Instructions to use ErrorAI/97b33462-44ae-4e18-a978-37f476a489f4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use ErrorAI/97b33462-44ae-4e18-a978-37f476a489f4 with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("beomi/polyglot-ko-12.8b-safetensors") model = PeftModel.from_pretrained(base_model, "ErrorAI/97b33462-44ae-4e18-a978-37f476a489f4") - Notebooks
- Google Colab
- Kaggle
Training in progress, step 144, checkpoint
Browse files
last-checkpoint/adapter_config.json
CHANGED
|
@@ -20,10 +20,10 @@
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
| 23 |
-
"dense_4h_to_h",
|
| 24 |
-
"dense_h_to_4h",
|
| 25 |
"dense",
|
| 26 |
-
"query_key_value"
|
|
|
|
|
|
|
| 27 |
],
|
| 28 |
"task_type": "CAUSAL_LM",
|
| 29 |
"use_dora": false,
|
|
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
|
|
|
|
|
|
| 23 |
"dense",
|
| 24 |
+
"query_key_value",
|
| 25 |
+
"dense_h_to_4h",
|
| 26 |
+
"dense_4h_to_h"
|
| 27 |
],
|
| 28 |
"task_type": "CAUSAL_LM",
|
| 29 |
"use_dora": false,
|
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 104902272
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7116b1c1314ad0972618f93cc3b28d151dda1ff0a92b623658450fd853da9040
|
| 3 |
size 104902272
|
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 sha256:5a7a0ab71ff7162a4bea3491cbf46e91dce78835f1dd98681fa1f3ed5d4dfb75
|
| 3 |
+
size 53622996
|
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 sha256:08753c2fa7a01c0b6662d53f842dd75c0945abf1fb5dd9c4697594f72a44af58
|
| 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 sha256:5e91368fc2ec2cb657c59caf19792cdc828c22dc3516914ebc4f3ddc1a5d7aa4
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,2044 +1,1027 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
-
"eval_steps":
|
| 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 |
-
"grad_norm": 0.
|
| 14 |
"learning_rate": 2e-05,
|
| 15 |
-
"loss": 13.
|
| 16 |
-
"step": 1
|
| 17 |
-
},
|
| 18 |
-
{
|
| 19 |
-
"epoch": 0.0008714596949891067,
|
| 20 |
-
"eval_loss": 3.2851531505584717,
|
| 21 |
-
"eval_runtime": 36.2933,
|
| 22 |
-
"eval_samples_per_second": 13.336,
|
| 23 |
-
"eval_steps_per_second": 6.668,
|
| 24 |
"step": 1
|
| 25 |
},
|
| 26 |
{
|
| 27 |
-
"epoch": 0.
|
| 28 |
-
"grad_norm": 0.
|
| 29 |
"learning_rate": 4e-05,
|
| 30 |
-
"loss": 12.
|
| 31 |
"step": 2
|
| 32 |
},
|
| 33 |
{
|
| 34 |
-
"epoch": 0.
|
| 35 |
-
"grad_norm": 0.
|
| 36 |
"learning_rate": 6e-05,
|
| 37 |
-
"loss":
|
| 38 |
"step": 3
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch": 0.
|
| 42 |
-
"grad_norm": 0.
|
| 43 |
"learning_rate": 8e-05,
|
| 44 |
-
"loss":
|
| 45 |
"step": 4
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 0.
|
| 49 |
-
"grad_norm": 0.
|
| 50 |
"learning_rate": 0.0001,
|
| 51 |
-
"loss": 12.
|
| 52 |
"step": 5
|
| 53 |
},
|
| 54 |
{
|
| 55 |
-
"epoch": 0.
|
| 56 |
-
"grad_norm": 1.
|
| 57 |
-
"learning_rate":
|
| 58 |
-
"loss":
|
| 59 |
"step": 6
|
| 60 |
},
|
| 61 |
{
|
| 62 |
-
"epoch": 0.
|
| 63 |
-
"grad_norm": 1.
|
| 64 |
-
"learning_rate":
|
| 65 |
-
"loss":
|
| 66 |
"step": 7
|
| 67 |
},
|
| 68 |
{
|
| 69 |
-
"epoch": 0.
|
| 70 |
-
"grad_norm": 1.
|
| 71 |
-
"learning_rate":
|
| 72 |
-
"loss": 12.
|
| 73 |
"step": 8
|
| 74 |
},
|
| 75 |
{
|
| 76 |
-
"epoch": 0.
|
| 77 |
-
"grad_norm": 2.
|
| 78 |
-
"learning_rate":
|
| 79 |
-
"loss":
|
| 80 |
"step": 9
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch": 0.
|
| 84 |
-
"grad_norm":
|
| 85 |
-
"learning_rate":
|
| 86 |
-
"loss": 11.
|
| 87 |
"step": 10
|
| 88 |
},
|
| 89 |
{
|
| 90 |
-
"epoch": 0.
|
| 91 |
-
"grad_norm":
|
| 92 |
-
"learning_rate":
|
| 93 |
-
"loss": 11.
|
| 94 |
"step": 11
|
| 95 |
},
|
| 96 |
{
|
| 97 |
-
"epoch": 0.
|
| 98 |
-
"grad_norm":
|
| 99 |
-
"learning_rate":
|
| 100 |
-
"loss": 10.
|
| 101 |
"step": 12
|
| 102 |
},
|
| 103 |
{
|
| 104 |
-
"epoch": 0.
|
| 105 |
-
"grad_norm": 3.
|
| 106 |
-
"learning_rate":
|
| 107 |
-
"loss": 10.
|
| 108 |
"step": 13
|
| 109 |
},
|
| 110 |
{
|
| 111 |
-
"epoch": 0.
|
| 112 |
-
"grad_norm":
|
| 113 |
-
"learning_rate":
|
| 114 |
-
"loss": 10.
|
| 115 |
"step": 14
|
| 116 |
},
|
| 117 |
{
|
| 118 |
-
"epoch": 0.
|
| 119 |
-
"grad_norm":
|
| 120 |
-
"learning_rate":
|
| 121 |
-
"loss":
|
| 122 |
"step": 15
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch": 0.
|
| 126 |
-
"grad_norm":
|
| 127 |
-
"learning_rate":
|
| 128 |
-
"loss":
|
| 129 |
"step": 16
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch": 0.
|
| 133 |
-
"grad_norm": 2.
|
| 134 |
-
"learning_rate":
|
| 135 |
-
"loss": 9.
|
| 136 |
"step": 17
|
| 137 |
},
|
| 138 |
{
|
| 139 |
-
"epoch": 0.
|
| 140 |
-
"grad_norm":
|
| 141 |
-
"learning_rate":
|
| 142 |
-
"loss":
|
| 143 |
"step": 18
|
| 144 |
},
|
| 145 |
{
|
| 146 |
-
"epoch": 0.
|
| 147 |
-
"grad_norm":
|
| 148 |
-
"learning_rate":
|
| 149 |
-
"loss":
|
| 150 |
"step": 19
|
| 151 |
},
|
| 152 |
{
|
| 153 |
-
"epoch": 0.
|
| 154 |
-
"grad_norm": 2.
|
| 155 |
-
"learning_rate":
|
| 156 |
-
"loss": 9.
|
| 157 |
"step": 20
|
| 158 |
},
|
| 159 |
{
|
| 160 |
-
"epoch": 0.
|
| 161 |
-
"grad_norm":
|
| 162 |
-
"learning_rate":
|
| 163 |
-
"loss":
|
| 164 |
"step": 21
|
| 165 |
},
|
| 166 |
{
|
| 167 |
-
"epoch": 0.
|
| 168 |
-
"grad_norm": 2.
|
| 169 |
-
"learning_rate":
|
| 170 |
-
"loss":
|
| 171 |
"step": 22
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch": 0.
|
| 175 |
-
"grad_norm":
|
| 176 |
-
"learning_rate":
|
| 177 |
-
"loss":
|
| 178 |
"step": 23
|
| 179 |
},
|
| 180 |
{
|
| 181 |
-
"epoch": 0.
|
| 182 |
-
"grad_norm":
|
| 183 |
-
"learning_rate":
|
| 184 |
-
"loss":
|
| 185 |
"step": 24
|
| 186 |
},
|
| 187 |
{
|
| 188 |
-
"epoch": 0.
|
| 189 |
-
"grad_norm":
|
| 190 |
-
"learning_rate":
|
| 191 |
-
"loss":
|
| 192 |
"step": 25
|
| 193 |
},
|
| 194 |
{
|
| 195 |
-
"epoch": 0.
|
| 196 |
-
"grad_norm":
|
| 197 |
-
"learning_rate":
|
| 198 |
-
"loss":
|
| 199 |
"step": 26
|
| 200 |
},
|
| 201 |
{
|
| 202 |
-
"epoch": 0.
|
| 203 |
-
"grad_norm":
|
| 204 |
-
"learning_rate":
|
| 205 |
-
"loss":
|
| 206 |
"step": 27
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch": 0.
|
| 210 |
-
"grad_norm":
|
| 211 |
-
"learning_rate":
|
| 212 |
-
"loss":
|
| 213 |
"step": 28
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch": 0.
|
| 217 |
-
"grad_norm":
|
| 218 |
-
"learning_rate":
|
| 219 |
-
"loss": 8.
|
| 220 |
"step": 29
|
| 221 |
},
|
| 222 |
{
|
| 223 |
-
"epoch": 0.
|
| 224 |
-
"grad_norm":
|
| 225 |
-
"learning_rate":
|
| 226 |
-
"loss": 8.
|
| 227 |
"step": 30
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch": 0.
|
| 231 |
-
"grad_norm":
|
| 232 |
-
"learning_rate":
|
| 233 |
-
"loss": 8.
|
| 234 |
"step": 31
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch": 0.
|
| 238 |
-
"grad_norm":
|
| 239 |
-
"learning_rate":
|
| 240 |
-
"loss": 8.
|
| 241 |
"step": 32
|
| 242 |
},
|
| 243 |
{
|
| 244 |
-
"epoch": 0.
|
| 245 |
-
"grad_norm":
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss": 8.
|
| 248 |
"step": 33
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch": 0.
|
| 252 |
-
"grad_norm":
|
| 253 |
-
"learning_rate":
|
| 254 |
-
"loss": 8.
|
| 255 |
"step": 34
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch": 0.
|
| 259 |
-
"grad_norm": 2.
|
| 260 |
-
"learning_rate":
|
| 261 |
-
"loss":
|
| 262 |
"step": 35
|
| 263 |
},
|
| 264 |
{
|
| 265 |
-
"epoch": 0.
|
| 266 |
-
"grad_norm":
|
| 267 |
-
"learning_rate":
|
| 268 |
-
"loss":
|
| 269 |
"step": 36
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch": 0.
|
| 273 |
-
"grad_norm": 2.
|
| 274 |
-
"learning_rate":
|
| 275 |
-
"loss": 8.
|
| 276 |
"step": 37
|
| 277 |
},
|
| 278 |
{
|
| 279 |
-
"epoch": 0.
|
| 280 |
-
"grad_norm": 2.
|
| 281 |
-
"learning_rate":
|
| 282 |
-
"loss":
|
| 283 |
"step": 38
|
| 284 |
},
|
| 285 |
{
|
| 286 |
-
"epoch": 0.
|
| 287 |
-
"grad_norm": 2.
|
| 288 |
-
"learning_rate":
|
| 289 |
-
"loss": 7.
|
| 290 |
"step": 39
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"epoch": 0.
|
| 294 |
-
"grad_norm": 2.
|
| 295 |
-
"learning_rate":
|
| 296 |
-
"loss":
|
| 297 |
"step": 40
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch": 0.
|
| 301 |
-
"grad_norm":
|
| 302 |
-
"learning_rate":
|
| 303 |
-
"loss": 7.
|
| 304 |
"step": 41
|
| 305 |
},
|
| 306 |
{
|
| 307 |
-
"epoch": 0.
|
| 308 |
-
"grad_norm":
|
| 309 |
-
"learning_rate":
|
| 310 |
-
"loss":
|
| 311 |
"step": 42
|
| 312 |
},
|
| 313 |
{
|
| 314 |
-
"epoch": 0.
|
| 315 |
-
"grad_norm": 2.
|
| 316 |
-
"learning_rate":
|
| 317 |
-
"loss":
|
| 318 |
"step": 43
|
| 319 |
},
|
| 320 |
{
|
| 321 |
-
"epoch": 0.
|
| 322 |
-
"grad_norm":
|
| 323 |
-
"learning_rate":
|
| 324 |
-
"loss":
|
| 325 |
"step": 44
|
| 326 |
},
|
| 327 |
{
|
| 328 |
-
"epoch": 0.
|
| 329 |
-
"grad_norm": 2.
|
| 330 |
-
"learning_rate":
|
| 331 |
-
"loss": 7.
|
| 332 |
"step": 45
|
| 333 |
},
|
| 334 |
{
|
| 335 |
-
"epoch": 0.
|
| 336 |
-
"grad_norm":
|
| 337 |
-
"learning_rate":
|
| 338 |
-
"loss": 7.
|
| 339 |
"step": 46
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch": 0.
|
| 343 |
-
"grad_norm":
|
| 344 |
-
"learning_rate":
|
| 345 |
-
"loss": 7.
|
| 346 |
"step": 47
|
| 347 |
},
|
| 348 |
{
|
| 349 |
-
"epoch": 0.
|
| 350 |
-
"grad_norm":
|
| 351 |
-
"learning_rate":
|
| 352 |
-
"loss": 7.
|
| 353 |
"step": 48
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch": 0.
|
| 357 |
-
"grad_norm": 3.
|
| 358 |
-
"learning_rate":
|
| 359 |
-
"loss":
|
| 360 |
"step": 49
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch": 0.
|
| 364 |
-
"grad_norm":
|
| 365 |
-
"learning_rate":
|
| 366 |
-
"loss":
|
| 367 |
"step": 50
|
| 368 |
},
|
| 369 |
{
|
| 370 |
-
"epoch": 0.
|
| 371 |
-
"grad_norm":
|
| 372 |
-
"learning_rate":
|
| 373 |
-
"loss":
|
| 374 |
"step": 51
|
| 375 |
},
|
| 376 |
{
|
| 377 |
-
"epoch": 0.
|
| 378 |
-
"grad_norm": 3.
|
| 379 |
-
"learning_rate":
|
| 380 |
-
"loss": 7.
|
| 381 |
"step": 52
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch": 0.
|
| 385 |
-
"grad_norm":
|
| 386 |
-
"learning_rate":
|
| 387 |
-
"loss":
|
| 388 |
"step": 53
|
| 389 |
},
|
| 390 |
{
|
| 391 |
-
"epoch": 0.
|
| 392 |
-
"grad_norm":
|
| 393 |
-
"learning_rate":
|
| 394 |
-
"loss":
|
| 395 |
"step": 54
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch": 0.
|
| 399 |
-
"grad_norm":
|
| 400 |
-
"learning_rate":
|
| 401 |
-
"loss": 7.
|
| 402 |
"step": 55
|
| 403 |
},
|
| 404 |
{
|
| 405 |
-
"epoch": 0.
|
| 406 |
-
"grad_norm":
|
| 407 |
-
"learning_rate":
|
| 408 |
-
"loss":
|
| 409 |
"step": 56
|
| 410 |
},
|
| 411 |
{
|
| 412 |
-
"epoch": 0.
|
| 413 |
-
"grad_norm":
|
| 414 |
-
"learning_rate":
|
| 415 |
-
"loss":
|
| 416 |
"step": 57
|
| 417 |
},
|
| 418 |
{
|
| 419 |
-
"epoch": 0.
|
| 420 |
-
"grad_norm":
|
| 421 |
-
"learning_rate":
|
| 422 |
-
"loss":
|
| 423 |
"step": 58
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"epoch": 0.
|
| 427 |
-
"grad_norm":
|
| 428 |
-
"learning_rate":
|
| 429 |
-
"loss":
|
| 430 |
"step": 59
|
| 431 |
},
|
| 432 |
{
|
| 433 |
-
"epoch": 0.
|
| 434 |
-
"grad_norm":
|
| 435 |
-
"learning_rate":
|
| 436 |
-
"loss":
|
| 437 |
"step": 60
|
| 438 |
},
|
| 439 |
{
|
| 440 |
-
"epoch": 0.
|
| 441 |
-
"grad_norm":
|
| 442 |
-
"learning_rate":
|
| 443 |
-
"loss": 7.
|
| 444 |
"step": 61
|
| 445 |
},
|
| 446 |
{
|
| 447 |
-
"epoch": 0.
|
| 448 |
-
"grad_norm":
|
| 449 |
-
"learning_rate":
|
| 450 |
-
"loss": 7.
