277,
|
| 1330 |
+
"learning_rate": 3e-05,
|
| 1331 |
+
"loss": 1.3738,
|
| 1332 |
"step": 189
|
| 1333 |
},
|
| 1334 |
{
|
| 1335 |
+
"epoch": 0.30303030303030304,
|
| 1336 |
+
"grad_norm": 7.969113826751709,
|
| 1337 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 1338 |
+
"loss": 1.7352,
|
| 1339 |
"step": 190
|
| 1340 |
},
|
| 1341 |
{
|
| 1342 |
+
"epoch": 0.30462519936204147,
|
| 1343 |
+
"grad_norm": 4.11577033996582,
|
| 1344 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 1345 |
+
"loss": 1.6171,
|
| 1346 |
"step": 191
|
| 1347 |
},
|
| 1348 |
{
|
| 1349 |
+
"epoch": 0.3062200956937799,
|
| 1350 |
+
"grad_norm": 4.727830410003662,
|
| 1351 |
+
"learning_rate": 2.4e-05,
|
| 1352 |
+
"loss": 1.3359,
|
| 1353 |
"step": 192
|
| 1354 |
},
|
| 1355 |
{
|
| 1356 |
+
"epoch": 0.3078149920255183,
|
| 1357 |
+
"grad_norm": 1.7988307476043701,
|
| 1358 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 1359 |
+
"loss": 0.834,
|
| 1360 |
"step": 193
|
| 1361 |
},
|
| 1362 |
{
|
| 1363 |
+
"epoch": 0.3094098883572568,
|
| 1364 |
+
"grad_norm": 4.455323696136475,
|
| 1365 |
+
"learning_rate": 2e-05,
|
| 1366 |
+
"loss": 1.2367,
|
| 1367 |
"step": 194
|
| 1368 |
},
|
| 1369 |
{
|
| 1370 |
+
"epoch": 0.31100478468899523,
|
| 1371 |
+
"grad_norm": 5.334070682525635,
|
| 1372 |
+
"learning_rate": 1.8e-05,
|
| 1373 |
+
"loss": 1.0581,
|
| 1374 |
"step": 195
|
| 1375 |
},
|
| 1376 |
{
|
| 1377 |
+
"epoch": 0.31259968102073366,
|
| 1378 |
+
"grad_norm": 4.967501640319824,
|
| 1379 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 1380 |
+
"loss": 2.2129,
|
| 1381 |
"step": 196
|
| 1382 |
},
|
| 1383 |
{
|
| 1384 |
+
"epoch": 0.3141945773524721,
|
| 1385 |
+
"grad_norm": 4.0120673179626465,
|
| 1386 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 1387 |
+
"loss": 1.2043,
|
| 1388 |
"step": 197
|
| 1389 |
},
|
| 1390 |
{
|
| 1391 |
+
"epoch": 0.3157894736842105,
|
| 1392 |
+
"grad_norm": 4.491064548492432,
|
| 1393 |
+
"learning_rate": 1.2e-05,
|
| 1394 |
+
"loss": 1.3383,
|
| 1395 |
"step": 198
|
| 1396 |
},
|
| 1397 |
{
|
| 1398 |
+
"epoch": 0.31738437001594894,
|
| 1399 |
+
"grad_norm": 3.6377949714660645,
|
| 1400 |
+
"learning_rate": 1e-05,
|
| 1401 |
+
"loss": 1.2878,
|
| 1402 |
"step": 199
|
| 1403 |
},
|
| 1404 |
{
|
| 1405 |
+
"epoch": 0.3189792663476874,
|
| 1406 |
+
"grad_norm": 3.7990877628326416,
|
| 1407 |
+
"learning_rate": 8.000000000000001e-06,
|
| 1408 |
+
"loss": 1.7178,
|
| 1409 |
"step": 200
|
| 1410 |
}
|
| 1411 |
],
|
|
|
|
| 1426 |
"attributes": {}
|
| 1427 |
}
|
| 1428 |
},
|
| 1429 |
+
"total_flos": 1717079281606656.0,
|
| 1430 |
"train_batch_size": 1,
|
| 1431 |
"trial_name": null,
|
| 1432 |
"trial_params": null
|
checkpoint-200/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5176
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0882660315b6d44be153a534ef331a1a652a772eed227ca85ccc42c14c629577
|
| 3 |
size 5176
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5176
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0882660315b6d44be153a534ef331a1a652a772eed227ca85ccc42c14c629577
|
| 3 |
size 5176
|