|
| 451 |
"step": 62
|
| 452 |
},
|
| 453 |
{
|
| 454 |
-
"epoch": 0.
|
| 455 |
-
"grad_norm":
|
| 456 |
-
"learning_rate":
|
| 457 |
-
"loss":
|
| 458 |
"step": 63
|
| 459 |
},
|
| 460 |
{
|
| 461 |
-
"epoch": 0.
|
| 462 |
-
"grad_norm":
|
| 463 |
-
"learning_rate":
|
| 464 |
-
"loss":
|
| 465 |
"step": 64
|
| 466 |
},
|
| 467 |
{
|
| 468 |
-
"epoch": 0.
|
| 469 |
-
"grad_norm":
|
| 470 |
-
"learning_rate":
|
| 471 |
-
"loss": 6.
|
| 472 |
"step": 65
|
| 473 |
},
|
| 474 |
{
|
| 475 |
-
"epoch": 0.
|
| 476 |
-
"grad_norm":
|
| 477 |
-
"learning_rate":
|
| 478 |
-
"loss":
|
| 479 |
"step": 66
|
| 480 |
},
|
| 481 |
{
|
| 482 |
-
"epoch": 0.
|
| 483 |
-
"grad_norm":
|
| 484 |
-
"learning_rate":
|
| 485 |
-
"loss": 7.
|
| 486 |
"step": 67
|
| 487 |
},
|
| 488 |
{
|
| 489 |
-
"epoch": 0.
|
| 490 |
-
"grad_norm":
|
| 491 |
-
"learning_rate":
|
| 492 |
-
"loss":
|
| 493 |
"step": 68
|
| 494 |
},
|
| 495 |
{
|
| 496 |
-
"epoch": 0.
|
| 497 |
-
"grad_norm":
|
| 498 |
-
"learning_rate":
|
| 499 |
-
"loss":
|
| 500 |
"step": 69
|
| 501 |
},
|
| 502 |
{
|
| 503 |
-
"epoch": 0.
|
| 504 |
-
"grad_norm":
|
| 505 |
-
"learning_rate":
|
| 506 |
-
"loss": 6.
|
| 507 |
"step": 70
|
| 508 |
},
|
| 509 |
{
|
| 510 |
-
"epoch": 0.
|
| 511 |
-
"grad_norm":
|
| 512 |
-
"learning_rate":
|
| 513 |
-
"loss": 6.
|
| 514 |
"step": 71
|
| 515 |
},
|
| 516 |
{
|
| 517 |
-
"epoch": 0.
|
| 518 |
-
"grad_norm":
|
| 519 |
-
"learning_rate":
|
| 520 |
-
"loss": 6.
|
| 521 |
"step": 72
|
| 522 |
},
|
| 523 |
{
|
| 524 |
-
"epoch": 0.
|
| 525 |
-
"grad_norm":
|
| 526 |
-
"learning_rate":
|
| 527 |
-
"loss":
|
| 528 |
"step": 73
|
| 529 |
},
|
| 530 |
{
|
| 531 |
-
"epoch": 0.
|
| 532 |
-
"grad_norm":
|
| 533 |
-
"learning_rate":
|
| 534 |
-
"loss": 6.
|
| 535 |
"step": 74
|
| 536 |
},
|
| 537 |
{
|
| 538 |
-
"epoch": 0.
|
| 539 |
-
"grad_norm": 2.
|
| 540 |
-
"learning_rate":
|
| 541 |
-
"loss": 6.
|
| 542 |
"step": 75
|
| 543 |
},
|
| 544 |
{
|
| 545 |
-
"epoch": 0.
|
| 546 |
-
"grad_norm":
|
| 547 |
-
"learning_rate":
|
| 548 |
-
"loss": 6.
|
| 549 |
"step": 76
|
| 550 |
},
|
| 551 |
{
|
| 552 |
-
"epoch": 0.
|
| 553 |
-
"grad_norm":
|
| 554 |
-
"learning_rate":
|
| 555 |
-
"loss":
|
| 556 |
"step": 77
|
| 557 |
},
|
| 558 |
{
|
| 559 |
-
"epoch": 0.
|
| 560 |
-
"grad_norm":
|
| 561 |
-
"learning_rate":
|
| 562 |
-
"loss": 6.
|
| 563 |
"step": 78
|
| 564 |
},
|
| 565 |
{
|
| 566 |
-
"epoch": 0.
|
| 567 |
-
"grad_norm":
|
| 568 |
-
"learning_rate":
|
| 569 |
-
"loss":
|
| 570 |
"step": 79
|
| 571 |
},
|
| 572 |
{
|
| 573 |
-
"epoch": 0.
|
| 574 |
-
"grad_norm":
|
| 575 |
-
"learning_rate":
|
| 576 |
-
"loss": 6.
|
| 577 |
"step": 80
|
| 578 |
},
|
| 579 |
{
|
| 580 |
-
"epoch": 0.
|
| 581 |
-
"grad_norm":
|
| 582 |
-
"learning_rate":
|
| 583 |
-
"loss": 6.
|
| 584 |
"step": 81
|
| 585 |
},
|
| 586 |
{
|
| 587 |
-
"epoch": 0.
|
| 588 |
-
"grad_norm": 3.
|
| 589 |
-
"learning_rate":
|
| 590 |
-
"loss": 6.
|
| 591 |
"step": 82
|
| 592 |
},
|
| 593 |
{
|
| 594 |
-
"epoch": 0.
|
| 595 |
-
"grad_norm":
|
| 596 |
-
"learning_rate":
|
| 597 |
-
"loss": 6.
|
| 598 |
"step": 83
|
| 599 |
},
|
| 600 |
{
|
| 601 |
-
"epoch": 0.
|
| 602 |
-
"grad_norm":
|
| 603 |
-
"learning_rate":
|
| 604 |
-
"loss": 6.
|
| 605 |
"step": 84
|
| 606 |
},
|
| 607 |
{
|
| 608 |
-
"epoch": 0.
|
| 609 |
-
"grad_norm":
|
| 610 |
-
"learning_rate":
|
| 611 |
-
"loss":
|
| 612 |
"step": 85
|
| 613 |
},
|
| 614 |
{
|
| 615 |
-
"epoch": 0.
|
| 616 |
-
"grad_norm": 3.
|
| 617 |
-
"learning_rate":
|
| 618 |
-
"loss": 6.
|
| 619 |
"step": 86
|
| 620 |
},
|
| 621 |
{
|
| 622 |
-
"epoch": 0.
|
| 623 |
-
"grad_norm":
|
| 624 |
-
"learning_rate":
|
| 625 |
-
"loss": 6.
|
| 626 |
"step": 87
|
| 627 |
},
|
| 628 |
{
|
| 629 |
-
"epoch": 0.
|
| 630 |
-
"grad_norm": 3.
|
| 631 |
-
"learning_rate":
|
| 632 |
-
"loss": 6.
|
| 633 |
"step": 88
|
| 634 |
},
|
| 635 |
{
|
| 636 |
-
"epoch": 0.
|
| 637 |
-
"grad_norm": 3.
|
| 638 |
-
"learning_rate":
|
| 639 |
-
"loss": 6.
|
| 640 |
"step": 89
|
| 641 |
},
|
| 642 |
{
|
| 643 |
-
"epoch": 0.
|
| 644 |
-
"grad_norm": 3.
|
| 645 |
-
"learning_rate":
|
| 646 |
-
"loss":
|
| 647 |
"step": 90
|
| 648 |
},
|
| 649 |
{
|
| 650 |
-
"epoch": 0.
|
| 651 |
-
"grad_norm": 3.
|
| 652 |
-
"learning_rate":
|
| 653 |
-
"loss":
|
| 654 |
"step": 91
|
| 655 |
},
|
| 656 |
{
|
| 657 |
-
"epoch": 0.
|
| 658 |
-
"grad_norm":
|
| 659 |
-
"learning_rate":
|
| 660 |
-
"loss": 6.
|
| 661 |
"step": 92
|
| 662 |
},
|
| 663 |
{
|
| 664 |
-
"epoch": 0.
|
| 665 |
-
"grad_norm": 2.
|
| 666 |
-
"learning_rate":
|
| 667 |
-
"loss": 6.
|
| 668 |
"step": 93
|
| 669 |
},
|
| 670 |
{
|
| 671 |
-
"epoch": 0.
|
| 672 |
-
"grad_norm": 3.
|
| 673 |
-
"learning_rate":
|
| 674 |
-
"loss":
|
| 675 |
"step": 94
|
| 676 |
},
|
| 677 |
{
|
| 678 |
-
"epoch": 0.
|
| 679 |
-
"grad_norm":
|
| 680 |
-
"learning_rate":
|
| 681 |
-
"loss": 6.
|
| 682 |
"step": 95
|
| 683 |
},
|
| 684 |
{
|
| 685 |
-
"epoch": 0.
|
| 686 |
-
"grad_norm":
|
| 687 |
-
"learning_rate":
|
| 688 |
-
"loss": 6.
|
| 689 |
"step": 96
|
| 690 |
},
|
| 691 |
{
|
| 692 |
-
"epoch": 0.
|
| 693 |
-
"grad_norm": 3.
|
| 694 |
-
"learning_rate":
|
| 695 |
-
"loss": 6.
|
| 696 |
"step": 97
|
| 697 |
},
|
| 698 |
{
|
| 699 |
-
"epoch": 0.
|
| 700 |
-
"grad_norm":
|
| 701 |
-
"learning_rate":
|
| 702 |
-
"loss":
|
| 703 |
"step": 98
|
| 704 |
},
|
| 705 |
{
|
| 706 |
-
"epoch": 0.
|
| 707 |
-
"grad_norm":
|
| 708 |
-
"learning_rate":
|
| 709 |
-
"loss": 6.
|
| 710 |
"step": 99
|
| 711 |
},
|
| 712 |
{
|
| 713 |
-
"epoch": 0.
|
| 714 |
-
"grad_norm":
|
| 715 |
-
"learning_rate":
|
| 716 |
-
"loss":
|
| 717 |
"step": 100
|
| 718 |
},
|
| 719 |
{
|
| 720 |
-
"epoch": 0.
|
| 721 |
-
"grad_norm":
|
| 722 |
-
"learning_rate":
|
| 723 |
-
"loss": 6.
|
| 724 |
"step": 101
|
| 725 |
},
|
| 726 |
{
|
| 727 |
-
"epoch": 0.
|
| 728 |
-
"grad_norm": 3.
|
| 729 |
-
"learning_rate":
|
| 730 |
-
"loss": 6.
|
| 731 |
"step": 102
|
| 732 |
},
|
| 733 |
{
|
| 734 |
-
"epoch": 0.
|
| 735 |
-
"grad_norm":
|
| 736 |
-
"learning_rate":
|
| 737 |
-
"loss":
|
| 738 |
"step": 103
|
| 739 |
},
|
| 740 |
{
|
| 741 |
-
"epoch": 0.
|
| 742 |
-
"grad_norm":
|
| 743 |
-
"learning_rate":
|
| 744 |
-
"loss":
|
| 745 |
"step": 104
|
| 746 |
},
|
| 747 |
{
|
| 748 |
-
"epoch": 0.
|
| 749 |
-
"grad_norm":
|
| 750 |
-
"learning_rate":
|
| 751 |
-
"loss":
|
| 752 |
"step": 105
|
| 753 |
},
|
| 754 |
{
|
| 755 |
-
"epoch": 0.
|
| 756 |
-
"grad_norm":
|
| 757 |
-
"learning_rate":
|
| 758 |
-
"loss":
|
| 759 |
"step": 106
|
| 760 |
},
|
| 761 |
{
|
| 762 |
-
"epoch": 0.
|
| 763 |
-
"grad_norm":
|
| 764 |
-
"learning_rate":
|
| 765 |
-
"loss": 6.
|
| 766 |
"step": 107
|
| 767 |
},
|
| 768 |
{
|
| 769 |
-
"epoch": 0.
|
| 770 |
-
"grad_norm":
|
| 771 |
-
"learning_rate":
|
| 772 |
-
"loss": 7.
|
| 773 |
"step": 108
|
| 774 |
},
|
| 775 |
{
|
| 776 |
-
"epoch": 0.
|
| 777 |
-
"grad_norm":
|
| 778 |
-
"learning_rate":
|
| 779 |
-
"loss": 6.
|
| 780 |
"step": 109
|
| 781 |
},
|
| 782 |
{
|
| 783 |
-
"epoch": 0.
|
| 784 |
-
"grad_norm":
|
| 785 |
-
"learning_rate":
|
| 786 |
-
"loss": 6.
|
| 787 |
"step": 110
|
| 788 |
},
|
| 789 |
{
|
| 790 |
-
"epoch": 0.
|
| 791 |
-
"grad_norm":
|
| 792 |
-
"learning_rate":
|
| 793 |
-
"loss":
|
| 794 |
"step": 111
|
| 795 |
},
|
| 796 |
{
|
| 797 |
-
"epoch": 0.
|
| 798 |
-
"grad_norm":
|
| 799 |
-
"learning_rate":
|
| 800 |
-
"loss": 6.
|
| 801 |
"step": 112
|
| 802 |
},
|
| 803 |
{
|
| 804 |
-
"epoch": 0.
|
| 805 |
-
"grad_norm":
|
| 806 |
-
"learning_rate":
|
| 807 |
-
"loss": 6.
|
| 808 |
"step": 113
|
| 809 |
},
|
| 810 |
{
|
| 811 |
-
"epoch": 0.
|
| 812 |
-
"grad_norm":
|
| 813 |
-
"learning_rate":
|
| 814 |
-
"loss": 6.
|
| 815 |
"step": 114
|
| 816 |
},
|
| 817 |
{
|
| 818 |
-
"epoch": 0.
|
| 819 |
-
"grad_norm":
|
| 820 |
-
"learning_rate":
|
| 821 |
-
"loss": 6.
|
| 822 |
"step": 115
|
| 823 |
},
|
| 824 |
{
|
| 825 |
-
"epoch": 0.
|
| 826 |
-
"grad_norm":
|
| 827 |
-
"learning_rate":
|
| 828 |
-
"loss": 6.
|
| 829 |
"step": 116
|
| 830 |
},
|
| 831 |
{
|
| 832 |
-
"epoch": 0.
|
| 833 |
-
"grad_norm":
|
| 834 |
-
"learning_rate":
|
| 835 |
-
"loss":
|
| 836 |
"step": 117
|
| 837 |
},
|
| 838 |
{
|
| 839 |
-
"epoch": 0.
|
| 840 |
-
"grad_norm":
|
| 841 |
-
"learning_rate":
|
| 842 |
-
"loss": 6.
|
| 843 |
"step": 118
|
| 844 |
},
|
| 845 |
{
|
| 846 |
-
"epoch": 0.
|
| 847 |
-
"grad_norm":
|
| 848 |
-
"learning_rate":
|
| 849 |
-
"loss":
|
| 850 |
"step": 119
|
| 851 |
},
|
| 852 |
{
|
| 853 |
-
"epoch": 0.
|
| 854 |
-
"grad_norm":
|
| 855 |
-
"learning_rate":
|
| 856 |
-
"loss": 6.
|
| 857 |
"step": 120
|
| 858 |
},
|
| 859 |
{
|
| 860 |
-
"epoch": 0.
|
| 861 |
-
"grad_norm":
|
| 862 |
-
"learning_rate":
|
| 863 |
-
"loss": 6.
|
| 864 |
"step": 121
|
| 865 |
},
|
| 866 |
{
|
| 867 |
-
"epoch": 0.
|
| 868 |
-
"grad_norm":
|
| 869 |
-
"learning_rate":
|
| 870 |
-
"loss": 6.
|
| 871 |
"step": 122
|
| 872 |
},
|
| 873 |
{
|
| 874 |
-
"epoch": 0.
|
| 875 |
-
"grad_norm":
|
| 876 |
-
"learning_rate":
|
| 877 |
-
"loss":
|
| 878 |
"step": 123
|
| 879 |
},
|
| 880 |
{
|
| 881 |
-
"epoch": 0.
|
| 882 |
-
"grad_norm":
|
| 883 |
-
"learning_rate":
|
| 884 |
-
"loss": 6.
|
| 885 |
"step": 124
|
| 886 |
},
|
| 887 |
{
|
| 888 |
-
"epoch": 0.
|
| 889 |
-
"grad_norm":
|
| 890 |
-
"learning_rate":
|
| 891 |
-
"loss":
|
| 892 |
"step": 125
|
| 893 |
},
|
| 894 |
{
|
| 895 |
-
"epoch": 0.
|
| 896 |
-
"grad_norm":
|
| 897 |
-
"learning_rate":
|
| 898 |
-
"loss": 6.
|
| 899 |
"step": 126
|
| 900 |
},
|
| 901 |
{
|
| 902 |
-
"epoch": 0.
|
| 903 |
-
"grad_norm":
|
| 904 |
-
"learning_rate":
|
| 905 |
-
"loss": 6.
|
| 906 |
"step": 127
|
| 907 |
},
|
| 908 |
{
|
| 909 |
-
"epoch": 0.
|
| 910 |
-
"grad_norm":
|
| 911 |
-
"learning_rate":
|
| 912 |
-
"loss": 6.
|
| 913 |
"step": 128
|
| 914 |
},
|
| 915 |
{
|
| 916 |
-
"epoch": 0.
|
| 917 |
-
"grad_norm":
|
| 918 |
-
"learning_rate":
|
| 919 |
-
"loss":
|
| 920 |
"step": 129
|
| 921 |
},
|
| 922 |
{
|
| 923 |
-
"epoch": 0.
|
| 924 |
-
"grad_norm":
|
| 925 |
-
"learning_rate":
|
| 926 |
-
"loss": 6.
|
| 927 |
"step": 130
|
| 928 |
},
|
| 929 |
{
|
| 930 |
-
"epoch": 0.
|
| 931 |
-
"grad_norm": 3.
|
| 932 |
-
"learning_rate":
|
| 933 |
-
"loss": 6.
|
| 934 |
"step": 131
|
| 935 |
},
|
| 936 |
{
|
| 937 |
-
"epoch": 0.
|
| 938 |
-
"grad_norm": 2.
|
| 939 |
-
"learning_rate":
|
| 940 |
-
"loss": 6.
|
| 941 |
"step": 132
|
| 942 |
},
|
| 943 |
{
|
| 944 |
-
"epoch": 0.
|
| 945 |
-
"grad_norm":
|
| 946 |
-
"learning_rate":
|
| 947 |
-
"loss": 6.
|
| 948 |
"step": 133
|
| 949 |
},
|
| 950 |
{
|
| 951 |
-
"epoch": 0.
|
| 952 |
-
"grad_norm":
|
| 953 |
-
"learning_rate":
|
| 954 |
-
"loss":
|
| 955 |
"step": 134
|
| 956 |
},
|
| 957 |
{
|
| 958 |
-
"epoch": 0.
|
| 959 |
-
"grad_norm":
|
| 960 |
-
"learning_rate":
|
| 961 |
-
"loss":
|
| 962 |
"step": 135
|
| 963 |
},
|
| 964 |
{
|
| 965 |
-
"epoch": 0.
|
| 966 |
-
"grad_norm": 2.
|
| 967 |
-
"learning_rate":
|
| 968 |
-
"loss": 6.
|
| 969 |
"step": 136
|
| 970 |
},
|
| 971 |
{
|
| 972 |
-
"epoch": 0.
|
| 973 |
-
"grad_norm": 2.
|
| 974 |
-
"learning_rate":
|
| 975 |
-
"loss": 6.
|
| 976 |
"step": 137
|
| 977 |
},
|
| 978 |
{
|
| 979 |
-
"epoch": 0.
|
| 980 |
-
"grad_norm": 3.
|
| 981 |
-
"learning_rate":
|
| 982 |
-
"loss": 6.
|
| 983 |
"step": 138
|
| 984 |
},
|
| 985 |
{
|
| 986 |
-
"epoch": 0.
|
| 987 |
-
"grad_norm": 3.
|
| 988 |
-
"learning_rate":
|
| 989 |
-
"loss": 6.
|
| 990 |
"step": 139
|
| 991 |
},
|
| 992 |
{
|
| 993 |
-
"epoch": 0.
|
| 994 |
-
"grad_norm": 3.
|
| 995 |
-
"learning_rate":
|
| 996 |
-
"loss": 7.
|
| 997 |
"step": 140
|
| 998 |
},
|
| 999 |
{
|
| 1000 |
-
"epoch": 0.
|
| 1001 |
-
"grad_norm": 3.
|
| 1002 |
-
"learning_rate":
|
| 1003 |
-
"loss": 6.
|
| 1004 |
"step": 141
|
| 1005 |
},
|
| 1006 |
{
|
| 1007 |
-
"epoch": 0.
|
| 1008 |
-
"grad_norm": 3.
|
| 1009 |
-
"learning_rate":
|
| 1010 |
-
"loss":
|
| 1011 |
"step": 142
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
-
"epoch": 0.
|
| 1015 |
-
"grad_norm": 3.
|
| 1016 |
-
"learning_rate":
|
| 1017 |
-
"loss":
|
| 1018 |
"step": 143
|
| 1019 |
},
|
| 1020 |
{
|
| 1021 |
-
"epoch": 0.
|
| 1022 |
-
"grad_norm": 3.
|
| 1023 |
-
"learning_rate":
|
| 1024 |
-
"loss":
|
| 1025 |
"step": 144
|
| 1026 |
-
},
|
| 1027 |
-
{
|
| 1028 |
-
"epoch": 0.12636165577342048,
|
| 1029 |
-
"grad_norm": 3.450144052505493,
|
| 1030 |
-
"learning_rate": 0.0001931353215606343,
|
| 1031 |
-
"loss": 6.2738,
|
| 1032 |
-
"step": 145
|
| 1033 |
-
},
|
| 1034 |
-
{
|
| 1035 |
-
"epoch": 0.1272331154684096,
|
| 1036 |
-
"grad_norm": 3.2067205905914307,
|
| 1037 |
-
"learning_rate": 0.00019303444770023184,
|
| 1038 |
-
"loss": 6.1241,
|
| 1039 |
-
"step": 146
|
| 1040 |
-
},
|
| 1041 |
-
{
|
| 1042 |
-
"epoch": 0.12810457516339868,
|
| 1043 |
-
"grad_norm": 3.24552059173584,
|
| 1044 |
-
"learning_rate": 0.0001929328648194321,
|
| 1045 |
-
"loss": 6.5917,
|
| 1046 |
-
"step": 147
|
| 1047 |
-
},
|
| 1048 |
-
{
|
| 1049 |
-
"epoch": 0.1289760348583878,
|
| 1050 |
-
"grad_norm": 3.2059834003448486,
|
| 1051 |
-
"learning_rate": 0.00019283057369240357,
|
| 1052 |
-
"loss": 6.7524,
|
| 1053 |
-
"step": 148
|
| 1054 |
-
},
|
| 1055 |
-
{
|
| 1056 |
-
"epoch": 0.1298474945533769,
|
| 1057 |
-
"grad_norm": 3.5542874336242676,
|
| 1058 |
-
"learning_rate": 0.00019272757509871232,
|
| 1059 |
-
"loss": 6.6428,
|
| 1060 |
-
"step": 149
|
| 1061 |
-
},
|
| 1062 |
-
{
|
| 1063 |
-
"epoch": 0.13071895424836602,
|
| 1064 |
-
"grad_norm": 3.5958824157714844,
|
| 1065 |
-
"learning_rate": 0.00019262386982331594,
|
| 1066 |
-
"loss": 6.1798,
|
| 1067 |
-
"step": 150
|
| 1068 |
-
},
|
| 1069 |
-
{
|
| 1070 |
-
"epoch": 0.13159041394335513,
|
| 1071 |
-
"grad_norm": 2.9370152950286865,
|
| 1072 |
-
"learning_rate": 0.00019251945865655782,
|
| 1073 |
-
"loss": 5.6334,
|
| 1074 |
-
"step": 151
|
| 1075 |
-
},
|
| 1076 |
-
{
|
| 1077 |
-
"epoch": 0.13246187363834422,
|
| 1078 |
-
"grad_norm": 3.0617597103118896,
|
| 1079 |
-
"learning_rate": 0.00019241434239416095,
|
| 1080 |
-
"loss": 6.4162,
|
| 1081 |
-
"step": 152
|
| 1082 |
-
},
|
| 1083 |
-
{
|
| 1084 |
-
"epoch": 0.13333333333333333,
|
| 1085 |
-
"grad_norm": 3.3482141494750977,
|
| 1086 |
-
"learning_rate": 0.00019230852183722182,
|
| 1087 |
-
"loss": 6.6717,
|
| 1088 |
-
"step": 153
|
| 1089 |
-
},
|
| 1090 |
-
{
|
| 1091 |
-
"epoch": 0.13420479302832244,
|
| 1092 |
-
"grad_norm": 3.2366411685943604,
|
| 1093 |
-
"learning_rate": 0.00019220199779220446,
|
| 1094 |
-
"loss": 6.7417,
|
| 1095 |
-
"step": 154
|
| 1096 |
-
},
|
| 1097 |
-
{
|
| 1098 |
-
"epoch": 0.13507625272331156,
|
| 1099 |
-
"grad_norm": 3.4498870372772217,
|
| 1100 |
-
"learning_rate": 0.0001920947710709342,
|
| 1101 |
-
"loss": 6.6616,
|
| 1102 |
-
"step": 155
|
| 1103 |
-
},
|
| 1104 |
-
{
|
| 1105 |
-
"epoch": 0.13594771241830064,
|
| 1106 |
-
"grad_norm": 3.071847438812256,
|
| 1107 |
-
"learning_rate": 0.0001919868424905915,
|
| 1108 |
-
"loss": 7.1704,
|
| 1109 |
-
"step": 156
|
| 1110 |
-
},
|
| 1111 |
-
{
|
| 1112 |
-
"epoch": 0.13681917211328976,
|
| 1113 |
-
"grad_norm": 2.8072800636291504,
|
| 1114 |
-
"learning_rate": 0.00019187821287370568,
|
| 1115 |
-
"loss": 6.2717,
|
| 1116 |
-
"step": 157
|
| 1117 |
-
},
|
| 1118 |
-
{
|
| 1119 |
-
"epoch": 0.13769063180827887,
|
| 1120 |
-
"grad_norm": 2.722471237182617,
|
| 1121 |
-
"learning_rate": 0.00019176888304814883,
|
| 1122 |
-
"loss": 6.1792,
|
| 1123 |
-
"step": 158
|
| 1124 |
-
},
|
| 1125 |
-
{
|
| 1126 |
-
"epoch": 0.13856209150326798,
|
| 1127 |
-
"grad_norm": 3.434514284133911,
|
| 1128 |
-
"learning_rate": 0.00019165885384712916,
|
| 1129 |
-
"loss": 6.574,
|
| 1130 |
-
"step": 159
|
| 1131 |
-
},
|
| 1132 |
-
{
|
| 1133 |
-
"epoch": 0.1394335511982571,
|
| 1134 |
-
"grad_norm": 3.995135545730591,
|
| 1135 |
-
"learning_rate": 0.00019154812610918501,
|
| 1136 |
-
"loss": 5.9547,
|
| 1137 |
-
"step": 160
|
| 1138 |
-
},
|
| 1139 |
-
{
|
| 1140 |
-
"epoch": 0.14030501089324618,
|
| 1141 |
-
"grad_norm": 3.9248478412628174,
|
| 1142 |
-
"learning_rate": 0.0001914367006781783,
|
| 1143 |
-
"loss": 6.9711,
|
| 1144 |
-
"step": 161
|
| 1145 |
-
},
|
| 1146 |
-
{
|
| 1147 |
-
"epoch": 0.1411764705882353,
|
| 1148 |
-
"grad_norm": 3.2751011848449707,
|
| 1149 |
-
"learning_rate": 0.00019132457840328811,
|
| 1150 |
-
"loss": 6.3862,
|
| 1151 |
-
"step": 162
|
| 1152 |
-
},
|
| 1153 |
-
{
|
| 1154 |
-
"epoch": 0.1420479302832244,
|
| 1155 |
-
"grad_norm": 2.9850356578826904,
|
| 1156 |
-
"learning_rate": 0.0001912117601390041,
|
| 1157 |
-
"loss": 6.1286,
|
| 1158 |
-
"step": 163
|
| 1159 |
-
},
|
| 1160 |
-
{
|
| 1161 |
-
"epoch": 0.14291938997821352,
|
| 1162 |
-
"grad_norm": 3.267690896987915,
|
| 1163 |
-
"learning_rate": 0.00019109824674512014,
|
| 1164 |
-
"loss": 6.1965,
|
| 1165 |
-
"step": 164
|
| 1166 |
-
},
|
| 1167 |
-
{
|
| 1168 |
-
"epoch": 0.1437908496732026,
|
| 1169 |
-
"grad_norm": 3.1468505859375,
|
| 1170 |
-
"learning_rate": 0.0001909840390867278,
|
| 1171 |
-
"loss": 6.865,
|
| 1172 |
-
"step": 165
|
| 1173 |
-
},
|
| 1174 |
-
{
|
| 1175 |
-
"epoch": 0.14466230936819172,
|
| 1176 |
-
"grad_norm": 2.709254503250122,
|
| 1177 |
-
"learning_rate": 0.00019086913803420966,
|
| 1178 |
-
"loss": 6.2319,
|
| 1179 |
-
"step": 166
|
| 1180 |
-
},
|
| 1181 |
-
{
|
| 1182 |
-
"epoch": 0.14553376906318083,
|
| 1183 |
-
"grad_norm": 3.203265905380249,
|
| 1184 |
-
"learning_rate": 0.0001907535444632326,
|
| 1185 |
-
"loss": 6.0057,
|
| 1186 |
-
"step": 167
|
| 1187 |
-
},
|
| 1188 |
-
{
|
| 1189 |
-
"epoch": 0.14640522875816994,
|
| 1190 |
-
"grad_norm": 2.994889736175537,
|
| 1191 |
-
"learning_rate": 0.0001906372592547413,
|
| 1192 |
-
"loss": 6.7063,
|
| 1193 |
-
"step": 168
|
| 1194 |
-
},
|
| 1195 |
-
{
|
| 1196 |
-
"epoch": 0.14727668845315905,
|
| 1197 |
-
"grad_norm": 3.4088401794433594,
|
| 1198 |
-
"learning_rate": 0.0001905202832949514,
|
| 1199 |
-
"loss": 5.8675,
|
| 1200 |
-
"step": 169
|
| 1201 |
-
},
|
| 1202 |
-
{
|
| 1203 |
-
"epoch": 0.14814814814814814,
|
| 1204 |
-
"grad_norm": 3.7288787364959717,
|
| 1205 |
-
"learning_rate": 0.00019040261747534283,
|
| 1206 |
-
"loss": 5.8542,
|
| 1207 |
-
"step": 170
|
| 1208 |
-
},
|
| 1209 |
-
{
|
| 1210 |
-
"epoch": 0.14901960784313725,
|
| 1211 |
-
"grad_norm": 4.366305828094482,
|
| 1212 |
-
"learning_rate": 0.00019028426269265299,
|
| 1213 |
-
"loss": 6.0556,
|
| 1214 |
-
"step": 171
|
| 1215 |
-
},
|
| 1216 |
-
{
|
| 1217 |
-
"epoch": 0.14989106753812637,
|
| 1218 |
-
"grad_norm": 3.3466691970825195,
|
| 1219 |
-
"learning_rate": 0.00019016521984886983,
|
| 1220 |
-
"loss": 6.6329,
|
| 1221 |
-
"step": 172
|
| 1222 |
-
},
|
| 1223 |
-
{
|
| 1224 |
-
"epoch": 0.15076252723311548,
|
| 1225 |
-
"grad_norm": 2.9854562282562256,
|
| 1226 |
-
"learning_rate": 0.00019004548985122514,
|
| 1227 |
-
"loss": 5.7907,
|
| 1228 |
-
"step": 173
|
| 1229 |
-
},
|
| 1230 |
-
{
|
| 1231 |
-
"epoch": 0.15163398692810456,
|
| 1232 |
-
"grad_norm": 4.434035778045654,
|
| 1233 |
-
"learning_rate": 0.0001899250736121874,
|
| 1234 |
-
"loss": 6.4903,
|
| 1235 |
-
"step": 174
|
| 1236 |
-
},
|
| 1237 |
-
{
|
| 1238 |
-
"epoch": 0.15250544662309368,
|
| 1239 |
-
"grad_norm": 3.617687225341797,
|
| 1240 |
-
"learning_rate": 0.00018980397204945522,
|
| 1241 |
-
"loss": 6.353,
|
| 1242 |
-
"step": 175
|
| 1243 |
-
},
|
| 1244 |
-
{
|
| 1245 |
-
"epoch": 0.1533769063180828,
|
| 1246 |
-
"grad_norm": 3.2313356399536133,
|
| 1247 |
-
"learning_rate": 0.00018968218608594986,
|
| 1248 |
-
"loss": 6.3903,
|
| 1249 |
-
"step": 176
|
| 1250 |
-
},
|
| 1251 |
-
{
|
| 1252 |
-
"epoch": 0.1542483660130719,
|
| 1253 |
-
"grad_norm": 3.2535009384155273,
|
| 1254 |
-
"learning_rate": 0.00018955971664980854,
|
| 1255 |
-
"loss": 6.1581,
|
| 1256 |
-
"step": 177
|
| 1257 |
-
},
|
| 1258 |
-
{
|
| 1259 |
-
"epoch": 0.15511982570806102,
|
| 1260 |
-
"grad_norm": 3.6953959465026855,
|
| 1261 |
-
"learning_rate": 0.00018943656467437724,
|
| 1262 |
-
"loss": 5.9416,
|
| 1263 |
-
"step": 178
|
| 1264 |
-
},
|
| 1265 |
-
{
|
| 1266 |
-
"epoch": 0.1559912854030501,
|
| 1267 |
-
"grad_norm": 3.7896955013275146,
|
| 1268 |
-
"learning_rate": 0.00018931273109820364,
|
| 1269 |
-
"loss": 6.3745,
|
| 1270 |
-
"step": 179
|
| 1271 |
-
},
|
| 1272 |
-
{
|
| 1273 |
-
"epoch": 0.1568627450980392,
|
| 1274 |
-
"grad_norm": 3.284451961517334,
|
| 1275 |
-
"learning_rate": 0.00018918821686502989,
|
| 1276 |
-
"loss": 6.5936,
|
| 1277 |
-
"step": 180
|
| 1278 |
-
},
|
| 1279 |
-
{
|
| 1280 |
-
"epoch": 0.15773420479302833,
|
| 1281 |
-
"grad_norm": 3.3078837394714355,
|
| 1282 |
-
"learning_rate": 0.00018906302292378549,
|
| 1283 |
-
"loss": 6.3766,
|
| 1284 |
-
"step": 181
|
| 1285 |
-
},
|
| 1286 |
-
{
|
| 1287 |
-
"epoch": 0.15860566448801744,
|
| 1288 |
-
"grad_norm": 2.928623676300049,
|
| 1289 |
-
"learning_rate": 0.00018893715022857998,
|
| 1290 |
-
"loss": 6.1763,
|
| 1291 |
-
"step": 182
|
| 1292 |
-
},
|
| 1293 |
-
{
|
| 1294 |
-
"epoch": 0.15947712418300652,
|
| 1295 |
-
"grad_norm": 3.517934560775757,
|
| 1296 |
-
"learning_rate": 0.0001888105997386958,
|
| 1297 |
-
"loss": 7.0927,
|
| 1298 |
-
"step": 183
|
| 1299 |
-
},
|
| 1300 |
-
{
|
| 1301 |
-
"epoch": 0.16034858387799564,
|
| 1302 |
-
"grad_norm": 5.320308685302734,
|
| 1303 |
-
"learning_rate": 0.0001886833724185809,
|
| 1304 |
-
"loss": 7.4936,
|
| 1305 |
-
"step": 184
|
| 1306 |
-
},
|
| 1307 |
-
{
|
| 1308 |
-
"epoch": 0.16122004357298475,
|
| 1309 |
-
"grad_norm": 3.246738910675049,
|
| 1310 |
-
"learning_rate": 0.00018855546923784124,
|
| 1311 |
-
"loss": 5.9404,
|
| 1312 |
-
"step": 185
|
| 1313 |
-
},
|
| 1314 |
-
{
|
| 1315 |
-
"epoch": 0.16209150326797386,
|
| 1316 |
-
"grad_norm": 3.304151773452759,
|
| 1317 |
-
"learning_rate": 0.00018842689117123376,
|
| 1318 |
-
"loss": 5.8871,
|
| 1319 |
-
"step": 186
|
| 1320 |
-
},
|
| 1321 |
-
{
|
| 1322 |
-
"epoch": 0.16296296296296298,
|
| 1323 |
-
"grad_norm": 3.847766399383545,
|
| 1324 |
-
"learning_rate": 0.00018829763919865858,
|
| 1325 |
-
"loss": 6.5487,
|
| 1326 |
-
"step": 187
|
| 1327 |
-
},
|
| 1328 |
-
{
|
| 1329 |
-
"epoch": 0.16383442265795206,
|
| 1330 |
-
"grad_norm": 3.1779181957244873,
|
| 1331 |
-
"learning_rate": 0.00018816771430515177,
|
| 1332 |
-
"loss": 6.2279,
|
| 1333 |
-
"step": 188
|
| 1334 |
-
},
|
| 1335 |
-
{
|
| 1336 |
-
"epoch": 0.16470588235294117,
|
| 1337 |
-
"grad_norm": 3.6784069538116455,
|
| 1338 |
-
"learning_rate": 0.00018803711748087773,
|
| 1339 |
-
"loss": 7.3388,
|
| 1340 |
-
"step": 189
|
| 1341 |
-
},
|
| 1342 |
-
{
|
| 1343 |
-
"epoch": 0.1655773420479303,
|
| 1344 |
-
"grad_norm": 3.748657703399658,
|
| 1345 |
-
"learning_rate": 0.00018790584972112174,
|
| 1346 |
-
"loss": 6.1629,
|
| 1347 |
-
"step": 190
|
| 1348 |
-
},
|
| 1349 |
-
{
|
| 1350 |
-
"epoch": 0.1664488017429194,
|
| 1351 |
-
"grad_norm": 3.3160455226898193,
|
| 1352 |
-
"learning_rate": 0.0001877739120262822,
|
| 1353 |
-
"loss": 5.6017,
|
| 1354 |
-
"step": 191
|
| 1355 |
-
},
|
| 1356 |
-
{
|
| 1357 |
-
"epoch": 0.16732026143790849,
|
| 1358 |
-
"grad_norm": 3.8499162197113037,
|
| 1359 |
-
"learning_rate": 0.00018764130540186328,
|
| 1360 |
-
"loss": 6.6374,
|
| 1361 |
-
"step": 192
|
| 1362 |
-
},
|
| 1363 |
-
{
|
| 1364 |
-
"epoch": 0.1681917211328976,
|
| 1365 |
-
"grad_norm": 3.7523679733276367,
|
| 1366 |
-
"learning_rate": 0.0001875080308584669,
|
| 1367 |
-
"loss": 6.3529,
|
| 1368 |
-
"step": 193
|
| 1369 |
-
},
|
| 1370 |
-
{
|
| 1371 |
-
"epoch": 0.1690631808278867,
|
| 1372 |
-
"grad_norm": 3.095731496810913,
|
| 1373 |
-
"learning_rate": 0.00018737408941178543,
|
| 1374 |
-
"loss": 6.5641,
|
| 1375 |
-
"step": 194
|
| 1376 |
-
},
|
| 1377 |
-
{
|
| 1378 |
-
"epoch": 0.16993464052287582,
|
| 1379 |
-
"grad_norm": 3.698686122894287,
|
| 1380 |
-
"learning_rate": 0.0001872394820825936,
|
| 1381 |
-
"loss": 5.9233,
|
| 1382 |
-
"step": 195
|
| 1383 |
-
},
|
| 1384 |
-
{
|
| 1385 |
-
"epoch": 0.17080610021786494,
|
| 1386 |
-
"grad_norm": 3.3824868202209473,
|
| 1387 |
-
"learning_rate": 0.00018710420989674091,
|
| 1388 |
-
"loss": 6.262,
|
| 1389 |
-
"step": 196
|
| 1390 |
-
},
|
| 1391 |
-
{
|
| 1392 |
-
"epoch": 0.17167755991285402,
|
| 1393 |
-
"grad_norm": 3.9130444526672363,
|
| 1394 |
-
"learning_rate": 0.0001869682738851438,
|
| 1395 |
-
"loss": 6.7302,
|
| 1396 |
-
"step": 197
|
| 1397 |
-
},
|
| 1398 |
-
{
|
| 1399 |
-
"epoch": 0.17254901960784313,
|
| 1400 |
-
"grad_norm": 4.361310958862305,
|
| 1401 |
-
"learning_rate": 0.00018683167508377774,
|
| 1402 |
-
"loss": 7.2424,
|
| 1403 |
-
"step": 198
|
| 1404 |
-
},
|
| 1405 |
-
{
|
| 1406 |
-
"epoch": 0.17342047930283225,
|
| 1407 |
-
"grad_norm": 3.0834221839904785,
|
| 1408 |
-
"learning_rate": 0.0001866944145336693,
|
| 1409 |
-
"loss": 5.7327,
|
| 1410 |
-
"step": 199
|
| 1411 |
-
},
|
| 1412 |
-
{
|
| 1413 |
-
"epoch": 0.17429193899782136,
|
| 1414 |
-
"grad_norm": 3.402498722076416,
|
| 1415 |
-
"learning_rate": 0.00018655649328088835,
|
| 1416 |
-
"loss": 6.1162,
|
| 1417 |
-
"step": 200
|
| 1418 |
-
},
|
| 1419 |
-
{
|
| 1420 |
-
"epoch": 0.17516339869281045,
|
| 1421 |
-
"grad_norm": 4.900463581085205,
|
| 1422 |
-
"learning_rate": 0.00018641791237654,
|
| 1423 |
-
"loss": 7.5089,
|
| 1424 |
-
"step": 201
|
| 1425 |
-
},
|
| 1426 |
-
{
|
| 1427 |
-
"epoch": 0.17603485838779956,
|
| 1428 |
-
"grad_norm": 3.0102016925811768,
|
| 1429 |
-
"learning_rate": 0.0001862786728767565,
|
| 1430 |
-
"loss": 6.6401,
|
| 1431 |
-
"step": 202
|
| 1432 |
-
},
|
| 1433 |
-
{
|
| 1434 |
-
"epoch": 0.17690631808278867,
|
| 1435 |
-
"grad_norm": 3.4192609786987305,
|
| 1436 |
-
"learning_rate": 0.00018613877584268944,
|
| 1437 |
-
"loss": 5.8015,
|
| 1438 |
-
"step": 203
|
| 1439 |
-
},
|
| 1440 |
-
{
|
| 1441 |
-
"epoch": 0.17777777777777778,
|
| 1442 |
-
"grad_norm": 3.483651876449585,
|
| 1443 |
-
"learning_rate": 0.00018599822234050141,
|
| 1444 |
-
"loss": 5.9493,
|
| 1445 |
-
"step": 204
|
| 1446 |
-
},
|
| 1447 |
-
{
|
| 1448 |
-
"epoch": 0.1786492374727669,
|
| 1449 |
-
"grad_norm": 3.289872884750366,
|
| 1450 |
-
"learning_rate": 0.00018585701344135807,
|
| 1451 |
-
"loss": 6.7852,
|
| 1452 |
-
"step": 205
|
| 1453 |
-
},
|
| 1454 |
-
{
|
| 1455 |
-
"epoch": 0.17952069716775598,
|
| 1456 |
-
"grad_norm": 3.382138252258301,
|
| 1457 |
-
"learning_rate": 0.00018571515022141976,
|
| 1458 |
-
"loss": 6.5197,
|
| 1459 |
-
"step": 206
|
| 1460 |
-
},
|
| 1461 |
-
{
|
| 1462 |
-
"epoch": 0.1803921568627451,
|
| 1463 |
-
"grad_norm": 3.1128950119018555,
|
| 1464 |
-
"learning_rate": 0.0001855726337618335,
|
| 1465 |
-
"loss": 6.764,
|
| 1466 |
-
"step": 207
|
| 1467 |
-
},
|
| 1468 |
-
{
|
| 1469 |
-
"epoch": 0.1812636165577342,
|
| 1470 |
-
"grad_norm": 2.732131004333496,
|
| 1471 |
-
"learning_rate": 0.00018542946514872477,
|
| 1472 |
-
"loss": 5.8167,
|
| 1473 |
-
"step": 208
|
| 1474 |
-
},
|
| 1475 |
-
{
|
| 1476 |
-
"epoch": 0.18213507625272332,
|
| 1477 |
-
"grad_norm": 3.122868537902832,
|
| 1478 |
-
"learning_rate": 0.00018528564547318902,
|
| 1479 |
-
"loss": 6.0786,
|
| 1480 |
-
"step": 209
|
| 1481 |
-
},
|
| 1482 |
-
{
|
| 1483 |
-
"epoch": 0.1830065359477124,
|
| 1484 |
-
"grad_norm": 2.85406231880188,
|
| 1485 |
-
"learning_rate": 0.00018514117583128347,
|
| 1486 |
-
"loss": 5.4824,
|
| 1487 |
-
"step": 210
|
| 1488 |
-
},
|
| 1489 |
-
{
|
| 1490 |
-
"epoch": 0.18387799564270152,
|
| 1491 |
-
"grad_norm": 2.9424877166748047,
|
| 1492 |
-
"learning_rate": 0.00018499605732401895,
|
| 1493 |
-
"loss": 6.106,
|
| 1494 |
-
"step": 211
|
| 1495 |
-
},
|
| 1496 |
-
{
|
| 1497 |
-
"epoch": 0.18474945533769063,
|
| 1498 |
-
"grad_norm": 3.1483314037323,
|
| 1499 |
-
"learning_rate": 0.0001848502910573511,
|
| 1500 |
-
"loss": 6.5205,
|
| 1501 |
-
"step": 212
|
| 1502 |
-
},
|
| 1503 |
-
{
|
| 1504 |
-
"epoch": 0.18562091503267975,
|
| 1505 |
-
"grad_norm": 3.3453025817871094,
|
| 1506 |
-
"learning_rate": 0.00018470387814217232,
|
| 1507 |
-
"loss": 6.1123,
|
| 1508 |
-
"step": 213
|
| 1509 |
-
},
|
| 1510 |
-
{
|
| 1511 |
-
"epoch": 0.18649237472766886,
|
| 1512 |
-
"grad_norm": 2.945319414138794,
|
| 1513 |
-
"learning_rate": 0.00018455681969430307,
|
| 1514 |
-
"loss": 5.8776,
|
| 1515 |
-
"step": 214
|
| 1516 |
-
},
|
| 1517 |
-
{
|
| 1518 |
-
"epoch": 0.18736383442265794,
|
| 1519 |
-
"grad_norm": 3.3796839714050293,
|
| 1520 |
-
"learning_rate": 0.0001844091168344835,
|
| 1521 |
-
"loss": 6.8702,
|
| 1522 |
-
"step": 215
|
| 1523 |
-
},
|
| 1524 |
-
{
|
| 1525 |
-
"epoch": 0.18823529411764706,
|
| 1526 |
-
"grad_norm": 2.9702494144439697,
|
| 1527 |
-
"learning_rate": 0.00018426077068836488,
|
| 1528 |
-
"loss": 5.8691,
|
| 1529 |
-
"step": 216
|
| 1530 |
-
},
|
| 1531 |
-
{
|
| 1532 |
-
"epoch": 0.18910675381263617,
|
| 1533 |
-
"grad_norm": 2.9386281967163086,
|
| 1534 |
-
"learning_rate": 0.0001841117823865009,
|
| 1535 |
-
"loss": 6.0198,
|
| 1536 |
-
"step": 217
|
| 1537 |
-
},
|
| 1538 |
-
{
|
| 1539 |
-
"epoch": 0.18997821350762528,
|
| 1540 |
-
"grad_norm": 2.764477491378784,
|
| 1541 |
-
"learning_rate": 0.00018396215306433922,
|
| 1542 |
-
"loss": 5.4371,
|
| 1543 |
-
"step": 218
|
| 1544 |
-
},
|
| 1545 |
-
{
|
| 1546 |
-
"epoch": 0.19084967320261437,
|
| 1547 |
-
"grad_norm": 3.959301471710205,
|
| 1548 |
-
"learning_rate": 0.00018381188386221275,
|
| 1549 |
-
"loss": 6.1633,
|
| 1550 |
-
"step": 219
|
| 1551 |
-
},
|
| 1552 |
-
{
|
| 1553 |
-
"epoch": 0.19172113289760348,
|
| 1554 |
-
"grad_norm": 3.0180444717407227,
|
| 1555 |
-
"learning_rate": 0.00018366097592533093,
|
| 1556 |
-
"loss": 5.5212,
|
| 1557 |
-
"step": 220
|
| 1558 |
-
},
|
| 1559 |
-
{
|
| 1560 |
-
"epoch": 0.1925925925925926,
|
| 1561 |
-
"grad_norm": 3.3983206748962402,
|
| 1562 |
-
"learning_rate": 0.00018350943040377105,
|
| 1563 |
-
"loss": 6.1409,
|
| 1564 |
-
"step": 221
|
| 1565 |
-
},
|
| 1566 |
-
{
|
| 1567 |
-
"epoch": 0.1934640522875817,
|
| 1568 |
-
"grad_norm": 3.186225652694702,
|
| 1569 |
-
"learning_rate": 0.0001833572484524695,
|
| 1570 |
-
"loss": 6.0513,
|
| 1571 |
-
"step": 222
|
| 1572 |
-
},
|
| 1573 |
-
{
|
| 1574 |
-
"epoch": 0.19433551198257082,
|
| 1575 |
-
"grad_norm": 3.2218050956726074,
|
| 1576 |
-
"learning_rate": 0.00018320443123121284,
|
| 1577 |
-
"loss": 6.6948,
|
| 1578 |
-
"step": 223
|
| 1579 |
-
},
|
| 1580 |
-
{
|
| 1581 |
-
"epoch": 0.1952069716775599,
|
| 1582 |
-
"grad_norm": 2.7722697257995605,
|
| 1583 |
-
"learning_rate": 0.0001830509799046292,
|
| 1584 |
-
"loss": 5.7941,
|
| 1585 |
-
"step": 224
|
| 1586 |
-
},
|
| 1587 |
-
{
|
| 1588 |
-
"epoch": 0.19607843137254902,
|
| 1589 |
-
"grad_norm": 3.123589515686035,
|
| 1590 |
-
"learning_rate": 0.0001828968956421791,
|
| 1591 |
-
"loss": 6.1152,
|
| 1592 |
-
"step": 225
|
| 1593 |
-
},
|
| 1594 |
-
{
|
| 1595 |
-
"epoch": 0.19694989106753813,
|
| 1596 |
-
"grad_norm": 2.928809404373169,
|
| 1597 |
-
"learning_rate": 0.0001827421796181468,
|
| 1598 |
-
"loss": 5.9507,
|
| 1599 |
-
"step": 226
|
| 1600 |
-
},
|
| 1601 |
-
{
|
| 1602 |
-
"epoch": 0.19782135076252724,
|
| 1603 |
-
"grad_norm": 3.9273805618286133,
|
| 1604 |
-
"learning_rate": 0.0001825868330116312,
|
| 1605 |
-
"loss": 5.8818,
|
| 1606 |
-
"step": 227
|
| 1607 |
-
},
|
| 1608 |
-
{
|
| 1609 |
-
"epoch": 0.19869281045751633,
|
| 1610 |
-
"grad_norm": 3.857663631439209,
|
| 1611 |
-
"learning_rate": 0.00018243085700653697,
|
| 1612 |
-
"loss": 5.8865,
|
| 1613 |
-
"step": 228
|
| 1614 |
-
},
|
| 1615 |
-
{
|
| 1616 |
-
"epoch": 0.19956427015250544,
|
| 1617 |
-
"grad_norm": 3.6917290687561035,
|
| 1618 |
-
"learning_rate": 0.00018227425279156536,
|
| 1619 |
-
"loss": 6.6875,
|
| 1620 |
-
"step": 229
|
| 1621 |
-
},
|
| 1622 |
-
{
|
| 1623 |
-
"epoch": 0.20043572984749455,
|
| 1624 |
-
"grad_norm": 3.376311779022217,
|
| 1625 |
-
"learning_rate": 0.0001821170215602053,
|
| 1626 |
-
"loss": 6.2749,
|
| 1627 |
-
"step": 230
|
| 1628 |
-
},
|
| 1629 |
-
{
|
| 1630 |
-
"epoch": 0.20130718954248367,
|
| 1631 |
-
"grad_norm": 3.929055690765381,
|
| 1632 |
-
"learning_rate": 0.00018195916451072428,
|
| 1633 |
-
"loss": 6.7266,
|
| 1634 |
-
"step": 231
|
| 1635 |
-
},
|
| 1636 |
-
{
|
| 1637 |
-
"epoch": 0.20217864923747278,
|
| 1638 |
-
"grad_norm": 3.159097671508789,
|
| 1639 |
-
"learning_rate": 0.0001818006828461591,
|
| 1640 |
-
"loss": 5.859,
|
| 1641 |
-
"step": 232
|
| 1642 |
-
},
|
| 1643 |
-
{
|
| 1644 |
-
"epoch": 0.20305010893246186,
|
| 1645 |
-
"grad_norm": 3.784909248352051,
|
| 1646 |
-
"learning_rate": 0.00018164157777430682,
|
| 1647 |
-
"loss": 6.3503,
|
| 1648 |
-
"step": 233
|
| 1649 |
-
},
|
| 1650 |
-
{
|
| 1651 |
-
"epoch": 0.20392156862745098,
|
| 1652 |
-
"grad_norm": 4.014029502868652,
|
| 1653 |
-
"learning_rate": 0.00018148185050771554,
|
| 1654 |
-
"loss": 6.4046,
|
| 1655 |
-
"step": 234
|
| 1656 |
-
},
|
| 1657 |
-
{
|
| 1658 |
-
"epoch": 0.2047930283224401,
|
| 1659 |
-
"grad_norm": 2.901723861694336,
|
| 1660 |
-
"learning_rate": 0.00018132150226367512,
|
| 1661 |
-
"loss": 6.1373,
|
| 1662 |
-
"step": 235
|
| 1663 |
-
},
|
| 1664 |
-
{
|
| 1665 |
-
"epoch": 0.2056644880174292,
|
| 1666 |
-
"grad_norm": 3.193726062774658,
|
| 1667 |
-
"learning_rate": 0.00018116053426420794,
|
| 1668 |
-
"loss": 6.2137,
|
| 1669 |
-
"step": 236
|
| 1670 |
-
},
|
| 1671 |
-
{
|
| 1672 |
-
"epoch": 0.2065359477124183,
|
| 1673 |
-
"grad_norm": 3.4831013679504395,
|
| 1674 |
-
"learning_rate": 0.0001809989477360595,
|
| 1675 |
-
"loss": 6.9075,
|
| 1676 |
-
"step": 237
|
| 1677 |
-
},
|
| 1678 |
-
{
|
| 1679 |
-
"epoch": 0.2074074074074074,
|
| 1680 |
-
"grad_norm": 2.7948129177093506,
|
| 1681 |
-
"learning_rate": 0.00018083674391068924,
|
| 1682 |
-
"loss": 5.8719,
|
| 1683 |
-
"step": 238
|
| 1684 |
-
},
|
| 1685 |
-
{
|
| 1686 |
-
"epoch": 0.20827886710239651,
|
| 1687 |
-
"grad_norm": 3.455291986465454,
|
| 1688 |
-
"learning_rate": 0.00018067392402426103,
|
| 1689 |
-
"loss": 6.0601,
|
| 1690 |
-
"step": 239
|
| 1691 |
-
},
|
| 1692 |
-
{
|
| 1693 |
-
"epoch": 0.20915032679738563,
|
| 1694 |
-
"grad_norm": 3.306884288787842,
|
| 1695 |
-
"learning_rate": 0.00018051048931763366,
|
| 1696 |
-
"loss": 6.5009,
|
| 1697 |
-
"step": 240
|
| 1698 |
-
},
|
| 1699 |
-
{
|
| 1700 |
-
"epoch": 0.21002178649237474,
|
| 1701 |
-
"grad_norm": 3.174142599105835,
|
| 1702 |
-
"learning_rate": 0.00018034644103635166,
|
| 1703 |
-
"loss": 5.1709,
|
| 1704 |
-
"step": 241
|
| 1705 |
-
},
|
| 1706 |
-
{
|
| 1707 |
-
"epoch": 0.21089324618736383,
|
| 1708 |
-
"grad_norm": 3.9862282276153564,
|
| 1709 |
-
"learning_rate": 0.00018018178043063555,
|
| 1710 |
-
"loss": 6.5155,
|
| 1711 |
-
"step": 242
|
| 1712 |
-
},
|
| 1713 |
-
{
|
| 1714 |
-
"epoch": 0.21176470588235294,
|
| 1715 |
-
"grad_norm": 3.0755395889282227,
|
| 1716 |
-
"learning_rate": 0.0001800165087553724,
|
| 1717 |
-
"loss": 6.4159,
|
| 1718 |
-
"step": 243
|
| 1719 |
-
},
|
| 1720 |
-
{
|
| 1721 |
-
"epoch": 0.21263616557734205,
|
| 1722 |
-
"grad_norm": 3.4993367195129395,
|
| 1723 |
-
"learning_rate": 0.0001798506272701064,
|
| 1724 |
-
"loss": 5.8788,
|
| 1725 |
-
"step": 244
|
| 1726 |
-
},
|
| 1727 |
-
{
|
| 1728 |
-
"epoch": 0.21350762527233116,
|
| 1729 |
-
"grad_norm": 3.3749759197235107,
|
| 1730 |
-
"learning_rate": 0.00017968413723902887,
|
| 1731 |
-
"loss": 6.5896,
|
| 1732 |
-
"step": 245
|
| 1733 |
-
},
|
| 1734 |
-
{
|
| 1735 |
-
"epoch": 0.21437908496732025,
|
| 1736 |
-
"grad_norm": 3.2183454036712646,
|
| 1737 |
-
"learning_rate": 0.0001795170399309692,
|
| 1738 |
-
"loss": 5.8676,
|
| 1739 |
-
"step": 246
|
| 1740 |
-
},
|
| 1741 |
-
{
|
| 1742 |
-
"epoch": 0.21525054466230936,
|
| 1743 |
-
"grad_norm": 3.839151620864868,
|
| 1744 |
-
"learning_rate": 0.00017934933661938466,
|
| 1745 |
-
"loss": 5.7902,
|
| 1746 |
-
"step": 247
|
| 1747 |
-
},
|
| 1748 |
-
{
|
| 1749 |
-
"epoch": 0.21612200435729848,
|
| 1750 |
-
"grad_norm": 5.124701976776123,
|
| 1751 |
-
"learning_rate": 0.00017918102858235102,
|
| 1752 |
-
"loss": 6.4626,
|
| 1753 |
-
"step": 248
|
| 1754 |
-
},
|
| 1755 |
-
{
|
| 1756 |
-
"epoch": 0.2169934640522876,
|
| 1757 |
-
"grad_norm": 3.1410319805145264,
|
| 1758 |
-
"learning_rate": 0.00017901211710255255,
|
| 1759 |
-
"loss": 5.8767,
|
| 1760 |
-
"step": 249
|
| 1761 |
-
},
|
| 1762 |
-
{
|
| 1763 |
-
"epoch": 0.2178649237472767,
|
| 1764 |
-
"grad_norm": 3.26542592048645,
|
| 1765 |
-
"learning_rate": 0.00017884260346727254,
|
| 1766 |
-
"loss": 5.8412,
|
| 1767 |
-
"step": 250
|
| 1768 |
-
},
|
| 1769 |
-
{
|
| 1770 |
-
"epoch": 0.21873638344226579,
|
| 1771 |
-
"grad_norm": 3.5896737575531006,
|
| 1772 |
-
"learning_rate": 0.0001786724889683833,
|
| 1773 |
-
"loss": 6.1054,
|
| 1774 |
-
"step": 251
|
| 1775 |
-
},
|
| 1776 |
-
{
|
| 1777 |
-
"epoch": 0.2196078431372549,
|
| 1778 |
-
"grad_norm": 3.036224842071533,
|
| 1779 |
-
"learning_rate": 0.00017850177490233633,
|
| 1780 |
-
"loss": 6.306,
|
| 1781 |
-
"step": 252
|
| 1782 |
-
},
|
| 1783 |
-
{
|
| 1784 |
-
"epoch": 0.220479302832244,
|
| 1785 |
-
"grad_norm": 2.9149057865142822,
|
| 1786 |
-
"learning_rate": 0.0001783304625701524,
|
| 1787 |
-
"loss": 5.7077,
|
| 1788 |
-
"step": 253
|
| 1789 |
-
},
|
| 1790 |
-
{
|
| 1791 |
-
"epoch": 0.22135076252723312,
|
| 1792 |
-
"grad_norm": 3.4123952388763428,
|
| 1793 |
-
"learning_rate": 0.00017815855327741185,
|
| 1794 |
-
"loss": 5.6122,
|
| 1795 |
-
"step": 254
|
| 1796 |
-
},
|
| 1797 |
-
{
|
| 1798 |
-
"epoch": 0.2222222222222222,
|
| 1799 |
-
"grad_norm": 3.3414759635925293,
|
| 1800 |
-
"learning_rate": 0.00017798604833424434,
|
| 1801 |
-
"loss": 6.2309,
|
| 1802 |
-
"step": 255
|
| 1803 |
-
},
|
| 1804 |
-
{
|
| 1805 |
-
"epoch": 0.22309368191721132,
|
| 1806 |
-
"grad_norm": 2.8341996669769287,
|
| 1807 |
-
"learning_rate": 0.00017781294905531907,
|
| 1808 |
-
"loss": 5.7753,
|
| 1809 |
-
"step": 256
|
| 1810 |
-
},
|
| 1811 |
-
{
|
| 1812 |
-
"epoch": 0.22396514161220044,
|
| 1813 |
-
"grad_norm": 2.8891713619232178,
|
| 1814 |
-
"learning_rate": 0.00017763925675983475,
|
| 1815 |
-
"loss": 6.259,
|
| 1816 |
-
"step": 257
|
| 1817 |
-
},
|
| 1818 |
-
{
|
| 1819 |
-
"epoch": 0.22483660130718955,
|
| 1820 |
-
"grad_norm": 3.7543976306915283,
|
| 1821 |
-
"learning_rate": 0.00017746497277150943,
|
| 1822 |
-
"loss": 6.2447,
|
| 1823 |
-
"step": 258
|
| 1824 |
-
},
|
| 1825 |
-
{
|
| 1826 |
-
"epoch": 0.22570806100217866,
|
| 1827 |
-
"grad_norm": 3.6341092586517334,
|
| 1828 |
-
"learning_rate": 0.00017729009841857053,
|
| 1829 |
-
"loss": 6.503,
|
| 1830 |
-
"step": 259
|
| 1831 |
-
},
|
| 1832 |
-
{
|
| 1833 |
-
"epoch": 0.22657952069716775,
|
| 1834 |
-
"grad_norm": 3.1001079082489014,
|
| 1835 |
-
"learning_rate": 0.00017711463503374466,
|
| 1836 |
-
"loss": 5.544,
|
| 1837 |
-
"step": 260
|
| 1838 |
-
},
|
| 1839 |
-
{
|
| 1840 |
-
"epoch": 0.22745098039215686,
|
| 1841 |
-
"grad_norm": 3.5646779537200928,
|
| 1842 |
-
"learning_rate": 0.00017693858395424742,
|
| 1843 |
-
"loss": 6.8872,
|
| 1844 |
-
"step": 261
|
| 1845 |
-
},
|
| 1846 |
-
{
|
| 1847 |
-
"epoch": 0.22832244008714597,
|
| 1848 |
-
"grad_norm": 3.1725387573242188,
|
| 1849 |
-
"learning_rate": 0.00017676194652177334,
|
| 1850 |
-
"loss": 5.8004,
|
| 1851 |
-
"step": 262
|
| 1852 |
-
},
|
| 1853 |
-
{
|
| 1854 |
-
"epoch": 0.22919389978213509,
|
| 1855 |
-
"grad_norm": 3.6738593578338623,
|
| 1856 |
-
"learning_rate": 0.00017658472408248551,
|
| 1857 |
-
"loss": 6.0577,
|
| 1858 |
-
"step": 263
|
| 1859 |
-
},
|
| 1860 |
-
{
|
| 1861 |
-
"epoch": 0.23006535947712417,
|
| 1862 |
-
"grad_norm": 3.147831439971924,
|
| 1863 |
-
"learning_rate": 0.0001764069179870055,
|
| 1864 |
-
"loss": 5.8112,
|
| 1865 |
-
"step": 264
|
| 1866 |
-
},
|
| 1867 |
-
{
|
| 1868 |
-
"epoch": 0.23093681917211328,
|
| 1869 |
-
"grad_norm": 3.1304612159729004,
|
| 1870 |
-
"learning_rate": 0.00017622852959040273,
|
| 1871 |
-
"loss": 6.0289,
|
| 1872 |
-
"step": 265
|
| 1873 |
-
},
|
| 1874 |
-
{
|
| 1875 |
-
"epoch": 0.2318082788671024,
|
| 1876 |
-
"grad_norm": 3.5180559158325195,
|
| 1877 |
-
"learning_rate": 0.0001760495602521847,
|
| 1878 |
-
"loss": 6.1329,
|
| 1879 |
-
"step": 266
|
| 1880 |
-
},
|
| 1881 |
-
{
|
| 1882 |
-
"epoch": 0.2326797385620915,
|
| 1883 |
-
"grad_norm": 2.9736528396606445,
|
| 1884 |
-
"learning_rate": 0.00017587001133628592,
|
| 1885 |
-
"loss": 5.8546,
|
| 1886 |
-
"step": 267
|
| 1887 |
-
},
|
| 1888 |
-
{
|
| 1889 |
-
"epoch": 0.23355119825708062,
|
| 1890 |
-
"grad_norm": 2.9651129245758057,
|
| 1891 |
-
"learning_rate": 0.0001756898842110582,
|
| 1892 |
-
"loss": 6.4554,
|
| 1893 |
-
"step": 268
|
| 1894 |
-
},
|
| 1895 |
-
{
|
| 1896 |
-
"epoch": 0.2344226579520697,
|
| 1897 |
-
"grad_norm": 4.1373515129089355,
|
| 1898 |
-
"learning_rate": 0.0001755091802492597,
|
| 1899 |
-
"loss": 6.0014,
|
| 1900 |
-
"step": 269
|
| 1901 |
-
},
|
| 1902 |
-
{
|
| 1903 |
-
"epoch": 0.23529411764705882,
|
| 1904 |
-
"grad_norm": 3.086876630783081,
|
| 1905 |
-
"learning_rate": 0.0001753279008280449,
|
| 1906 |
-
"loss": 6.4445,
|
| 1907 |
-
"step": 270
|
| 1908 |
-
},
|
| 1909 |
-
{
|
| 1910 |
-
"epoch": 0.23616557734204793,
|
| 1911 |
-
"grad_norm": 3.251668930053711,
|
| 1912 |
-
"learning_rate": 0.00017514604732895363,
|
| 1913 |
-
"loss": 6.9109,
|
| 1914 |
-
"step": 271
|
| 1915 |
-
},
|
| 1916 |
-
{
|
| 1917 |
-
"epoch": 0.23703703703703705,
|
| 1918 |
-
"grad_norm": 3.4233386516571045,
|
| 1919 |
-
"learning_rate": 0.000174963621137901,
|
| 1920 |
-
"loss": 6.6209,
|
| 1921 |
-
"step": 272
|
| 1922 |
-
},
|
| 1923 |
-
{
|
| 1924 |
-
"epoch": 0.23790849673202613,
|
| 1925 |
-
"grad_norm": 3.802621603012085,
|
| 1926 |
-
"learning_rate": 0.0001747806236451666,
|
| 1927 |
-
"loss": 7.2266,
|
| 1928 |
-
"step": 273
|
| 1929 |
-
},
|
| 1930 |
-
{
|
| 1931 |
-
"epoch": 0.23877995642701524,
|
| 1932 |
-
"grad_norm": 3.2463502883911133,
|
| 1933 |
-
"learning_rate": 0.00017459705624538383,
|
| 1934 |
-
"loss": 6.7707,
|
| 1935 |
-
"step": 274
|
| 1936 |
-
},
|
| 1937 |
-
{
|
| 1938 |
-
"epoch": 0.23965141612200436,
|
| 1939 |
-
"grad_norm": 2.789658784866333,
|
| 1940 |
-
"learning_rate": 0.00017441292033752948,
|
| 1941 |
-
"loss": 6.0205,
|
| 1942 |
-
"step": 275
|
| 1943 |
-
},
|
| 1944 |
-
{
|
| 1945 |
-
"epoch": 0.24052287581699347,
|
| 1946 |
-
"grad_norm": 3.016808271408081,
|
| 1947 |
-
"learning_rate": 0.00017422821732491293,
|
| 1948 |
-
"loss": 5.7279,
|
| 1949 |
-
"step": 276
|
| 1950 |
-
},
|
| 1951 |
-
{
|
| 1952 |
-
"epoch": 0.24139433551198258,
|
| 1953 |
-
"grad_norm": 3.44633150100708,
|
| 1954 |
-
"learning_rate": 0.00017404294861516557,
|
| 1955 |
-
"loss": 6.5382,
|
| 1956 |
-
"step": 277
|
| 1957 |
-
},
|
| 1958 |
-
{
|
| 1959 |
-
"epoch": 0.24226579520697167,
|
| 1960 |
-
"grad_norm": 3.792668342590332,
|
| 1961 |
-
"learning_rate": 0.00017385711562022986,
|
| 1962 |
-
"loss": 7.1705,
|
| 1963 |
-
"step": 278
|
| 1964 |
-
},
|
| 1965 |
-
{
|
| 1966 |
-
"epoch": 0.24313725490196078,
|
| 1967 |
-
"grad_norm": 3.15065860748291,
|
| 1968 |
-
"learning_rate": 0.00017367071975634882,
|
| 1969 |
-
"loss": 5.6286,
|
| 1970 |
-
"step": 279
|
| 1971 |
-
},
|
| 1972 |
-
{
|
| 1973 |
-
"epoch": 0.2440087145969499,
|
| 1974 |
-
"grad_norm": 3.124851703643799,
|
| 1975 |
-
"learning_rate": 0.00017348376244405512,
|
| 1976 |
-
"loss": 5.9117,
|
| 1977 |
-
"step": 280
|
| 1978 |
-
},
|
| 1979 |
-
{
|
| 1980 |
-
"epoch": 0.244880174291939,
|
| 1981 |
-
"grad_norm": 3.9259023666381836,
|
| 1982 |
-
"learning_rate": 0.00017329624510816017,
|
| 1983 |
-
"loss": 6.4439,
|
| 1984 |
-
"step": 281
|
| 1985 |
-
},
|
| 1986 |
-
{
|
| 1987 |
-
"epoch": 0.2457516339869281,
|
| 1988 |
-
"grad_norm": 3.3216142654418945,
|
| 1989 |
-
"learning_rate": 0.00017310816917774343,
|
| 1990 |
-
"loss": 6.8465,
|
| 1991 |
-
"step": 282
|
| 1992 |
-
},
|
| 1993 |
-
{
|
| 1994 |
-
"epoch": 0.2466230936819172,
|
| 1995 |
-
"grad_norm": 4.405170917510986,
|
| 1996 |
-
"learning_rate": 0.0001729195360861414,
|
| 1997 |
-
"loss": 6.1221,
|
| 1998 |
-
"step": 283
|
| 1999 |
-
},
|
| 2000 |
-
{
|
| 2001 |
-
"epoch": 0.24749455337690632,
|
| 2002 |
-
"grad_norm": 3.6341638565063477,
|
| 2003 |
-
"learning_rate": 0.0001727303472709368,
|
| 2004 |
-
"loss": 6.5787,
|
| 2005 |
-
"step": 284
|
| 2006 |
-
},
|
| 2007 |
-
{
|
| 2008 |
-
"epoch": 0.24836601307189543,
|
| 2009 |
-
"grad_norm": 3.294029712677002,
|
| 2010 |
-
"learning_rate": 0.00017254060417394741,
|
| 2011 |
-
"loss": 6.3653,
|
| 2012 |
-
"step": 285
|
| 2013 |
-
},
|
| 2014 |
-
{
|
| 2015 |
-
"epoch": 0.24923747276688454,
|
| 2016 |
-
"grad_norm": 3.6027772426605225,
|
| 2017 |
-
"learning_rate": 0.00017235030824121542,
|
| 2018 |
-
"loss": 5.6906,
|
| 2019 |
-
"step": 286
|
| 2020 |
-
},
|
| 2021 |
-
{
|
| 2022 |
-
"epoch": 0.25010893246187366,
|
| 2023 |
-
"grad_norm": 3.3305227756500244,
|
| 2024 |
-
"learning_rate": 0.000172159460922996,
|
| 2025 |
-
"loss": 6.9769,
|
| 2026 |
-
"step": 287
|
| 2027 |
-
},
|
| 2028 |
-
{
|
| 2029 |
-
"epoch": 0.25010893246187366,
|
| 2030 |
-
"eval_loss": 1.5338541269302368,
|
| 2031 |
-
"eval_runtime": 35.2309,
|
| 2032 |
-
"eval_samples_per_second": 13.738,
|
| 2033 |
-
"eval_steps_per_second": 6.869,
|
| 2034 |
-
"step": 287
|
| 2035 |
}
|
| 2036 |
],
|
| 2037 |
"logging_steps": 1,
|
| 2038 |
-
"max_steps":
|
| 2039 |
"num_input_tokens_seen": 0,
|
| 2040 |
"num_train_epochs": 2,
|
| 2041 |
-
"save_steps":
|
| 2042 |
"stateful_callbacks": {
|
| 2043 |
"TrainerControl": {
|
| 2044 |
"args": {
|
|
@@ -2051,8 +1034,8 @@
|
|
| 2051 |
"attributes": {}
|
| 2052 |
}
|
| 2053 |
},
|
| 2054 |
-
"total_flos":
|
| 2055 |
-
"train_batch_size":
|
| 2056 |
"trial_name": null,
|
| 2057 |
"trial_params": null
|
| 2058 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.25098039215686274,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 144,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.0017429193899782135,
|
| 13 |
+
"grad_norm": 0.6158298850059509,
|
| 14 |
"learning_rate": 2e-05,
|
| 15 |
+
"loss": 13.2165,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.003485838779956427,
|
| 20 |
+
"grad_norm": 0.6245508790016174,
|
| 21 |
"learning_rate": 4e-05,
|
| 22 |
+
"loss": 12.78,
|
| 23 |
"step": 2
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.00522875816993464,
|
| 27 |
+
"grad_norm": 0.6068661212921143,
|
| 28 |
"learning_rate": 6e-05,
|
| 29 |
+
"loss": 12.9627,
|
| 30 |
"step": 3
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.006971677559912854,
|
| 34 |
+
"grad_norm": 0.7556893229484558,
|
| 35 |
"learning_rate": 8e-05,
|
| 36 |
+
"loss": 12.5849,
|
| 37 |
"step": 4
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.008714596949891068,
|
| 41 |
+
"grad_norm": 0.9619686603546143,
|
| 42 |
"learning_rate": 0.0001,
|
| 43 |
+
"loss": 12.759,
|
| 44 |
"step": 5
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.01045751633986928,
|
| 48 |
+
"grad_norm": 1.2424684762954712,
|
| 49 |
+
"learning_rate": 9.999923789621597e-05,
|
| 50 |
+
"loss": 12.9,
|
| 51 |
"step": 6
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.012200435729847494,
|
| 55 |
+
"grad_norm": 1.634145736694336,
|
| 56 |
+
"learning_rate": 9.999695160809596e-05,
|
| 57 |
+
"loss": 13.1499,
|
| 58 |
"step": 7
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.013943355119825708,
|
| 62 |
+
"grad_norm": 1.9859766960144043,
|
| 63 |
+
"learning_rate": 9.999314120533557e-05,
|
| 64 |
+
"loss": 12.7814,
|
| 65 |
"step": 8
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.01568627450980392,
|
| 69 |
+
"grad_norm": 2.207068920135498,
|
| 70 |
+
"learning_rate": 9.998780680409161e-05,
|
| 71 |
+
"loss": 11.7586,
|
| 72 |
"step": 9
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.017429193899782137,
|
| 76 |
+
"grad_norm": 2.345550298690796,
|
| 77 |
+
"learning_rate": 9.998094856697883e-05,
|
| 78 |
+
"loss": 11.8078,
|
| 79 |
"step": 10
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.01917211328976035,
|
| 83 |
+
"grad_norm": 3.4373486042022705,
|
| 84 |
+
"learning_rate": 9.997256670306479e-05,
|
| 85 |
+
"loss": 11.3877,
|
| 86 |
"step": 11
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.02091503267973856,
|
| 90 |
+
"grad_norm": 3.628711223602295,
|
| 91 |
+
"learning_rate": 9.996266146786344e-05,
|
| 92 |
+
"loss": 10.8927,
|
| 93 |
"step": 12
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.022657952069716776,
|
| 97 |
+
"grad_norm": 3.0676188468933105,
|
| 98 |
+
"learning_rate": 9.99512331633275e-05,
|
| 99 |
+
"loss": 10.8846,
|
| 100 |
"step": 13
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.024400871459694988,
|
| 104 |
+
"grad_norm": 2.7230966091156006,
|
| 105 |
+
"learning_rate": 9.993828213783915e-05,
|
| 106 |
+
"loss": 10.6591,
|
| 107 |
"step": 14
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.026143790849673203,
|
| 111 |
+
"grad_norm": 3.5522327423095703,
|
| 112 |
+
"learning_rate": 9.992380878619938e-05,
|
| 113 |
+
"loss": 11.0064,
|
| 114 |
"step": 15
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.027886710239651415,
|
| 118 |
+
"grad_norm": 3.1329822540283203,
|
| 119 |
+
"learning_rate": 9.990781354961605e-05,
|
| 120 |
+
"loss": 10.156,
|
| 121 |
"step": 16
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.02962962962962963,
|
| 125 |
+
"grad_norm": 2.490328788757324,
|
| 126 |
+
"learning_rate": 9.989029691569037e-05,
|
| 127 |
+
"loss": 9.7129,
|
| 128 |
"step": 17
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.03137254901960784,
|
| 132 |
+
"grad_norm": 2.357733726501465,
|
| 133 |
+
"learning_rate": 9.987125941840204e-05,
|
| 134 |
+
"loss": 10.3977,
|
| 135 |
"step": 18
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.033115468409586055,
|
| 139 |
+
"grad_norm": 2.3284122943878174,
|
| 140 |
+
"learning_rate": 9.985070163809305e-05,
|
| 141 |
+
"loss": 9.8478,
|
| 142 |
"step": 19
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.034858387799564274,
|
| 146 |
+
"grad_norm": 2.3385725021362305,
|
| 147 |
+
"learning_rate": 9.982862420144985e-05,
|
| 148 |
+
"loss": 9.6358,
|
| 149 |
"step": 20
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.036601307189542485,
|
| 153 |
+
"grad_norm": 1.8942724466323853,
|
| 154 |
+
"learning_rate": 9.980502778148437e-05,
|
| 155 |
+
"loss": 9.9016,
|
| 156 |
"step": 21
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.0383442265795207,
|
| 160 |
+
"grad_norm": 2.210202693939209,
|
| 161 |
+
"learning_rate": 9.977991309751346e-05,
|
| 162 |
+
"loss": 9.7391,
|
| 163 |
"step": 22
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.04008714596949891,
|
| 167 |
+
"grad_norm": 1.7446099519729614,
|
| 168 |
+
"learning_rate": 9.975328091513696e-05,
|
| 169 |
+
"loss": 10.0383,
|
| 170 |
"step": 23
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.04183006535947712,
|
| 174 |
+
"grad_norm": 1.7494003772735596,
|
| 175 |
+
"learning_rate": 9.972513204621429e-05,
|
| 176 |
+
"loss": 9.6563,
|
| 177 |
"step": 24
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.04357298474945534,
|
| 181 |
+
"grad_norm": 1.59136962890625,
|
| 182 |
+
"learning_rate": 9.96954673488399e-05,
|
| 183 |
+
"loss": 9.6376,
|
| 184 |
"step": 25
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.04531590413943355,
|
| 188 |
+
"grad_norm": 1.7087982892990112,
|
| 189 |
+
"learning_rate": 9.966428772731688e-05,
|
| 190 |
+
"loss": 9.1243,
|
| 191 |
"step": 26
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.047058823529411764,
|
| 195 |
+
"grad_norm": 1.673592209815979,
|
| 196 |
+
"learning_rate": 9.963159413212953e-05,
|
| 197 |
+
"loss": 9.4495,
|
| 198 |
"step": 27
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.048801742919389976,
|
| 202 |
+
"grad_norm": 1.5657862424850464,
|
| 203 |
+
"learning_rate": 9.959738755991435e-05,
|
| 204 |
+
"loss": 9.5537,
|
| 205 |
"step": 28
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.050544662309368195,
|
| 209 |
+
"grad_norm": 1.446624517440796,
|
| 210 |
+
"learning_rate": 9.95616690534297e-05,
|
| 211 |
+
"loss": 8.733,
|
| 212 |
"step": 29
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.05228758169934641,
|
| 216 |
+
"grad_norm": 1.5286850929260254,
|
| 217 |
+
"learning_rate": 9.95244397015239e-05,
|
| 218 |
+
"loss": 8.8059,
|
| 219 |
"step": 30
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.05403050108932462,
|
| 223 |
+
"grad_norm": 1.6234805583953857,
|
| 224 |
+
"learning_rate": 9.948570063910216e-05,
|
| 225 |
+
"loss": 8.7743,
|
| 226 |
"step": 31
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.05577342047930283,
|
| 230 |
+
"grad_norm": 1.5004093647003174,
|
| 231 |
+
"learning_rate": 9.944545304709191e-05,
|
| 232 |
+
"loss": 8.7718,
|
| 233 |
"step": 32
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.05751633986928104,
|
| 237 |
+
"grad_norm": 1.8447928428649902,
|
| 238 |
+
"learning_rate": 9.940369815240688e-05,
|
| 239 |
+
"loss": 8.7953,
|
| 240 |
"step": 33
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.05925925925925926,
|
| 244 |
+
"grad_norm": 1.665925145149231,
|
| 245 |
+
"learning_rate": 9.936043722790956e-05,
|
| 246 |
+
"loss": 8.4996,
|
| 247 |
"step": 34
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.06100217864923747,
|
| 251 |
+
"grad_norm": 2.4476370811462402,
|
| 252 |
+
"learning_rate": 9.931567159237251e-05,
|
| 253 |
+
"loss": 8.2559,
|
| 254 |
"step": 35
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.06274509803921569,
|
| 258 |
+
"grad_norm": 1.7158671617507935,
|
| 259 |
+
"learning_rate": 9.92694026104382e-05,
|
| 260 |
+
"loss": 8.1353,
|
| 261 |
"step": 36
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.0644880174291939,
|
| 265 |
+
"grad_norm": 2.119276523590088,
|
| 266 |
+
"learning_rate": 9.922163169257721e-05,
|
| 267 |
+
"loss": 8.6706,
|
| 268 |
"step": 37
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 0.06623093681917211,
|
| 272 |
+
"grad_norm": 2.184993267059326,
|
| 273 |
+
"learning_rate": 9.917236029504547e-05,
|
| 274 |
+
"loss": 8.4756,
|
| 275 |
"step": 38
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.06797385620915032,
|
| 279 |
+
"grad_norm": 2.250312089920044,
|
| 280 |
+
"learning_rate": 9.91215899198397e-05,
|
| 281 |
+
"loss": 7.8425,
|
| 282 |
"step": 39
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.06971677559912855,
|
| 286 |
+
"grad_norm": 2.339944362640381,
|
| 287 |
+
"learning_rate": 9.906932211465173e-05,
|
| 288 |
+
"loss": 8.1978,
|
| 289 |
"step": 40
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 0.07145969498910676,
|
| 293 |
+
"grad_norm": 2.22255802154541,
|
| 294 |
+
"learning_rate": 9.901555847282122e-05,
|
| 295 |
+
"loss": 7.4908,
|
| 296 |
"step": 41
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 0.07320261437908497,
|
| 300 |
+
"grad_norm": 2.190056800842285,
|
| 301 |
+
"learning_rate": 9.896030063328717e-05,
|
| 302 |
+
"loss": 8.4106,
|
| 303 |
"step": 42
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 0.07494553376906318,
|
| 307 |
+
"grad_norm": 2.3875746726989746,
|
| 308 |
+
"learning_rate": 9.890355028053793e-05,
|
| 309 |
+
"loss": 8.1042,
|
| 310 |
"step": 43
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 0.0766884531590414,
|
| 314 |
+
"grad_norm": 2.2626776695251465,
|
| 315 |
+
"learning_rate": 9.884530914455984e-05,
|
| 316 |
+
"loss": 7.38,
|
| 317 |
"step": 44
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.0784313725490196,
|
| 321 |
+
"grad_norm": 2.90751314163208,
|
| 322 |
+
"learning_rate": 9.87855790007845e-05,
|
| 323 |
+
"loss": 7.5586,
|
| 324 |
"step": 45
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 0.08017429193899782,
|
| 328 |
+
"grad_norm": 2.6105775833129883,
|
| 329 |
+
"learning_rate": 9.872436167003467e-05,
|
| 330 |
+
"loss": 7.8514,
|
| 331 |
"step": 46
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 0.08191721132897603,
|
| 335 |
+
"grad_norm": 2.965604305267334,
|
| 336 |
+
"learning_rate": 9.86616590184687e-05,
|
| 337 |
+
"loss": 7.1101,
|
| 338 |
"step": 47
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.08366013071895424,
|
| 342 |
+
"grad_norm": 2.8931803703308105,
|
| 343 |
+
"learning_rate": 9.859747295752372e-05,
|
| 344 |
+
"loss": 7.7739,
|
| 345 |
"step": 48
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 0.08540305010893247,
|
| 349 |
+
"grad_norm": 3.355397939682007,
|
| 350 |
+
"learning_rate": 9.853180544385736e-05,
|
| 351 |
+
"loss": 8.0234,
|
| 352 |
"step": 49
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 0.08714596949891068,
|
| 356 |
+
"grad_norm": 4.1606292724609375,
|
| 357 |
+
"learning_rate": 9.8464658479288e-05,
|
| 358 |
+
"loss": 8.4085,
|
| 359 |
"step": 50
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.08888888888888889,
|
| 363 |
+
"grad_norm": 4.537642002105713,
|
| 364 |
+
"learning_rate": 9.839603411073388e-05,
|
| 365 |
+
"loss": 7.8124,
|
| 366 |
"step": 51
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 0.0906318082788671,
|
| 370 |
+
"grad_norm": 3.6303694248199463,
|
| 371 |
+
"learning_rate": 9.832593443015067e-05,
|
| 372 |
+
"loss": 7.2467,
|
| 373 |
"step": 52
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 0.09237472766884532,
|
| 377 |
+
"grad_norm": 2.6730170249938965,
|
| 378 |
+
"learning_rate": 9.825436157446762e-05,
|
| 379 |
+
"loss": 8.0992,
|
| 380 |
"step": 53
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 0.09411764705882353,
|
| 384 |
+
"grad_norm": 1.9828307628631592,
|
| 385 |
+
"learning_rate": 9.818131772552249e-05,
|
| 386 |
+
"loss": 7.385,
|
| 387 |
"step": 54
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 0.09586056644880174,
|
| 391 |
+
"grad_norm": 1.7050368785858154,
|
| 392 |
+
"learning_rate": 9.810680510999504e-05,
|
| 393 |
+
"loss": 7.4899,
|
| 394 |
"step": 55
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 0.09760348583877995,
|
| 398 |
+
"grad_norm": 2.1304142475128174,
|
| 399 |
+
"learning_rate": 9.80308259993391e-05,
|
| 400 |
+
"loss": 8.3856,
|
| 401 |
"step": 56
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.09934640522875816,
|
| 405 |
+
"grad_norm": 2.171189069747925,
|
| 406 |
+
"learning_rate": 9.795338270971341e-05,
|
| 407 |
+
"loss": 7.5037,
|
| 408 |
"step": 57
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 0.10108932461873639,
|
| 412 |
+
"grad_norm": 1.9037686586380005,
|
| 413 |
+
"learning_rate": 9.78744776019109e-05,
|
| 414 |
+
"loss": 7.4089,
|
| 415 |
"step": 58
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 0.1028322440087146,
|
| 419 |
+
"grad_norm": 1.5761909484863281,
|
| 420 |
+
"learning_rate": 9.779411308128685e-05,
|
| 421 |
+
"loss": 7.507,
|
| 422 |
"step": 59
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 0.10457516339869281,
|
| 426 |
+
"grad_norm": 1.6125203371047974,
|
| 427 |
+
"learning_rate": 9.771229159768547e-05,
|
| 428 |
+
"loss": 6.9208,
|
| 429 |
"step": 60
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 0.10631808278867103,
|
| 433 |
+
"grad_norm": 1.8575252294540405,
|
| 434 |
+
"learning_rate": 9.762901564536522e-05,
|
| 435 |
+
"loss": 7.2879,
|
| 436 |
"step": 61
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 0.10806100217864924,
|
| 440 |
+
"grad_norm": 1.8381884098052979,
|
| 441 |
+
"learning_rate": 9.754428776292287e-05,
|
| 442 |
+
"loss": 7.546,
|
| 443 |
"step": 62
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.10980392156862745,
|
| 447 |
+
"grad_norm": 1.5687748193740845,
|
| 448 |
+
"learning_rate": 9.745811053321596e-05,
|
| 449 |
+
"loss": 7.1789,
|
| 450 |
"step": 63
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 0.11154684095860566,
|
| 454 |
+
"grad_norm": 1.734289288520813,
|
| 455 |
+
"learning_rate": 9.737048658328427e-05,
|
| 456 |
+
"loss": 7.1747,
|
| 457 |
"step": 64
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 0.11328976034858387,
|
| 461 |
+
"grad_norm": 1.8572449684143066,
|
| 462 |
+
"learning_rate": 9.728141858426952e-05,
|
| 463 |
+
"loss": 6.9627,
|
| 464 |
"step": 65
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 0.11503267973856209,
|
| 468 |
+
"grad_norm": 1.816596508026123,
|
| 469 |
+
"learning_rate": 9.719090925133408e-05,
|
| 470 |
+
"loss": 7.4948,
|
| 471 |
"step": 66
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 0.11677559912854031,
|
| 475 |
+
"grad_norm": 1.9843227863311768,
|
| 476 |
+
"learning_rate": 9.709896134357815e-05,
|
| 477 |
+
"loss": 7.282,
|
| 478 |
"step": 67
|
| 479 |
},
|
| 480 |
{
|
| 481 |
+
"epoch": 0.11851851851851852,
|
| 482 |
+
"grad_norm": 1.6963810920715332,
|
| 483 |
+
"learning_rate": 9.700557766395566e-05,
|
| 484 |
+
"loss": 6.757,
|
| 485 |
"step": 68
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.12026143790849673,
|
| 489 |
+
"grad_norm": 2.103762149810791,
|
| 490 |
+
"learning_rate": 9.691076105918885e-05,
|
| 491 |
+
"loss": 6.83,
|
| 492 |
"step": 69
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 0.12200435729847495,
|
| 496 |
+
"grad_norm": 1.6644117832183838,
|
| 497 |
+
"learning_rate": 9.681451441968144e-05,
|
| 498 |
+
"loss": 6.9097,
|
| 499 |
"step": 70
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 0.12374727668845316,
|
| 503 |
+
"grad_norm": 2.2518982887268066,
|
| 504 |
+
"learning_rate": 9.671684067943057e-05,
|
| 505 |
+
"loss": 6.5565,
|
| 506 |
"step": 71
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 0.12549019607843137,
|
| 510 |
+
"grad_norm": 1.9176573753356934,
|
| 511 |
+
"learning_rate": 9.66177428159373e-05,
|
| 512 |
+
"loss": 6.3409,
|
| 513 |
"step": 72
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 0.1272331154684096,
|
| 517 |
+
"grad_norm": 2.9997940063476562,
|
| 518 |
+
"learning_rate": 9.651722385011592e-05,
|
| 519 |
+
"loss": 7.3301,
|
| 520 |
"step": 73
|
| 521 |
},
|
| 522 |
{
|
| 523 |
+
"epoch": 0.1289760348583878,
|
| 524 |
+
"grad_norm": 1.927817940711975,
|
| 525 |
+
"learning_rate": 9.641528684620178e-05,
|
| 526 |
+
"loss": 6.7444,
|
| 527 |
"step": 74
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 0.13071895424836602,
|
| 531 |
+
"grad_norm": 2.941283941268921,
|
| 532 |
+
"learning_rate": 9.631193491165797e-05,
|
| 533 |
+
"loss": 6.7375,
|
| 534 |
"step": 75
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 0.13246187363834422,
|
| 538 |
+
"grad_norm": 2.0440499782562256,
|
| 539 |
+
"learning_rate": 9.620717119708047e-05,
|
| 540 |
+
"loss": 6.2593,
|
| 541 |
"step": 76
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 0.13420479302832244,
|
| 545 |
+
"grad_norm": 2.585775375366211,
|
| 546 |
+
"learning_rate": 9.610099889610223e-05,
|
| 547 |
+
"loss": 7.0414,
|
| 548 |
"step": 77
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 0.13594771241830064,
|
| 552 |
+
"grad_norm": 2.524921178817749,
|
| 553 |
+
"learning_rate": 9.599342124529575e-05,
|
| 554 |
+
"loss": 6.6689,
|
| 555 |
"step": 78
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 0.13769063180827887,
|
| 559 |
+
"grad_norm": 2.768930196762085,
|
| 560 |
+
"learning_rate": 9.588444152407441e-05,
|
| 561 |
+
"loss": 7.0204,
|
| 562 |
"step": 79
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 0.1394335511982571,
|
| 566 |
+
"grad_norm": 2.33314847946167,
|
| 567 |
+
"learning_rate": 9.577406305459251e-05,
|
| 568 |
+
"loss": 6.5685,
|
| 569 |
"step": 80
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 0.1411764705882353,
|
| 573 |
+
"grad_norm": 2.3114585876464844,
|
| 574 |
+
"learning_rate": 9.566228920164406e-05,
|
| 575 |
+
"loss": 6.4868,
|
| 576 |
"step": 81
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 0.14291938997821352,
|
| 580 |
+
"grad_norm": 3.182647705078125,
|
| 581 |
+
"learning_rate": 9.554912337256007e-05,
|
| 582 |
+
"loss": 6.8026,
|
| 583 |
"step": 82
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"epoch": 0.14466230936819172,
|
| 587 |
+
"grad_norm": 3.128938913345337,
|
| 588 |
+
"learning_rate": 9.543456901710483e-05,
|
| 589 |
+
"loss": 6.7505,
|
| 590 |
"step": 83
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 0.14640522875816994,
|
| 594 |
+
"grad_norm": 2.4824540615081787,
|
| 595 |
+
"learning_rate": 9.531862962737065e-05,
|
| 596 |
+
"loss": 6.9284,
|
| 597 |
"step": 84
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 0.14814814814814814,
|
| 601 |
+
"grad_norm": 3.8796262741088867,
|
| 602 |
+
"learning_rate": 9.520130873767141e-05,
|
| 603 |
+
"loss": 6.3305,
|
| 604 |
"step": 85
|
| 605 |
},
|
| 606 |
{
|
| 607 |
+
"epoch": 0.14989106753812637,
|
| 608 |
+
"grad_norm": 3.027245283126831,
|
| 609 |
+
"learning_rate": 9.508260992443491e-05,
|
| 610 |
+
"loss": 6.4731,
|
| 611 |
"step": 86
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 0.15163398692810456,
|
| 615 |
+
"grad_norm": 3.0274035930633545,
|
| 616 |
+
"learning_rate": 9.49625368060937e-05,
|
| 617 |
+
"loss": 6.5764,
|
| 618 |
"step": 87
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 0.1533769063180828,
|
| 622 |
+
"grad_norm": 3.6287505626678467,
|
| 623 |
+
"learning_rate": 9.484109304297493e-05,
|
| 624 |
+
"loss": 6.7021,
|
| 625 |
"step": 88
|
| 626 |
},
|
| 627 |
{
|
| 628 |
+
"epoch": 0.15511982570806102,
|
| 629 |
+
"grad_norm": 3.2184479236602783,
|
| 630 |
+
"learning_rate": 9.471828233718862e-05,
|
| 631 |
+
"loss": 6.9716,
|
| 632 |
"step": 89
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 0.1568627450980392,
|
| 636 |
+
"grad_norm": 3.7180967330932617,
|
| 637 |
+
"learning_rate": 9.459410843251494e-05,
|
| 638 |
+
"loss": 6.5295,
|
| 639 |
"step": 90
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 0.15860566448801744,
|
| 643 |
+
"grad_norm": 3.2844398021698,
|
| 644 |
+
"learning_rate": 9.446857511428999e-05,
|
| 645 |
+
"loss": 6.7034,
|
| 646 |
"step": 91
|
| 647 |
},
|
| 648 |
{
|
| 649 |
+
"epoch": 0.16034858387799564,
|
| 650 |
+
"grad_norm": 2.817380428314209,
|
| 651 |
+
"learning_rate": 9.434168620929045e-05,
|
| 652 |
+
"loss": 6.5006,
|
| 653 |
"step": 92
|
| 654 |
},
|
| 655 |
{
|
| 656 |
+
"epoch": 0.16209150326797386,
|
| 657 |
+
"grad_norm": 2.6760520935058594,
|
| 658 |
+
"learning_rate": 9.421344558561688e-05,
|
| 659 |
+
"loss": 6.1821,
|
| 660 |
"step": 93
|
| 661 |
},
|
| 662 |
{
|
| 663 |
+
"epoch": 0.16383442265795206,
|
| 664 |
+
"grad_norm": 3.8171274662017822,
|
| 665 |
+
"learning_rate": 9.408385715257589e-05,
|
| 666 |
+
"loss": 6.163,
|
| 667 |
"step": 94
|
| 668 |
},
|
| 669 |
{
|
| 670 |
+
"epoch": 0.1655773420479303,
|
| 671 |
+
"grad_norm": 3.1886677742004395,
|
| 672 |
+
"learning_rate": 9.395292486056087e-05,
|
| 673 |
+
"loss": 6.6912,
|
| 674 |
"step": 95
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 0.16732026143790849,
|
| 678 |
+
"grad_norm": 3.741964101791382,
|
| 679 |
+
"learning_rate": 9.382065270093164e-05,
|
| 680 |
+
"loss": 6.4306,
|
| 681 |
"step": 96
|
| 682 |
},
|
| 683 |
{
|
| 684 |
+
"epoch": 0.1690631808278867,
|
| 685 |
+
"grad_norm": 3.4298031330108643,
|
| 686 |
+
"learning_rate": 9.368704470589271e-05,
|
| 687 |
+
"loss": 6.1829,
|
| 688 |
"step": 97
|
| 689 |
},
|
| 690 |
{
|
| 691 |
+
"epoch": 0.17080610021786494,
|
| 692 |
+
"grad_norm": 4.670931339263916,
|
| 693 |
+
"learning_rate": 9.355210494837046e-05,
|
| 694 |
+
"loss": 6.248,
|
| 695 |
"step": 98
|
| 696 |
},
|
| 697 |
{
|
| 698 |
+
"epoch": 0.17254901960784313,
|
| 699 |
+
"grad_norm": 5.248985767364502,
|
| 700 |
+
"learning_rate": 9.341583754188887e-05,
|
| 701 |
+
"loss": 6.8693,
|
| 702 |
"step": 99
|
| 703 |
},
|
| 704 |
{
|
| 705 |
+
"epoch": 0.17429193899782136,
|
| 706 |
+
"grad_norm": 6.1399359703063965,
|
| 707 |
+
"learning_rate": 9.327824664044418e-05,
|
| 708 |
+
"loss": 7.4957,
|
| 709 |
"step": 100
|
| 710 |
},
|
| 711 |
{
|
| 712 |
+
"epoch": 0.17603485838779956,
|
| 713 |
+
"grad_norm": 4.436333656311035,
|
| 714 |
+
"learning_rate": 9.313933643837825e-05,
|
| 715 |
+
"loss": 6.8875,
|
| 716 |
"step": 101
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 0.17777777777777778,
|
| 720 |
+
"grad_norm": 3.8972339630126953,
|
| 721 |
+
"learning_rate": 9.299911117025071e-05,
|
| 722 |
+
"loss": 6.743,
|
| 723 |
"step": 102
|
| 724 |
},
|
| 725 |
{
|
| 726 |
+
"epoch": 0.17952069716775598,
|
| 727 |
+
"grad_norm": 2.9377124309539795,
|
| 728 |
+
"learning_rate": 9.285757511070988e-05,
|
| 729 |
+
"loss": 7.4621,
|
| 730 |
"step": 103
|
| 731 |
},
|
| 732 |
{
|
| 733 |
+
"epoch": 0.1812636165577342,
|
| 734 |
+
"grad_norm": 4.165389060974121,
|
| 735 |
+
"learning_rate": 9.271473257436239e-05,
|
| 736 |
+
"loss": 7.7563,
|
| 737 |
"step": 104
|
| 738 |
},
|
| 739 |
{
|
| 740 |
+
"epoch": 0.1830065359477124,
|
| 741 |
+
"grad_norm": 2.9757657051086426,
|
| 742 |
+
"learning_rate": 9.257058791564174e-05,
|
| 743 |
+
"loss": 7.1465,
|
| 744 |
"step": 105
|
| 745 |
},
|
| 746 |
{
|
| 747 |
+
"epoch": 0.18474945533769063,
|
| 748 |
+
"grad_norm": 1.6938395500183105,
|
| 749 |
+
"learning_rate": 9.242514552867555e-05,
|
| 750 |
+
"loss": 6.9904,
|
| 751 |
"step": 106
|
| 752 |
},
|
| 753 |
{
|
| 754 |
+
"epoch": 0.18649237472766886,
|
| 755 |
+
"grad_norm": 2.4745097160339355,
|
| 756 |
+
"learning_rate": 9.227840984715154e-05,
|
| 757 |
+
"loss": 6.9228,
|
| 758 |
"step": 107
|
| 759 |
},
|
| 760 |
{
|
| 761 |
+
"epoch": 0.18823529411764706,
|
| 762 |
+
"grad_norm": 2.1547858715057373,
|
| 763 |
+
"learning_rate": 9.213038534418244e-05,
|
| 764 |
+
"loss": 7.2495,
|
| 765 |
"step": 108
|
| 766 |
},
|
| 767 |
{
|
| 768 |
+
"epoch": 0.18997821350762528,
|
| 769 |
+
"grad_norm": 1.9106367826461792,
|
| 770 |
+
"learning_rate": 9.198107653216961e-05,
|
| 771 |
+
"loss": 6.4882,
|
| 772 |
"step": 109
|
| 773 |
},
|
| 774 |
{
|
| 775 |
+
"epoch": 0.19172113289760348,
|
| 776 |
+
"grad_norm": 2.465655565261841,
|
| 777 |
+
"learning_rate": 9.183048796266547e-05,
|
| 778 |
+
"loss": 6.3857,
|
| 779 |
"step": 110
|
| 780 |
},
|
| 781 |
{
|
| 782 |
+
"epoch": 0.1934640522875817,
|
| 783 |
+
"grad_norm": 1.8584837913513184,
|
| 784 |
+
"learning_rate": 9.167862422623475e-05,
|
| 785 |
+
"loss": 6.5691,
|
| 786 |
"step": 111
|
| 787 |
},
|
| 788 |
{
|
| 789 |
+
"epoch": 0.1952069716775599,
|
| 790 |
+
"grad_norm": 2.387545108795166,
|
| 791 |
+
"learning_rate": 9.15254899523146e-05,
|
| 792 |
+
"loss": 6.67,
|
| 793 |
"step": 112
|
| 794 |
},
|
| 795 |
{
|
| 796 |
+
"epoch": 0.19694989106753813,
|
| 797 |
+
"grad_norm": 2.282320022583008,
|
| 798 |
+
"learning_rate": 9.13710898090734e-05,
|
| 799 |
+
"loss": 6.308,
|
| 800 |
"step": 113
|
| 801 |
},
|
| 802 |
{
|
| 803 |
+
"epoch": 0.19869281045751633,
|
| 804 |
+
"grad_norm": 2.2023017406463623,
|
| 805 |
+
"learning_rate": 9.121542850326849e-05,
|
| 806 |
+
"loss": 6.8043,
|
| 807 |
"step": 114
|
| 808 |
},
|
| 809 |
{
|
| 810 |
+
"epoch": 0.20043572984749455,
|
| 811 |
+
"grad_norm": 2.360322952270508,
|
| 812 |
+
"learning_rate": 9.105851078010266e-05,
|
| 813 |
+
"loss": 6.5446,
|
| 814 |
"step": 115
|
| 815 |
},
|
| 816 |
{
|
| 817 |
+
"epoch": 0.20217864923747278,
|
| 818 |
+
"grad_norm": 2.0918686389923096,
|
| 819 |
+
"learning_rate": 9.090034142307955e-05,
|
| 820 |
+
"loss": 6.3559,
|
| 821 |
"step": 116
|
| 822 |
},
|
| 823 |
{
|
| 824 |
+
"epoch": 0.20392156862745098,
|
| 825 |
+
"grad_norm": 2.4008851051330566,
|
| 826 |
+
"learning_rate": 9.074092525385777e-05,
|
| 827 |
+
"loss": 6.5768,
|
| 828 |
"step": 117
|
| 829 |
},
|
| 830 |
{
|
| 831 |
+
"epoch": 0.2056644880174292,
|
| 832 |
+
"grad_norm": 2.7958297729492188,
|
| 833 |
+
"learning_rate": 9.058026713210397e-05,
|
| 834 |
+
"loss": 6.381,
|
| 835 |
"step": 118
|
| 836 |
},
|
| 837 |
{
|
| 838 |
+
"epoch": 0.2074074074074074,
|
| 839 |
+
"grad_norm": 2.0112011432647705,
|
| 840 |
+
"learning_rate": 9.041837195534462e-05,
|
| 841 |
+
"loss": 6.3069,
|
| 842 |
"step": 119
|
| 843 |
},
|
| 844 |
{
|
| 845 |
+
"epoch": 0.20915032679738563,
|
| 846 |
+
"grad_norm": 2.51285982131958,
|
| 847 |
+
"learning_rate": 9.025524465881683e-05,
|
| 848 |
+
"loss": 6.1612,
|
| 849 |
"step": 120
|
| 850 |
},
|
| 851 |
{
|
| 852 |
+
"epoch": 0.21089324618736383,
|
| 853 |
+
"grad_norm": 2.0926644802093506,
|
| 854 |
+
"learning_rate": 9.009089021531778e-05,
|
| 855 |
+
"loss": 6.8539,
|
| 856 |
"step": 121
|
| 857 |
},
|
| 858 |
{
|
| 859 |
+
"epoch": 0.21263616557734205,
|
| 860 |
+
"grad_norm": 2.5901362895965576,
|
| 861 |
+
"learning_rate": 8.99253136350532e-05,
|
| 862 |
+
"loss": 6.6751,
|
| 863 |
"step": 122
|
| 864 |
},
|
| 865 |
{
|
| 866 |
+
"epoch": 0.21437908496732025,
|
| 867 |
+
"grad_norm": 2.451211452484131,
|
| 868 |
+
"learning_rate": 8.97585199654846e-05,
|
| 869 |
+
"loss": 6.4976,
|
| 870 |
"step": 123
|
| 871 |
},
|
| 872 |
{
|
| 873 |
+
"epoch": 0.21612200435729848,
|
| 874 |
+
"grad_norm": 2.4457273483276367,
|
| 875 |
+
"learning_rate": 8.959051429117551e-05,
|
| 876 |
+
"loss": 6.479,
|
| 877 |
"step": 124
|
| 878 |
},
|
| 879 |
{
|
| 880 |
+
"epoch": 0.2178649237472767,
|
| 881 |
+
"grad_norm": 2.1082019805908203,
|
| 882 |
+
"learning_rate": 8.942130173363627e-05,
|
| 883 |
+
"loss": 6.1699,
|
| 884 |
"step": 125
|
| 885 |
},
|
| 886 |
{
|
| 887 |
+
"epoch": 0.2196078431372549,
|
| 888 |
+
"grad_norm": 2.3551406860351562,
|
| 889 |
+
"learning_rate": 8.925088745116816e-05,
|
| 890 |
+
"loss": 6.3931,
|
| 891 |
"step": 126
|
| 892 |
},
|
| 893 |
{
|
| 894 |
+
"epoch": 0.22135076252723312,
|
| 895 |
+
"grad_norm": 2.5933122634887695,
|
| 896 |
+
"learning_rate": 8.907927663870593e-05,
|
| 897 |
+
"loss": 6.4131,
|
| 898 |
"step": 127
|
| 899 |
},
|
| 900 |
{
|
| 901 |
+
"epoch": 0.22309368191721132,
|
| 902 |
+
"grad_norm": 2.589916944503784,
|
| 903 |
+
"learning_rate": 8.890647452765954e-05,
|
| 904 |
+
"loss": 6.0784,
|
| 905 |
"step": 128
|
| 906 |
},
|
| 907 |
{
|
| 908 |
+
"epoch": 0.22483660130718955,
|
| 909 |
+
"grad_norm": 2.3914597034454346,
|
| 910 |
+
"learning_rate": 8.873248638575471e-05,
|
| 911 |
+
"loss": 5.9878,
|
| 912 |
"step": 129
|
| 913 |
},
|
| 914 |
{
|
| 915 |
+
"epoch": 0.22657952069716775,
|
| 916 |
+
"grad_norm": 2.7471046447753906,
|
| 917 |
+
"learning_rate": 8.855731751687233e-05,
|
| 918 |
+
"loss": 6.3665,
|
| 919 |
"step": 130
|
| 920 |
},
|
| 921 |
{
|
| 922 |
+
"epoch": 0.22832244008714597,
|
| 923 |
+
"grad_norm": 3.0270299911499023,
|
| 924 |
+
"learning_rate": 8.838097326088667e-05,
|
| 925 |
+
"loss": 6.5195,
|
| 926 |
"step": 131
|
| 927 |
},
|
| 928 |
{
|
| 929 |
+
"epoch": 0.23006535947712417,
|
| 930 |
+
"grad_norm": 2.6412923336029053,
|
| 931 |
+
"learning_rate": 8.820345899350275e-05,
|
| 932 |
+
"loss": 6.2827,
|
| 933 |
"step": 132
|
| 934 |
},
|
| 935 |
{
|
| 936 |
+
"epoch": 0.2318082788671024,
|
| 937 |
+
"grad_norm": 2.825366973876953,
|
| 938 |
+
"learning_rate": 8.802478012609234e-05,
|
| 939 |
+
"loss": 6.2036,
|
| 940 |
"step": 133
|
| 941 |
},
|
| 942 |
{
|
| 943 |
+
"epoch": 0.23355119825708062,
|
| 944 |
+
"grad_norm": 2.7652394771575928,
|
| 945 |
+
"learning_rate": 8.78449421055291e-05,
|
| 946 |
+
"loss": 5.8052,
|
| 947 |
"step": 134
|
| 948 |
},
|
| 949 |
{
|
| 950 |
+
"epoch": 0.23529411764705882,
|
| 951 |
+
"grad_norm": 2.718158483505249,
|
| 952 |
+
"learning_rate": 8.766395041402244e-05,
|
| 953 |
+
"loss": 6.5726,
|
| 954 |
"step": 135
|
| 955 |
},
|
| 956 |
{
|
| 957 |
+
"epoch": 0.23703703703703705,
|
| 958 |
+
"grad_norm": 2.7716219425201416,
|
| 959 |
+
"learning_rate": 8.74818105689505e-05,
|
| 960 |
+
"loss": 6.2441,
|
| 961 |
"step": 136
|
| 962 |
},
|
| 963 |
{
|
| 964 |
+
"epoch": 0.23877995642701524,
|
| 965 |
+
"grad_norm": 2.9765782356262207,
|
| 966 |
+
"learning_rate": 8.729852812269192e-05,
|
| 967 |
+
"loss": 6.2913,
|
| 968 |
"step": 137
|
| 969 |
},
|
| 970 |
{
|
| 971 |
+
"epoch": 0.24052287581699347,
|
| 972 |
+
"grad_norm": 3.5550148487091064,
|
| 973 |
+
"learning_rate": 8.711410866245647e-05,
|
| 974 |
+
"loss": 6.0119,
|
| 975 |
"step": 138
|
| 976 |
},
|
| 977 |
{
|
| 978 |
+
"epoch": 0.24226579520697167,
|
| 979 |
+
"grad_norm": 3.569737434387207,
|
| 980 |
+
"learning_rate": 8.692855781011493e-05,
|
| 981 |
+
"loss": 6.2118,
|
| 982 |
"step": 139
|
| 983 |
},
|
| 984 |
{
|
| 985 |
+
"epoch": 0.2440087145969499,
|
| 986 |
+
"grad_norm": 3.7226216793060303,
|
| 987 |
+
"learning_rate": 8.674188122202756e-05,
|
| 988 |
+
"loss": 7.3339,
|
| 989 |
"step": 140
|
| 990 |
},
|
| 991 |
{
|
| 992 |
+
"epoch": 0.2457516339869281,
|
| 993 |
+
"grad_norm": 3.0813937187194824,
|
| 994 |
+
"learning_rate": 8.655408458887171e-05,
|
| 995 |
+
"loss": 6.3654,
|
| 996 |
"step": 141
|
| 997 |
},
|
| 998 |
{
|
| 999 |
+
"epoch": 0.24749455337690632,
|
| 1000 |
+
"grad_norm": 3.1687822341918945,
|
| 1001 |
+
"learning_rate": 8.63651736354684e-05,
|
| 1002 |
+
"loss": 6.189,
|
| 1003 |
"step": 142
|
| 1004 |
},
|
| 1005 |
{
|
| 1006 |
+
"epoch": 0.24923747276688454,
|
| 1007 |
+
"grad_norm": 3.1596567630767822,
|
| 1008 |
+
"learning_rate": 8.617515412060771e-05,
|
| 1009 |
+
"loss": 5.9898,
|
| 1010 |
"step": 143
|
| 1011 |
},
|
| 1012 |
{
|
| 1013 |
+
"epoch": 0.25098039215686274,
|
| 1014 |
+
"grad_norm": 3.297518730163574,
|
| 1015 |
+
"learning_rate": 8.598403183687328e-05,
|
| 1016 |
+
"loss": 5.6006,
|
| 1017 |
"step": 144
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1018 |
}
|
| 1019 |
],
|
| 1020 |
"logging_steps": 1,
|
| 1021 |
+
"max_steps": 574,
|
| 1022 |
"num_input_tokens_seen": 0,
|
| 1023 |
"num_train_epochs": 2,
|
| 1024 |
+
"save_steps": 144,
|
| 1025 |
"stateful_callbacks": {
|
| 1026 |
"TrainerControl": {
|
| 1027 |
"args": {
|
|
|
|
| 1034 |
"attributes": {}
|
| 1035 |
}
|
| 1036 |
},
|
| 1037 |
+
"total_flos": 2.1051451631468544e+17,
|
| 1038 |
+
"train_batch_size": 4,
|
| 1039 |
"trial_name": null,
|
| 1040 |
"trial_params": null
|
| 1041 |
}
|
last-checkpoint/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 6776
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1953c2fb4a5133466f4bc3f2225fe0f18ba77aa13ef3c64bbad22efa5fd6c36
|
| 3 |
size 6776
|